Documents
Resources
Learning Center
Upload
Plans & pricing Sign in
Sign Out

radar_ew_2004_25_to_29_Oct_2004_490pages_

VIEWS: 83 PAGES: 490

									       CEP COURSE
           ON
 RADAR ELECTRONIC WARFARE


             25 October – 29 October 2004




  Defence Research & Development Organisation

DEFENCE ELECTRONICS RESEARCH LABORATORY

             HYDERABAD – 500 005
ORGANISING COMMITTEE




Chairman &
Course Director :
                        Shri PRATULANANDA PAL




Members :
                        Shri P. RAGHAVENDRA RAO
                        Shri M. PRAHLADA RAO
                        Smt. P. PAPAJI


Member Secretary &
Course Co-Ordinator :
                        Shri J. SHANKAR RAO
                           FACULTY

                    DLRL, HYDERABAD


Shri G Kumaraswamy Rao                Dr. NN Sastry

Dr. RC Agarwal                        Shri BR Gandhi

Smt Renuka Chitrakar                  Shri KR Sundaram

Shri Pratulananda Pal                 Shri R Rama Rao

Shri RBR Prasad                       Shri J Shanker Rao

Shri K SivaKrishna Reddy              Shri Siva Raju Dadi

Shri P Raghavendra Rao                Shri S Vijaya Kumar

Shri HR Krishna Kumar                 Shri MK Das

Shri N Srinivas Rao                   Shri DD Sarma

Shri V Sobha Shankar                  Shri Y Gopala Krishna

Shri Niranjan Prasad                  Shri R Anand

Smt Y Hemalatha                       Smt UVV Krishnaveni

Smt SV UmaMaheswari                   Shri KSVM Shyam Kumar

Shri AK Singh                         Shri MP Narender

Smt M Santha

                        Other Labs & BEL
Shri K VishnuVardhana Rao                   Sc-G (Retd)

Shri G Kuloor                               LRDE, Bangalore

Shri Bhima Shankar                    RCMA, Hyderabad

Cdr Prasad (Retd)                     BEL, Hyderabad
                             CONTENTS
       Chapter No   Topic                         Page No
I                   An Overview of Radar EW       I – 1 to I - 34
II                  Electronic Support Systems    II – 1 to II - 34
III                 ELINT Systems                 III – 1 to III - 28
IV                  DIFM & Digital Receiver       IV – 1 to IV - 23
V(1)                DF for EW Systems             V(1) – 1 to V (1) - 11
V(2)                Analysis of Butler Matrix V(2) – 1 to V(2) - 7
                    Based DF Systems
V(3)                Interferometer DF Systems V(3) – 1 to V(3) - 13
                    for Radar EW
VI                  HW & SW Design of EW VI – 1 to VI - 20
                    Display
VII                 ESM Processors                VII – 1 to VII - 21
VIII                Electronic Attack Systems     VIII – 1 to VIII - 21
IX                  DSP Based Servo System        IX – 1 to IX - 21
X                   Techniques Generator          X – 1 to X - 19
XI                  High Power MW Transmitter     XI – 1 to XI - 21
XII                 MM Wave Radar System –               XII – 1 to XII - 9
                    EW Aspects
XIII                ECM Effectiveness             XIII – 1 to XIII - 24
XIV                 Test & Evaluation of EW XIV – 1 to XIV - 26
                    Systems
XV                  Frequency Memory Loop         XV – 1 to XV - 20
XVI                 RFP Technique for unique XVI – 1 to XVI - 26
                    Identification of emitter
                    system
XVII                TWT – Fundamentals            XVII – 1 to XVII - 14
XVIII               Decoy System                  XVIII – 1 to XVIII - 22
XIX                 Multisensor Interface for IEW XIX – 1 to XIX - 13
                    system
         CHAPTER – I




 AN OVER VIEW OF RADAR EW


Shri. G. KUMARASWAMY RAO, Sc-‘H’
            DIRECTOR, DLRL




      CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                   I
                          CHAPTER – I




            AN OVER VIEW OF RADAR EW




                           CONTENTS




1.   HISTORICAL BACKGROUND

2.   INTRODUCTION

3.   ELECTRONIC SUPPORT ( ES )

4.   ELECTRONIC ATTACK ( EA )

5.   ELECTRONIC PROTECTION ( EP)

6.   CONCLUSION




       CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                  II - 1
                  AN OVER VIEW OF RADAR EW
                   Shri G Kumaraswamy Rao, Sc’H’
                    DIRECTOR, DLRL, Hyderabad



1.     Historical Background :
       It was about 5.30 p.m. on a Saturday afternoon in October 1967 the
Eliat destroyer of Israel which was on a routine patrol Mission, was hit by
two Styx radar-guided missiles and sank. Forty Seven crew members of
the Eliat died.    The enemy was underestimated and the COMINT system
failed. The tunable microwave APR-9 receiver of Korean war vintage on
the ship Eliat could hardly match the capability of Styx which was a fire
and forget surface-to-surface missile. It has on board a targeting radar
which switches on only in the mid course to its target. That gave the Eliat
crew less than a minute and half to react. What went wrong? Would the
tragedy have been avoided? That was the wake-up-call for the EW
designers and it triggered maximum awareness of importance of EW.


       On May 4, 1982 an AM-39 Excocet missile fired from an Argentine
Super Etendard ripped into the British HMS Sheffield, during the Falkland
war.    24 British sailors were killed and 24 wounded. The destroyer was
detected by an SP-2H Neptune maritime patrol aircraft earlier that
morning. The two SUs which launched the Exocet missiles traveled at
low altitude ie 30 m to avoid detection by the Sheffield's radar and
maintained strict radio silence.        They popped up to 160 m twice to
determine the co-ordinates of Sheffield. The Scheffield radar failed to
forewarn the ships crew about the SUs. Then the EW Radar Warner also
did not do the job.      According to one report, the ships radar warning
systems were programmed to identity the Exocet missiles as friendly.
This incident baffled the EW military planners. The destruction of the
Sheffield taught not only the British Navy, but navies all over the world, a
valuable lesson.    More and sophisticated EW systems are necessary to
forewarn the impending dangers. Rapidly deployable counter measures
are critical to the life of military vessels - and to the lives of those abode.
           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                      I-1
2.     Introduction :
       The extensive application of EM spectrum to Communication,
Radar and Navigation provided the armed forces with powerful tools.
Radio Communication provided coordination between forces, radio
navigation gives accurate location of the deployed forces and radar
performed surveillance of the battle space to verify force deployments and
detect hostile forces. The EM spectrum is so extensively exploited, today
without its use, the survivability of armed forces is jeopardized.


       EW is the science of manipulation and control of EM environment
for its own survivability but denies or limits it to the adversary.           EW
technology extracts essential information from the EM environment. This
information is then exploited to influence adversary's capability to
coordinate its activities, to restrict its communication media, to deny the
use of radar for weapon launching or guiding. EW enhances the
survivability of own forces by denying the use of EM spectrum by the
enemy.


       EW      is   broadly     classified    based    on (i) frequency spectrum;
(ii)functionality; (iii) intended role.


2.1    Frequency Spectrum Classification:
       EW is divided into three groups based on frequency                      (i)
Communication EW; (ii) Radar EW; (iii) EO/IR EW; (iv) Hybrid EO/IR-
RFEW


       (i)      Communication EW:             Communication links are required to
disseminate, voice, digital data, FAX etc. between land forces, aircrafts and
ships. They may use HF (3-30 MHz) VHF (30-300 MHz) and UHF (300 MHz
- 3 GH).        Communication link data rates depend on link bandwidth,
modulation technique and signal to noise ratio.                  Advances in computer
technology have enabled increased communication link capacity for handling
and processing data. The high data rates permit transmission from satellites

             CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                        I-2
and between precision weapons and launch platforms. Communication EW
involves interception, direction finding and analysis of hostile emissions,
whether by voice or data link.                           Analysis of intercepted signal provides
valuable information for command and control purposes. This real time data
is necessary to counter the enemy's communication system by jamming.


       (ii)        Radar EW: Armed forces use radar for both defensive and
offensive weapon systems. Reflected R.F. echoes of the target, are used
to measure target range, bearing and elevation and determine target
location. Radar uses RF transmission ranging from high frequency (HF)
to millimeter waves (30 MHz - 95 GHz). The frequencies are designated
by various alphabetical letters. Both old and new designations are given
at Table 1. RF can be pulsed or continuous wave (CW). Radar function
includes      target       detection,          identification,        acquisition,   tracking   and
navigation. Radar extracts range, bearing and speed of a target. Radar
information is used for launch and control of a weapon like missile, air
defence gun etc. Modern advancements include Phase Array Antennas,
Complex modulation on the radar pulse, Low probability of Intercept
Radars, improved signal processing to extract data from highly corrupted
echo signal etc.


       Radar EW involves extraction of detailed information of Radar
Signals emitted, use of this information either to formulate Electronic
order of Battle (EoB), or provide the information to a jammer to operate
in an efficient way.


               L               S               C                 X         Ku        K          Ka
               D           E       F       G         H       I             J                    K
(Ghz) 1                2       3       4         6       8       10   12        18   20              26




                Table No 1 : Radar Frequency Designations



              CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         I-3
      (iii)     EO/IR (Electro optic/Infrared) EW:                 There has been a
steady growth in EO/IR guided weapons. EM spectrum extends beyond
Millimetric Waves. The IR encompasses wave length from 1000 (300
GHz) to 1 micron. In general hot targets like jet engines emit IR energy
in the 0.75 u to 3 u range.          EW systems and threats receivers use IR
energy to detect, identify, locate and guide missiles to radiating objects.
IR guided missile detects the IR signature of the aircraft and home on to
the emitter of the IR energy. The plume, the tail pipe, heated leading
edges of wings, or the IR image of the aircraft itself can be source of IR
for locking on of the missile using an IR seeker.                    EW involves the
detection and location of an incoming missile by a missile warning
receiver and deflection of the same by launching chaff etc.


      (iv)      Hybrid IR/EO-RF: Present trend is to develop a hybrid
receiver which fuzes the data obtained from Radar, IR/EO sensors, ESM
sensor etc. to obtain more accurate identification and location of a missile
emitter. These hybrids provide a high resolution three-dimensional target
information that greatly complicate EW response.


2.2   Functionality Classification:
      EW is classified based on functionality into three groups
(i)Electronic Support (ES); (ii) Electronic Attack (EA); (iii) Electronic
Self Protection (EP).


      (i)       Electronic Support (ES): ES           also    is    known   by   ESM
(Electronic Support Measure)             ES involves search, intercept, locate,
record and analyze radiated EM energy for the purpose of exploiting 'the
radiation information either for formulating EOB (Electronic Order of
Battle) or to provide the real time information to EA system. ES provides
surveillance and warning information derived from intercepted EM
environment emissions.


      (ii)      Electronic Attack (EA): EA also is known name ECM
(Electronic Counter Measure).           EA involves action taken to prevent or

             CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                        I-4
reduce enemy's effective use of EM spectrum.                   It can be active like
Jammers, or it can be passive like chaff.


       (iii)     Electronic Self Protection (EP): EP also is known by old
name ECCM (Electronic Counter Counter Measures).                             EP involves
actions taken to ensure friendly use of EM spectrum despite the use of
ECM. EP protects own platform against EA by the adversary.


2.3    ROLE CLASSIFICATION
       EW can be classified based on the role it has been assigned to carry
out (i) Tactical (ii) Strategic.


      (i) Tactical Role : Tactical Role for EW encompasses the use of
information obtained through ES in real time for immediate use like in
case of electronic jamming etc. As such the ES equipment should be wide
open in frequency and should cover wide angle 0 to 360° in space.


       (ii)      Strategic Role :        Closely related to tactical role is the use
of EW for signal intelligence (sigint) gathering, for strategic role.
SIGINT is more strategically oriented although its information is often of
tactical importance. Sigint provide important information for design of
EA and EP.          Sigint can be COMINT (Communication Intelligence) or
ELINT (Electronic Intelligence).              ELINT involves the collection of
technical information on the radar emitters of the adversaries whenever
they are switched on, recording them, and analyzing them. The receivers
used will be more sensitive than the ES Rxs, and need not be wide open as
in the tactical case. Elint have the objective of securing the maximum
possible data on EM environment.                The information can be used to
determine the enemy's EoB, it is also used to form a threat library for use
in ES and EA equipment.




              CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         I-5
3. Electronic Support (ES) :
      ES also known by ESM, involve actions taken to search, intercept,
locate, record and analyze radiated EM energy.                  The information is
employed for threat recognition and to use it in tactical employment in EA
equipment. The ES function is for real time use, whereas Elint Rxs are
used for intelligence collection, which can be subsequently used. Elint
Rxs perform fine grain analysis of emitters of interest. If the data cannot
be analysed, it can be stored and analyzed at a later time.


EW Rxs generally measures quantitatively the following parameters


      a)       Frequency
      b)       Angle of Arrival (AOA) or D.F. (Direction Finding)
      c)       Pulsewidth (PA)
      d)       Pulse Repition Frequency (PRF)
      e)       Time of Arrival (TOA)


The EW Rxs can be classified by their (a) Application and (b) Structure.


3.1   EW Rx by Application: (a) RWR (Radar Warning Rxs); (b) ESM
Rxs; (c) ECM Rxs; d) Elint Rxs.


      (a)      RWR :         A RWR detects the weapon radar and provides
warning to the pilot.      As soon as the radar locks on to the target, the
receiver with a moderate sensitivity detects the main beam. It provides
the frequency, direction, characteristics of emitter             which is used for
sorting and identification. The RWR Rx will have a wide frequency and
spatial coverage.      RWRs are generally the simplest form of ESM Rx
consisting of a low sensitivity equipment (generally -40 dBm sensitivity).
It is preset to cover the expected characteristics of threats. It uses the
range advantage to indicate a threat.


      (b)      ESM Rxs :       An ESM Rx is used to obtain all the
information about the emitters to determine the EOB (Electronic Order of

            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                       I-6
Battle). The Rx needs to collect all the information of all the radars in the
environment. So it is wide open in frequency and have wide spatial angle.
They are more complex than RWRs, have more sensitivities to intercept
radars through side lobes. They have higher Direction Finding Accuracy.
The information of ESM can be provided to EA for its jamming.


       (c)      ECM Rxs : An ECM Rx works in conjunction with EA
system. To concentrate the energy of a jammer towards the victim radar,
frequency as well as AoA information of the radar is required.                  ECM Rx
provide these. The Rx should interface with EA and quickly set the EA
system to operate in its high efficiency mode.


       (d)      ELINT Rxs : An Elint Rx works for collection of data and
information useful for strategic planning. It does fine-grain analysis and
therefore has a very large sensitivity compared to ESM Rx.                            Its
instantaneous bandwidth is far less than the ESM Rx since it covers one or
a few signals of interest. If the data cannot be analyzed at the collection
station, it can be stored and analyzed at a later time. It normally works in
peace time to collect as much data as possible about the adversary


3.2    EW Rx by Structure :
         a) Crystal Video;          b)    Superheterodyne;         c)       Instantaneous
Frequency Measurement (IFM); d)                  Channelised; e)            Compressive
(microscan) f) Bragg Cell and g) Digital Rx.


       a)       Crystal Video Rx (CVR) : Crystal Video Rx is the simplest of all
types. It is a diode (crystal) detector whose output is amplified to an adequate
level by a Video Amplifier. To obtain wide dynamic range a log-video amplifier
is used. The detector operates in 'Square Low Region', so that output voltage is a
function of the input power. The output of the crystal video receiver is a series of
pulses with amplitude proportional to R.F. signal power with the same start and
stop times. The sensitivity of a CVR is usually in the range of -35 to -50 dBm.



             CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                        I-7
      The block diagram of Crystal Video Rx is shown at Fig 1.




                                                                      Log
              Band               Pre                                 Video         ESM
            Pass Filter        Amplifier          Detector          Amplifier    Processo



                          Fig 1 : CRYSTAL VIDEO Rx


      b)      Super Heterodyne Rx:            The      Super      heterodyne    Rx
heterodynes with RF frequency to convert the input to an IF band.               A
tuned (Local Oscillator) LO is used to shift the input R.F. A fixed IF is
more desirable where in the necessary gain and filter selectivity can be
easily provided. Fig. 2 depicts the block diagram of a Superhet Rx. A
Tunable Bandpass filter is used after the antenna. This is controlled along
with the L.O. to select only the portion of the input spectrum that is
converted to the IF bandwidth. This gives isolation from other signals to
the Rx. Typically the sensitivity of a Superhet Rx is from -70 to -90 dbm
depending on the IF bandwidth.             Frequency accuracy of the Rx is
dependent on oscillator stability and resolution of IF bandwidth. Because
of narrow bandwidth, this type of receiver is used in conjunction with
other types of ESM Rxs.




                                Fig 2 : SUPERHET Rx


      c)      IFM Receiver: An IFM Rx uses delay lines and measures
the phase difference to measure its frequency. R.F. is split into a delayed
path with delay T and an undelayed path. It θ is the phase angle between
delayed and undelayed waves, θ = ωT. If phase angle is measured and

           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                      I-8
the delay time T is known, frequency can be computed.                     In an IFM
receiver the phase angle θ is measured using the relation below:


                      A       =       X Sin θ
                      B       =       X Cos θ
                                            –1
                      θ       =       tan        A/B    =ω T
       Where x is the amplitude information. The signal is passed through
a limiting Amplifier before applied to the R.F. power divider. One of the
outputs of power divider is given to the phase correlator. Other input to
the correlator is the delayed signal by a specific time T. The correlator
multiplies the undelayed signal and the delayed signal and produces Sine
and cosine Video outputs.         They are digitized in a 8 bit quantizer and
phase ‘θ’ is computed. No. of delay lines and associated correlaters are
used to obtain good accuracy and resolution.                 The longest delay line
determines the frequency accuracy and resolution, whereas the shortest




                                                       π/2
Cost ωt

                                                        Cos θ
                                                                            Sin θ

                          θ = ωT

                                     Fig 3 : I F M


delay line determines the unambiguous bandwidth. Frequency measurement can
be done by conventional analog ways by comparing the amplitude in analog
comparators. A Gray Code is obtained which indicates the frequency of the
incoming signal. Another approach is to digitize the correlator output by an A/D
converter. The digitized data is fed to a ROM which perform ω =1/T tan-1 (Sin
ωT/Cos ωT). The frequency is thus directly computed. The IFM will encounter
many problems when there are no. of emitters in space. The problems are (i)
Simultaneous signals ie pulse on pulse; (ii) Overlapping Signals; (iii) Pulse on

           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                      I-9
CW condition etc. These problems have been solved to certain extent and lot is
needed to be done with the present high density emitter environment.


         (iv)      Channelised Rx : The ideal Rx for EW is channelised Rx. It
consists of a set of fixed frequency Rx with their passbands set such that the upper
edge of the 3dB bandwidth of one Rx is same as the lower edge of the 3 dB
bandwidth of the next. It provides a demodulated output for signals in each
channel.       It can have narrow bandwidth to provide excellent sensitivity.      It
provides 100% probability of intercept for signals within its frequency range. It
also provides full feature reception for multiple simultaneous signals, as long as
they are in different frequency channels. The problem of course is in complexity
of implementation and huge no. of fixed Tuned Rxs required making it very bulky
and unwieldy. For 1 MHz of isolation across frequency range of 2-4 GHz,


                                          Fixed Tuned Rx
                 Multiplexer
                                          Fixed Tuned Rx

                                          Fixed Tuned Rx

                         Fig 4 : CHANNELISED Rx


2000 channels are required.             2000 separate Rxs are required making it
bulky.     However, with the miniaturization in packaging technology, in
future the size and weight can be brought down appreciably.


         (v)       Compressive Rx: This Rx is also called microscan Rx. It is
basically a superhet Rx that is rapidly tuned.                  The rate of scan in a
compressive Rx is much faster than superheterodynde Rx. The output is
passed through a compressive filter that has a delay proportional to
frequency. The delay versus frequency slope exactly compensates for the
receivers sweep rate.            Thus the output of the Rx is coherently time
compressed to make a strong signal spike.




                CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                          I - 10
                                                                                 Compressive
                  Filter B.P.                           IF Filter                   Filter

                                                                Sweep                  Detector
                                        L.O                    Generator

                                     Fig 5 : COMPRESSIVE Rx


          (vi)      Bragg Cell Rx : Bragg Cell Rx is a spectrum analyzer capable of
handling simultaneous signals. The RF signal is applied to a 'Bragg Cell'. The
cell deflects the light in proportion to the R.F. frequency of the signal. The light
is focused on the photodiode detector array. The spatial distribution across the
photo diode array represents the instantaneous.            Fourier transform of the input
signal.     The array detects the deflection angles of the diffracted beam and
produces output signals from which a digital read out of all signal frequencies
present are obtained. The Bragg Cell has limited dynamic range of 25 dB.




                                                                          BRAGG
                                                                                Cell              L
                                              Laser                                               I
                                                                                                  G
                                                                                                  H
                                                                                                  T
                  Filter B.P.                                                                     D
                                            RF                                                    E
                                                                                                  T
                                                                                                  E
                                                                                                  C
                                                                                                  T
                                        Fig 6 : BRAGG CELL                                        O
                                                                                                  R


          (vii)     Digital Rxs : This is the Rx for future. It digitizes the RF signal
to be processed in a computer. Software can be written to simulate any type of
filter or demodulator giving lot of flexibility and optimization. The problem of
Digital Rx is in the availability of fast (A/D) Analog to Digital Converters. For
capturing the frequency we need a minimum of two samples. Present day digital
Rx is working with 1 GHz bandwidth with 8 bit (48 dB) dynamic accuracy.



                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                           I - 11
                       Rx
                       Front End                     A/D                     Computer


                                       Fig 7 : DIGITAL Rx




3.3    Paraments Measured by ES Rxs :
       ES Receiver measures the following quantitatively.
       (i)       Frequency
       (ii)      Angle of Arrival (AOA)
       (iii)     Pulse width (PA)
       (iv)      Time of Arrival


       (i)       Frequency Measurement : The frequency information is used for
both sorting and jamming. By frequency comparison of pulses received, pulse trains
of various radars can be sorted out. Knowing the frequency, jammer can be tuned to
transmit in that frequency range.


       (ii)      Angle of Arrival (AOA) / Direction Finding :                 It is often
important to determine the location of the emitter.            A direction finding (DF)
system is useful in locating the signal source. A DF system gives the direction of
emitter. Two or more D.F. systems are necessary to obtain the location of the
emitter by triangulation. Alternatively the D.F. platform can move in space, and
taking D.F. measurement at different times, it is possible to locate the emitter.
There are no. of ways of determing the AOA. They are broadly divided into a)
Rotary D.F; b) Amplitude Comparison; c) Phase Comparison/Interferometer; d)
Time Difference of Arrival T DOA.


       a)        Rotary D.F (RDF) : The most simple method for measuring AoA
       is to rotate a narrow beam antenna at a fast speed. As the antenna beam
       sweep over the emitter, it traces out the antenna pattern on an
       oscilloscopic display. By selecting the peak of the pattern, the AOA of the

              CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                        I - 12
emitter can be determined.        An omni-directional antenna is generally
required to avoid the peak response of a sidelobe being interpreted as the
direction of emitter.     Obviously the system is slow as it employs a
mechanical rotating platform. Secondly the system is not suitable in a
multi emitter environment. Normally RDF is used in conjunction with
superhet Rx its sensitivity is very high.


b)      Amplitude Comparison : Amplitude comparison technique
is extensively used due to its lower complexity and cost.               The
principle is to derive a ratio from a pair of independent receiving
antenna channels.        Typically four or six antenna elements and
receiver channels are used to obtain 360 degree coverage.
Wideband logarithmic video detectors provide the signals for
comparison and angle determination.                The monopulse ratio is
obtained by subtracting the detected logarithmic signals and the




                                                             Antenna
                                                             Pattern




                                                             Receiver




                    Fig 8 : A Four Quadrant Amplitude D.F.

     CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                               I - 13
bearing is completed from the value of the ratio. Four quadrant
amplitude comparison systems are simple, low cost and cover from
                                                                             o
0.5 to 18 GHz.         However, the accuracies obtained are poor (5
RMS) and have very low sensitivity (-50 dBm).                  No. of antennas
can be increased to obtain higher accuracy at the cost of simplicity
and reliability. The system accuracy depends upon the error shape
and the amplitude imbalance between antennas. The error slope in
turn is a function of antenna beamwidth and squint angle between
two antennas.        The fig shows a four quadrant amplitude AOA
system.


(c)      Phase Comparison (Interferometry)
         A two element phase comparison is shown in fig.9




                              θ




                                           θ
                              d




                                     Φ


       Fig 9 : Two Element Phase Comparison D.F.


          The two antennas are separated by a distance ' d'. A E M wave
front coming at an angle O, will produce a phase O at the output of
phase correlator.
θ= 2 π d Sineθ/ λ Where λ is the wavelength of the wave.

      CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                I - 14
                  It is seen if 'd' is larger the accuracy is better.              The
       maximum           distance    between      two    antennas     without   causing
       ambiguity is half the signal wavelength. To obtain more accuracy
       without ambiguity, a multiple number of antennas with non uniform
       spacing is used. The long distances provide the accuracy and the
       short distances resolve the ambiguity. The D.F. error depends upon
       the frequency, phase measurement error and noise.                         Phase
       comparison D.F. systems comparatively give better accuracies than
       Amplitude comparison systems. The RMS accuracy could be
       around 2 o .


       (d)        Phase Comparison (Digital Bearing Discriminator) : This
       uses 16 antennas in a circular array. Each antenna feeds one input
       of a Bulter Matrix. Phase Angles O and 4 O are obtained as output
       of Butler Matrix.          They are digitized and ambiguity resolved to
       obtain a very high RMS bearing accuracies of 2 o . Sensitivity and
       dynamic range of this type of DBDs are far better than Amplitude
       Comparison D.F.


       (iii)      Time Difference of Arrival : D.F. also can be measured by
computing the time of arrival of a signal at different locations. These type of
systems do not depend upon frequency. The D.F. accuracy depends upon the
accuracy with which time difference can be measured. The D.F. accuracy is
improved by using large baselines. As such these systems are used in ships, are
on aircraft with large wing span. Accuracies can be very high of the order of 1
degree RMS.


3.4    Future Trend of ES Rxs
       i)         LPI Radar Detection : The performance of Rxs require to
be enhanced to cater to the present day modern radar especially LPI (Low
Probability of Intercept) radars. LPI Radars (a) use low peak power and
large time width and large frequency bandwidth; (b) use waveforms which
is difficult to identify such as pseudonoise.              In this type of radars the
               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         I - 15
transmitted pulse is as much as 20 dB smaller in peak power but as much
greater in average power.


The power spectrum of the stretched pulse is spread over a wider
frequency range. This spread spectrum will be below the sensitivity of the
ES Rx. So the future Rx should have low sensitivities between –85 to –
90 dbm to detect the LPI Radars. The research and development of EW
Rxs should concentrate on channelizers, compressive Rxs etc. These have
the potential for high sensitivities and good performance.


       (ii)      Hybrid Rxs : The EM environment has become so complex,
a single ES Rx is not going to give the required performance. Hybrid Rx
consisting of several Rxs perform better.                These Rxs will functions
together simultaneously and they are integrated by a common processor.
RWR, MAWR (Missile Approach Warning Rx) and a EO/IRRx in a single
package is not uncommon these days.


       (iii)     Miniaturization : The technological advancement in MIC
(Microwave Integrated Circuits) and the concept of super components will
make the size of the EW Rx very small, at the same time improving the
reliability, weight and size. Channelised Rxs may become less bulky.




       (iv)      High Speed ICs :          The availability of high speed A/Ds,
D/As etc. will revolutionize the concepts and provide the possibility of
digitizing the input RF signals directly. Thus the realization of Digital
Rx, with the associated flexible software etc. will solve many of the
limitations of the present day Rxs.


       (v)       Simulation : Developing an EW Rx is a very expensive and
time consuming job.          Some of the mistakes in the hardware could be
avoided if a model of a Rx is conceived and it is subjected to all simulated
inputs. Based on the outcome actual hardware will be built. This will cut
short effort and time.

              CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                        I - 16
4.    Electronic Attack (EA)


      EA is also known by name ECM (Electronic Counter Measure).
The aim of EA is to interfere with the enemy’s effective use of EM
spectrum. Sometimes EA also includes the use of high levels of radiated
power or directed energy to physically damage enemy assets. Jamming is
called ‘Soft Kill’, since it temporarily makes enemy assets ineffective.


      Jamming is to place an interfering signal into an enemy’s receiver
alongwith the desired signal. Jamming is effective only when the enemy’s
Rx is unable to recover the information from the desired signal. Radar
which accomplishes the task of early warning, acquisition, tracking and
guiding a weapon is a sophisticated system normally possesses the anti
jamming features. EA should accomplish its job by overcoming the EP
(Electronic Protection) features of the radar. It is a continuing game of
one-man up show between EA and EP. The Electronic device to achieve
this purpose is a RF transmitter called ‘Jammer’.


Some of the features which makes the jammer effective are as follows :


i)    RF frequency of jammer should match with the radar frequency.
ii)   Interference should be continuous, if necessary broad band of
      frequencies, or no.of jammers are used to counter search radars,
      Air     defence    radars,    track    radars,    weapon     guidance   radars
      simultaneously.


4.1   Burn Through Range
      Since jamming is one way transmission, it has distinct power
advantage over radar reflected power.              However, when the distance
between the radar and the jammer keeps on decreasing, the jammer power
received at radar increases by the square of the distance, whereas the echo
power received at the radar increases by fourth power. Therefore at some
distance between Radar and Jammer, the echo power is just greater than
            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                      I - 17
the jamming power and the radar starts to see through the jamming. This
distance is called the ‘Burn Through Range’. The Radar peak power and
the power of Jammer play an important role in determining this range.


4.2       Jammer Characteristics
          Different types of jammers are employed for different types of
radars.         Basically jamming is divided into three categories (i) noise
jamming           (ii) deception / confusion jamming (iii) modulated jamming.
The power of Jammer alone does not determine the effectiveness of that
Jammer. The type of Jamming used is extremely significant.


4.3       Noise Jamming
          Noise Jamming involves modulating an RF carrier wave with noise
and transmitting the carrier wave towards the radar. The jamming signal
has far greater power than the power of echo signal.                     Since the radar
receiver is extremely sensitive, this saturates the Rx. If the S/N is less
than 1, the target return is lost. The jamming becomes effective. There
are six types of Noise Jammers. They are (i) barrage (ii) spot (iii) sweep
(iv) sweep lock-on (v) cover pulse (vi) side lobe.


          i)       Barrage Jammers :           Barrage jammers are Wideband noise
transmitters. They deny to the adversary use of wide portion of EM specturm of
frequencies. This type of jammers have two advantages (a) No. of enemy radars
can be jammed and (b) frequency agile radars can be jammed. The disadvantage
is that its power density goes down because of large bandwidth. At the radar
frequency of interest the power of the jammer may not be effective for jamming.


          ii)      Spot Jammer : Spot Jammer have very narrow bandwidth.
The chief advantage of spot jammer is that their output is concentrated in
a narrow band and as such the jamming will be effective.                            The
disadvantage is that it can jam only one emitter and if more no. of radar
are their in space that many no. of spot jammer are to be carried. Also it
is not effective against Frequency Agile radar.


                CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                          I - 18
      iii)      Sweep Jammers : Sweep Jammers transmit a narrow band
spot jamming signal but the spot frequency is swept back and forth over
the desired band. The advantage is that all radars get covered by the high
power density. But the disadvantage is that it is not continuous. However
fast sweep jammers can overcome this to a certain extent. Fast sweep
jamming approximates a burst of energy and causes oscillation in the
receiver amplifiers which lasts until the sweep jammer once again passes
through the radar frequency and sustains the oscillation.


      iv)       Sweep Lock-on Jammers :                  This type of jammer is
essentially a swept jammer with additional feature of lock-on capability.
It consists of a jammer and a receiver. The jammer and receiver are swept
over the same frequency band.            When the Rx encounters a signal, the
frequency sweep is halted and the jammer becomes a spot jammer at this
frequency. A look through facility is provided, wherein the Rx can be
made to start sweeping again when the original signal being jammed
disappears.


      v)        Cover Pulse Jammers : This is also called ‘smart noise’
jammer. A repeater type of jammer is used in the transponder mode. The
received signal is used to set on the frequency of a low power oscillator.
PRI of radar is established. The time of arrival of the subsequent radar
pulse is predicted and the oscillator is gated ahead by a few microseconds
to a few microseconds later.          The oscillator is noise modulated, signal
amplified and transmitted. This type of jammers is effective for a steady
PRF. By time-sharing the system, no. of radars can be jammed.


      vi)       Sidelobe Jamming :           The objective in jamming a search
radar through the sidelobe is to make a large sector of the radar display
unusable.




             CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                       I - 19
4.4    Deception Jamming
       The noise jammers discussed so far, create jamming strobes on the
radar scope and thereby hide the target echo and deny the range
information. However, the centre of the jamming strobes gives out the
angular information which is what is required for a weapon guidance. So
to corrupt the angle tracking of the radar, deceptive type of jamming is
used. This will inject angle / range errors into the radar.


       i)        Range Gate Deception : The Jammer Receiver receives the
radar pulse, amplifies and retransmits. The jammer signal appears on top
of the echo received by the radar. The AGC of the radar lowers the gain
of its amplifier since the jammer signal is stronger. The range circuit with
the lowered gain, will then only see the jammer signals and not the echo.
In effect the range gate is captured. Now the jammer retransmits the radar
signal received, with an increasing time delay. The radar follows the false
signal and thus the true range of the target is denied. The jammer range
gate pull off rate should not exceed the radars tracking rate, at the same
time it should be fast enough for better protection. Modern radars use
leading edge tracking which defeats the range gate pull off technique. To
overcome this problem ‘Rage gate pull in’ technique is used. For this, it
is necessary to know the PRI of the radar pulse to predict the arrival time
of the next pulse. However, this technique requires great sophistication
for employment against staggered pulse train and cannot work at all
against randomly timed pulses.


       ii)       Angle Deception :          Normally tracking radars use either
conical      scan   or   monopulse      tracking     receivers    to   obtain   angular
information.


       a)        Conical Scan Radar : Angle Deception : A conical scan
       radar scans its antenna beam in a circular motion.                    A cone is
       formed in space. Information from the scan is used to move the
       antenna in a servo loop so that the target is in the centre. If the
       target is slightly non-centered, a sinusoidal output is observed.

              CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                        I - 20
                        Fig 10: Inverse Gain Jamming


The modulation envelope of the echo signal is detected, inverted
and presented as jamming signal.               Echo and Jammer signal get
combined in space and the resultant input at Radar tracker is a
constant amplified signal. The tracker stops moving and angular
error is thus introduced. In this technique the gain of the repeater
is varied inversely with the radar signal strength received.              The
power output required from the jammer is quite modest.


b)      Conical Scan Radar : Combined Range & Angle
Deception : The        effectiveness      of    the   inverse   gain   angular
modulation technique can be increased by first pulling the range
gate away from the target echo and then applying the angle
deception.


c)      Monopulse Radar : Jamming of this type of radar is most
difficult.    The reason is that the Rxs gets all the information
required to track a target from a single return pulse. It doesnot
need a series of pulses. If cover pulse jamming is used in this case,
the monopulse radar will treat the jammer as a beacon, and the
tracking becomes much easier.

     CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                               I - 21
i)        Formation Jamming : Two or more aircrafts are required to
          be simultaneously in the beam of the monopulse.                     The
          jamming signals sent by the two aircrafts should be same.
          The tracker wanders between these two targets.


ii)       Blinking : For this two or more jammers are required. In a
          ship they can be mounted on port and start board.                   The
          jammers may employ cover pulses. The jammers are turned
          on and off in sequence. The physical separation should be
          such that they are at the extreme ends of the monopulse
          beam. If blinking is proper, the tracker will transfer from
          one jammer to another and in the process loose track because
          of over shoots of servo system.


iii)      Cross Polarisation : The jammer will sense the polarisation
          of incoming signal and transmit a cross polarised signal that
          is much stronger than echo.            When the radar receives a
          signal that is polarised at right angles to the polarization of
          radar    transmitter,     erroneous      angular      information    is
          generated.


iv)       Cross Eye : This consists of two sets of repeaters arranged
          so that the two transmitting antennas as viewed from the
          radar appear to be 180° out of phase. The two transmitting
          antennas are located at some distance apart. They can be
          mounted on the farthest wing tips of an aircraft. The signals
          from the two repeaters will be 180° out of phase when they
          reach the radar tracking antennas independent of the
          direction to the radar. This causes a null in the combined
          response of the radars sensors just where the radar tracking
          circuit would expect a peak. If these is a null, where these
          is supposed to be a peak the tracking signal will be distorted
          and the angle information provided becomes erroneous.


       CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                 I - 22
       d)         CW Radar – Velocity Gate Pull Off (VGPO) :                   CW
       Doppler type radars are jammed by stealing the velocity gate. By
       changing the frequency of the repeater by a sawbooth modulation, a
       VGPO is accomplished.


4.5    Expendable Jammer
       Expendable devices include chaff and decoy :


       a)      Chaff : Chaff is an extremely effective EA device. Chaff is
made of aluminium strip or aluminium coated nylon or fiber glass. It is
packaged in small units. They are light enough to carry and dispensed in
large quantities during a sortie. Chaff consists of large number of dipole
reflectors which are designed to match the half wavelength of victims
radar’s frequency.      It re-radiates the EM energy and create false radar
echo. Chaff is made to respond to no. of radar frequencies by simply
packing different lengths of dipoles in the same bundle. An aircraft can
safely pass through a air defence net, by creating a multitude of echoes
over a large area by confusing the radar operators by dispensing the chaff
in a burst or in random.


       b)      Decoy :     By duplicating the speed, altitude and course of
penetrating aircraft, a deceptive target can be introduced into the radar
system. This could be a small expendable aircraft like vehicle / glider
launched from the mother aircraft.             They carry corner reflectors to
enhance their RCS (Radar Cross Section) and provide a echo return
similar to a penetrating aircraft.


4.6    Airborne EA
       Airborne      EA      consists     of    a)    Self     Protection   Jammer
b) Stand off Jammer.




            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                      I - 23
a)     Self Protection Jammer (SPJ)
       SPJ is illustrated at Fig : 11




                              Fig : 11 SPJ and SOJ


Self Protection Jammer is provided on platform being targeted by a radar.


b)     Stand Off Jammer (SOJ)
       SOJ remains out of range of radar but consists of high power Jammer
to protect platform that has entered the enemy’s radar range for military
operations.4.7         J/S RATIO
       J/S is the ratio of the jamming signal strength to the strength of the
desired signal. Fig : 12 depicts the J/S ratio diagram




                 Fig 12 : J/S Ratio Receiver Pass Band


       The desired J/S required for effective jamming can vary from 0 to
40 dB depending on the type of jamming and modulation. 10 dB J/S is a
reasonable figure in many applications. J/S is directly proportional with
a) Jammer Power, b) Jammer Antenna Gain c) Radar to target distance 4

           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     I - 24
and inversely proportional with a) Radar Power b) Jammer-to-Radar
distance 2 c) Radar Antenna Gain.


4.8   Frequency Memory Loop
      The most common method of implementing RGPO is to use a
Frequency Memory Loop shown at Fig : 12a




               Fig 12a : Frequency Memory Loop




      A portion of the received pulse is stored in the delay line through
which it is recirculated for the memory period ‘td’, which is less than
shortest pulse to be handled. The initial position of the pulse is amplified
in the input limiting amplifier and passed through the delay line.       As the
signal reaches the line output, the switches are thrown open and the pulse
with length ‘td’ is recirculated again and again through the memory loop.
Gated Amplifier is used to give the output at the appropriate time and for
the desired pulse duration.



          CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                    I - 25
4.9    Digital RF Memory (DRFM)
       Various defects in the analog FML is overcome in the DRFM. The
RF is first converted to IF, sampled and a high speed A/D is used. The
digital output of A/D can be stored without any degradation for any length
of time. The signal is replayed from memory whenever needed and up-
converted to the transmission frequency using the same LO which is used
for down conversion.       Present DRFM has an IF of 600 MHz and the
limitation is mainly the speed of A/D required, since if the signal is to be
reconstructed the sampling should be twice of the input frequency. The
DRFM is a very versatile tool to obtain various types of modulations for
EA system.


4.10   EA Techniques Vs. Radar Tracking Techniques :
       There are multiplicity of EA techniques. They generally fall into
either noise (active) or deceptive (active) or chaff / decoys / (passive).
The answer to “what technique is required to be used against which
radar?”   is not easily answerable.         It is only combination of no. of
techniques judiciously used that is going to make the EA effective. With
the advancements in radar the complexity of integrating these techniques
becomes a complex job. A computer controlled ‘Technique Generator’
system will be an effective tool to incorporate the automatic application of
these techniques either independently or in combination.



4.11 Assessment OF Jamming Effectiveness Against Radars
       Presently the following two methods are mostly used :


i)     Algorithm Method : Space energy relation is computed alongwith
jamming equation. Range where jammer is effective is calculated. For
this transmitter power, antenna gain and jamming signal shapes are taken
into consideration along with the radar characteristics.           This method is
suitable for one to one case.




          CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                    I - 26
ii)    Weapon System Effectiveness Method : Radar and its EP
characteristics are part of the weapon.            A such their effect on weapon
effectiveness should also be considered. The probability of survival of a
fighter plane depends on the jamming system effectiveness index.
Effective index is computed by statistical methods using simulation
techniques.      For this earlier figures of loss rates that have taken place
during combat are made use of.


4.12   Future Trends in EA Systems
       A desired jammer system should consist of reconnaissance,
warning receivers and jamming radiations, all of them integrated and
controlled by a computer. It will have large power, wide frequency band,
fast response, capability to jam multiple targets and self-adapting
capability.


i)     Power : Power is the basic index of the jammer. There are three
ways to raise power a) Raise the power of single TWT from the present
value b) Install no. of low power TWTs in each element of a phased array
antenna.      With large no. of these TWTs, a higher ERP is obtained by
combining the power in space.             By judicious control of amplitude and
phase the direction of the beam can be controlled. Such systems are now
known by MBJ (Multi Beam Jammer) Phased Array Systems. However,
the Phase Array Jammers work in limited band of frequencies c) Develop
Microwave Broadband Solid State Amplifiers and use Solid Array
jamming system. With passage of time, power outputs from these Solid
State Amplifiers are dramatically increasing.


ii)    Power Management : More efficient power management is
required in view of limited power available especially in airborne
applications.      The power transmitted should be with in the detection
bandwidth of victim radar, and also should be in the region of the radar.
This ensures maximum interference.



              CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                        I - 27
iii)    Modulation Techniques : Appropriate and effective modulations
are transmitted to confuse or deceive the radar. Real time jamming effect
is carried out and automatically strategies changed dynamically. This is
necessitated in view of very complex advancements in the EP fields of
radar. In order to jam CW and agile frequency radars, frequency tracking
velocity and accuracy are required to be enormously increased.


iv)     Frequency Band : Very broadband frequency jammers working
from MHz going to millimetric way frequencies are                 necessary   since
MMW Missile Launch Radars are already in existence.


v)      Sensor Integration : Integration of data from ES, infrared / optics
and missile warning Rxs is necessary to arrive at faster decision to deploy
the required jammers with necessary modulations, launch chaff, flares,
decoys etc., either singly or combinedly.


vi)     Stealth : Stealth vehicles have very low RCS. Any EA system on
board, these vehicles will be required to operate in a manner that prevents
radiation giving way to detection of stealth vehicles.             EA systems are
required to be separated before they are activated.


vii)    Artificial Intelligence : Fixed algorithms for EA management
become obsolete with advancement of radar.               The ability to learn and
adjust to dynamic changing environment should be incorporated in future
EA systems. Learning algorithms are required to be incorporated.


viii)   Integrated Distributed EA : There may be no. of EA systems
deployed in a hostile area of conflict. They presently work autonomously.
If co-ordination and integration is introduced between these stations, this
will lead to effective use of EA resources. One way is to employ a stand-
off vehicle like AWACS to monitor, determine and deploy the type of
technique etc., to jam each radar.




           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     I - 28
5.0    Electronic Protection (E.P) :
       E.P is also known by ECCM (Electronic Counter Counter
Measure). E.A systems are used against EP system and Vice versa. The
anti jamming techniques used in radars constantly confront the EA
techniques used in Jammer systems. This is like a chess game, which is
never ending. We can only expect it will become more vigorous in future.
Radar designers would strive to preserve the capabilities of Radar while
the EA designers’ tries to deny its capabilities.



5.1    Radars:
       All radars operate on the basis of a few principles. Radar broadly
consists of a Transmitter, Receiver, a processor and a Display Indicator.
All radars can be classified into three categories based on functionality
(i) Search Radars (ii) Track Radars (iii) Track while scan Radars.
Further they can be subdivided into (i) Early Warning Radars (ii) Height
Finding Radars (iii) Acquisition Radars (iv) Fire Control Radars (v)
Terminal Guidance Radars (vi) Missile Fuze. Based on the method of
transmission radars can be categorised into (i) Pulse (ii) Pulse Doppler
(iii) CW Doppler radars.


5.2    Radar Concept :
       Radar basically have three main functions (i) Measure the Range
of the target (ii) Measure the angle at which the target is located (iii)
Measure the velocity of the target.


(i)    Range     :      The measurement of range is based on the simple
principle ie. measurement of time elapsed from the transmission of the
signal to detection of the reflected signal at the radar. Since Radar signals
travel at the speed of light, the time elapsed in travel, will give the
distance between the radar and the target.


(ii)   Angle :       When the main beam of the antenna is pointed in the
direction of the target, return echo is received at the radar receiver. This
           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     I - 29
is taken as the angle of location of the target.              The angle at which
maximum signal power is received is the target azimuth angle. Normally
Search radar provides the azimuth angle alone.              The tracking antenna
searches a volume of space in azimuth and elevation and determines the
azimuth and elevation angles. Since the data rate of the tracking radar is
very high compared to that of search radar, high angular accuracies are
obtained, which is necessary for fire control systems.


(iii)   Velocity: Doppler Radars are used to compute the velocity of the
target. FMCW is an example.


5.3     EP Techniques:
        EP involves action taken to ensure the effective use of EM
spectrum despite the use of EA by the enemy. The types of EP techniques
available are very large in number. These techniques are integrated into
radar systems in four primary areas (i) Transmitter (ii) Receiver (iii)
Antenna (iv) System as a whole. These techniques are also called fixes.
Fixes are features built into a radar system, or modification to existing
system that overcomes the effectiveness of EA.


5.4     Transmitter Fixes :
(i)     Frequency Agility : The frequency of operation of radar is changed at
a very fast rate and the change in frequency is as wide as possible. This will
make Spot Jammers ineffective.            Frequency agile radars compel the EA
system to use barrage or Sweep Jamming which are less power efficient.
(ii)    Diplexing : Diplexing is the use of two transmitters and two Receivers
whose frequency of operation is quite apart say X and Ka band. Here if one Rx is
jammed, output from the other is used for tracking. Diplexing forces the Jammer
systems to divide the power between the frequencies.


(iii)   Power Add : Transmission of same frequency simultaneously by two
transmitters increases the power output of radar. This extends the burn – through
range of the system and increases the requirement of J/S ratio.


           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     I - 30
(iv)    Long Pulse Duration : Increasing PD means increase of average
power resulting in higher echo strength. However with higher PD range
resolution decreases.


(v)     Pulse Compression : This increases average transmitted power without
an increase of peak power, and with no loss of range resolution. The pulse is
stretched during transmission and the echo is compressed in the receiver. With
proper coding this is a very effective measure against EA systems. The radars
using this technique are called LPI (Low Probability of Intercept) radars. Since
the jammer will never know the coding employed by the radar, this EP technique
is very effective against Jammers.


(vi)    Staggered PRF : The jammer sends synchronous Jammer pulses,
which will be received as targets at different ranges, if the radar uses
staggered PRF.       Video integration at Radar can distinguish the false
jamming signals and are easily removed from the Rx.


(vi)    Jittered PRF : This is similar to staggered PRF except that the pulse PRF
        varies randomly and is used to overcome the Synchronous Jamming.


5.5     Receiver Fixes
(i)     Manual Gain Control : At some ranges it is possible echo is stronger
than the jammer signal, but the receiver may be saturated. The gain of the Rx can
be reduced manually until the Radar display indicates the target echo properly.


(ii)    IAGC ( Instantaneous Automatic Gain Control) :                     In case a
noise or CW jammer is used against the radar, IAGC samples the average
noise level at the output of the receiver and by raising or lowering the IF
gain the output level is maintained constant.


(iii)   Logarithmic Receiver : Small signals such as the radar echo will have
high amplification and large signals such as jamming signals, receive low
amplification.

            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                      I - 31
(iv)      Fast Time Constant (FTC) : FTC uses a time constant that is just
longer than the radar pulse.          Normal target return pulses pass without
distortion, whereas longer pulses or clutter from Jammers are reduced in
length.


(v)       Dicke – Fix Receiver : The Dicke – Fix Rx contains a wideband
amplifier, a limiter and narrow band amplifier. The jamming signal is
amplified along with the echo, in the Wideband amplifier. Because of
wideband, the ringing is reduced. The amplified signals are limited ie.
Noise and echos are held below a set amplitude. The limited signal is fed
to a narrow band amplifier. This is turned to the Center frequency of the
return pulse.      So noise in the Narrow band filter will receive less
amplification than the echo signal. The Dicke – Fix Rx is used to reduce
noise, fast sweep and narrow pulse jamming.


(vi)      Pulse Width Discrimination (PWD) : This is a technique used to
discriminate the radar echo pulse from shorter or longer Jammer pulses. Control
signals suppress the incorrect length pulses.


5.6       Antenna Fixes
(i)       Side Lobe Cancellation (SLC) : All directional antennas have sidelobes
of various strengths. Energy from Jammer entering sidelobe is displayed on the
scope at the azimuth indicated by the main lobe. A SLC system consists of an
omni antenna and a receiver signal from omni channel is subtracted from the main
lobe signal so that the sidelobe returns are cancelled. Only target returns are sent
to the Video amplifiers and to the Radar Display.


(ii)      Sidelobe Blanking (SLB) : SLB eliminates unwanted sidelobes returns by
a blanking technique.       As soon as sidelobe jamming is detected, a gate is
generated that turns off the main receiver. SLB will cause the loss of valid
targets, SLB may still be effective because it blanks the main receiver only for
short time intervals, this minimizes the loss of valid target data to the SLB.




             CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                       I - 32
5.7     Future Trends in EP
        Among the important development that are taking place a few are
given below.


 (i)    High Transmitter Power : The ERP of EA systems have direct relation
on the amount of Transmitted power by the radar. Large Radar powers will
increase the required ERP of EA systems, increasing the complexity, size and cost
of EA systems.


(ii)    Electrically Steerable Antennas : Phased arrays have given greater
flexibility in antenna pattern generation. This allows high speed beam steering
and allows nulling of externally generated intereference. EA systems will not be
able to predict the beam positions which is required for effective use of EA
techniques.


(iii)   Sensor Fusion : Other sensors like EO/IO sensors, ES sensors are
used to complement or supplement the radar system, especially when the
radar performance is degraded on account of jamming.


(iv)    Multi Static Radar : Radar Receivers are located far away from the
transmitter. This provide greater ranges of detection by placing the receivers
nearer to the engagement line, with this bistatic/multistatic radar, location of the
transmitter is useless since EA is intended for the radar receiver.


(v)     LPI : LPI radars transmit signals which are not detectable by ES
systems. This denies the capability of EA systems for effective jamming.


(vi)    Ultra Agile Carrier Frequencies : Random pulse to pulse frequency
agility is the most difficult to counter with EA systems.


(vii)   Deceptive Transmissions : Radar designers suggest modulating the radar
transmitted signal. This confuses the EA receiver analysis circuits. Amplitude
modulations on the transmitted signal are erroneously detected by the EA receiver
as the mutation frequency.
              CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                        I - 33
(viii) Intra Pulse Modulation : LPI radars use frequency modulation or phase
modulation within the pulse. This complicates the RF memory circuits, so to
maintain coherency, Digital RF memories are required which are very complex
and costly.


(ix)     Ultra Low Side Lobe Antennas : Research is yielding results and ultra
low side lobe Antennas are being used in expensive radars. This necessitates the
EA Jammers to jam through mainlobe of the radar antenna thereby requiring large
power.


  (x)    Multifunction Antenna : Single Antenna is used to perform the job of
several sensor functions. If the antenna is designed to include a passive mode of
operation, the radar will track the EA jammer signal without use of radar
transmission.     Operating passively on the jammer signal during periods of
interference will provide the required angle information of the jammer mounted
on the target.


6.       Conclusion
         During the past 50 years, an extremely hostile environment has
developed with proliferation of radar guided AAA guns, Surface to Air
Missiles, Air to Air Missiles etc. This threat environment has necessitated
enormous budgets being spent on development of Electronic Warfare
equipment of wide varieties.            Further technological innovations in the
filed of radar and its associated anti jamming techniques, has brought out
the requirement of further innovations in the field of Electronic Attack
systems both in Microwave, Milli-metric and Electro-optic and IR range
of frequencies.        The tussel between the radar designers and the EA
planners is going to intensify. This will culminate in the development of
new systems, techniques, components, miniaturization etc.                    Future
systems will be more software intensive using artificial intelligence for its
adaptability and survivability. Mulit sensors systems will be invariably
used with new data fusion algorithms.                 The future for EW is very
promising with immense challenges to be met in the days ahead.


              CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                        I - 34
            CHAPTER – II


ELECTRONIC SUPPORT SYSTEMS

      Shri. B.R. GANDHI, Sc-‘G’
                  &
      Shri. R. RAMA RAO, Sc-‘F’




  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                              II
                               CHAPTER –II



               ELECTRONIC SUPPORT SYSTEMS



                                 CONTENTS


1.    INTRODUCTION

2.    FUNCTIONS OF AN ESM SYSTEM

3.    FEATURES OF A MODERN ESM SYSTEM

4.    ESM SYSTEM CONFIGURATION

5.    THREAT ENVIRONMENT

6.    DESIGN CONSIDERATIONS OF AN ESM SYSTEM

7.    DESIGN OF AN ESM SYSTEM - A CASE STUDY

8.    ADVANCED ESM SYSTEMS

9.    CONCLUSION

10.   REFERENCES




            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                       II -1
                     ELECTRONIC SUPPORT SYSTEMS

                       R.Rama Rao,Sc’F’ and B.R.Gandhi, Sc’G’
                                DLRL, Hyderabad


1. INTRODUCTION


        Electronic Warfare is the practice of technical opportunism and expediency,
exploiting weakness in an enemy’s use of electronics for his weapons and sensors, and
cleverly taking advantage of features of enemy equipment design or his use of electronic
equipment.


EW is classified into three categories:


        Electronic Support Measures (ESM): ESM is a passive EW System that uses
enemy      transmissions   to   support   surveillance    operations    by       monitoring   the
electromagnetic (EM) spectrum.


        Electronic Counter Measures (ECM): ECM is an active EW involving actions
taken to prevent or reduce the enemy’s effective use of the EM spectrum or to destroy the
enemy’s forces.


        Electronic Counter-Counter Measures (ECCM): ECCM is also an active EW
involving actions taken to ensure friendly effective use of the EM spectrum despite the
enemy’s use of ECM.


The above EW definitions are broadened in the present-day electromagnetic scenario as
follows:


        EW is also defined as a military action involving the use of electromagnetic
energy to determine, exploit, reduce or prevent hostile use of the EM spectrum and action
which retains friendly use of the EM spectrum.




                  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                             II -2
          Electronic Support (ES): ES focuses on action taken for the purpose of real-time
threat recognition in support of immediate decisions involving Electronic Attack (EA),
Electronic Protection (EP), Weapon avoidance, targeting or other tactical employment of
forces.


          Electronic Attack (EA): In addition to ECM, Electronic Attack function includes
the use of directed energy weapons (lasers, microwave radiation, particle beams), anti-
radiation missiles and EM pulse (nuclear weapons, destruction of electronics) to destroy
enemy electronic equipment.


          Electronic Protection (EP): In addition to ECCM, Electronic Protection is
expanded to include the use of such measures as Electromagnetic Control (EMCON),
electromagnetic hardening, EW frequency deconfliction and communication security
(COMSEC).


          Defence Suppression: Electronic attack against military surveillance and tracking
radars generally involves lethal actions as well as electronic suppression techniques.
These lethal actions have as a primary objective, the physical destruction of a radiating
emitter, which usually is a component of an enemy defence system. The most prominent
lethal defence suppression system is the ARM/HARM.


          In this paper, an overview of ESM systems is discussed covering the Receiver
techniques, the design aspects and the design of an ESM system in detail.

2. FUNCTIONS OF AN ESM SYSTEM

          An ESM System receives the radar pulses emitted by the various radars in the
scenario, measures the parameters of these pulses and then uses these parameters to sort
out from which of the radars the pulses emanated. Finally, the emitter is identified and
information is presented to the operator. From identification of the radars present, an
inference is then made as to which platforms are present in the scenario and of their
deployment, thus facilitating surveillance operations to concentrate attention on more
stressing radars and the tactical deployment of own forces for necessary follow-up
actions. The key functions of the ESM system are illustrated in fig 1.


                   CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                              II -3
                                             Pulse de-interleaving, emitter
               Freq, Bearing PW,             association, scan analysis,          Operator
               TOA, Amplitude etc            library identification               Interface


    Receive             Measure                     Recognize                      Display




                             Fig 1: Functions of ESM system

2.1 ESM Vs SIGINT

       ESM is for tactical purposes that require immediate actions as contrasted with
similar functions which are performed for intelligence gathering such as SIGINT and its
constituent parts of ELINT, COMINT and RINT.

       ELINT deals with non-communication (radar) signals.
       COMINT deals with communication signals.
       RINT (Radiation Intelligence) deals with the intelligence derived from potentially
       hostile communications and weapon systems by virtue of their unintended
       spurious emissions even when in a non-transmitting mode of operation.


       ELINT/COMINT Receivers are used for intelligence collection purposes, which
generally involve subsequent or non-real-time analysis of the intercepted data.

2.2 ESM Vs Radar Warning Receiver (RWR)

       ESM systems are differentiated from RWR in that the ESM systems are more
complex and are used for reconnaissance or surveillance missions to map enemy radar by
measuring all the parameters of the radar pulse. The RWRs are in general less complex
and mainly used to warn of attack by SAMs & AAMs and anti aircraft gun systems.
RWRs can be classified as the simplest form of ESM system, which is preset to cover the
bands/characteristics of expected threats. The operator has a display showing functional
radar type, direction (course) and relative range. The identification algorithms used in
RWR for threat warning might be based on the information gathered by ESM system.




                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                            II -4
        However, the difference between ESM & RWR is narrowing down as the
technology is advancing to include more features, if not all, of ESM system in RWR.


        ESM Rx can be treated as the combination of possessing ELINT and RWR
features.

3. FEATURES OF A MODERN ESM SYSTEM

        A modern/advanced ESM system should possess mainly the following
capabilities:


1. Capable of handling complex/exotic radar signals such as overlapping, agile,
    stagger/jitter, pulse compression, LPI etc.
2. Instantaneous coverage in spatial and frequency domain ensuring 100% probability
    of interception.
3. High through put rate (processing speed)- One million pulses per second.
4. High sensitivity to provide adequate Range Advantage over the radars.
5. High DF accuracy to facilitate accurate fixing of the intercepted radar.
6. Accurate measurement of frequency and other pulse parameters to enable proper de-
    interleaving of the intercepted emitters precisely in a dense RF signal environment.
7. Low reaction time to present the intercepted radars data quickly to the operator on the
    system display in the required MMI format.
8. Compact in size and weight for adoption on the to variety of platforms.

4. ESM SYSTEM CONFIGURATION

        The basic configuration of an ESM System is shown in fig 2.




                  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                             II -5
  Omni Antenna
                              RF

                                       Freq Meas        Freq   P               P
                    RF FERx            Superhet/               U               R   Emitter File
                                       DIFM                    L               O
DF Antenna
                                                               S               C
                    Video                                      E               E
                                       DF Meas                                 S
  1                 RF FERx            ADF/PDF/         DOA    F      PDW                    DISPLAY
                                                                               S
                                       TDOA/                   O               O
                            V          RDF              PW,    R               R
                                                        PA,    M
                                       --------------   TOA    A
  n
                  RF FERx              Pulse Meas              T
                                       PW, PA,
                                       TOA


                                   Fig 2: Configuration of ESM System

The constituents of an ESM System are


1. Antennas
             Omni
             Directional
2. Receiver Front End
3. Frequency Measurement Receiver
4. DOA Measurement Receiver
5. Pulse Parameter Measurement Unit
6. Pulse Formatter
7. Processor
8. System Display


       The RF signals transmitted by radars are received by the antennas. The antennas
are of two types – Omni and Directional.


       1. The Omni antenna produces an uniform (circular) response in all directions
             and the RF signal coming out of this antenna is used for frequency
             measurement.




                    CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                               II -6
       2. The Directional or Direction Finding antennas exhibits directional
           characteristic, which is used to determine the Direction Of Arrival (DOA) of
           an intercepted signal.


       The next stage after the antennas is RF Front End Receiver. Each antenna is
followed by a Front End Receiver (FER). This provides necessary RF amplification and
video detection & amplification to enable the measurement of the intercepted signal
parameters by the subsequent stages.
       The Frequency measurement is done on the RF signal intercepted by the Omni
antenna in the omni chain. The techniques used for frequency measurement in general are
Superheterodyne and Instantaneous Frequency Measurement. Selection of the type of
receiver depends upon the system specifications/requirements. If high sensitivity and
frequency measurement accuracies are of prime concern, superhet is the choice. On the
other hand, if instantaneous coverage over the large frequency band of interest is desired
- the option is IFM/DIFM. The DIFM is mostly used in wide-open ESM Systems.


       The Direction Finding, also termed Direction of Arrival (DOA) or Angle of
Arrival (AOA) of the emitter is computed from the video signals provided by the FERs
used in the DF chain. The choice of DF measurement technique again depends on the
system specs/requirements. The technology options available are : Amplitude comparison
DF(ADF), Phase comparison DF(PDF), Time Difference Of Arrival DF(TDOA/TDF)
and Rotary DF(RDF).


       Another measurement stage is the pulse parameters measurement unit in which
Pulse Width (PW), Pulse Amplitude (PA) and Time of Arrival (TOA) of the intercepted
RF signal are measured.


       Next stage is the pulse Formatter, which receives all the parameters of the
intercepted emitter, pulse by pulse, formats these parameters in the form of Pulse
Descriptor Word (PDW) and passes it on to the processor.


       The most important stage of an ESM System is the Processor. The Processor
receives the interleaved parameters of the intercepted signals pulse by pulse, de-


                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                            II -7
interleaves the pulse trains, measures PRF of the de-interleaved signals emitter wise,
compares these emitters with the pre-programmed parameters in the library and provides
necessary identification and threat warning.


       The emitter files containing all the information of the intercepted RF Signals are
then passed on to the system Display for Visual Presentation to the operator. The emitters
are presented in different formats, eg., tactical mode, tote mode, f-α mode etc, to facilitate
easy interpretation of the data by the operator and also provides the necessary command
& control interface to enable easy & quick Man-Machine Interface (MMI).

5. THREAT ENVIRONMENT

       The frequency range of most weapon-related radar signals is in the microwave
region of 2 to 18 GHz. There is also minor activity in the millimetric wave window bands
around 26 GHz, 36GHz, 94GHz, 140GHz and 220 GHz. These frequency bands are
associated with short-range weapons, and hence most mm wave extensions of ES systems
extend to only 40 GHz. Also there is activity in the EO/IR laser radar bands.


       Also, there are radars in the VHF & UHF bands. These are generally search radars
having low resolution and are not threats to conventional military platforms. However,
the ability of these radars to detect stealth targets (LPI radar) has increased the importance
in the 80 MHz to 1 GHz band, because they can designate targets for attack by fire
control systems.


       The search radars in the lower microwave region in 0.2 to 2GHz are important
since they provide initial acquisition of targets for weapon systems. In some EW systems,
these signals are needed to set on Jammers and the ESM frequency coverage range is
further lowered to 200 MHz to intercept these radars.


       Most of the radars are of pulsed type, with PW varying from 50 ns to hundreds of
µs. The radars with longer PWs are generally of the pulse compression type. ESM
systems have difficulty in decoding these types of signals since they must be processed by
a matched filter whose exact characteristic is unknown to the ESM system.




                   CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                              II -8
         Radars of pulsed Doppler type have high-duty cycles producing up to 500,000
pulses per sec. The large number of pulses thus produced overload the ESM signal
processor. To combat this, special receivers are employed that search for and lock on to
high-duty factor signals. These signals are then processed separately and removed from
the signal set (i.e. by trapping). The remaining signals which are mostly pulsed can be
processed by the usual ESM signal processor.


         PRF ranges from 50 Hz to 500 KHz. They can be stagger or Jitter type. MTI
radars use Jittered PRF that varies from pulse to pulse.


         LPI radars signals are severely mismatched to the ESM systems. A sensitive ESM
system     (-80 dbm) can detect these signals, but at a considerably smaller range that can
be detected by the radar. This is the essence of LPI operation.


         A general solution to the LPI radar intercept problem for the ESM system would
be to employ a matched filter that is tuned to the LPI radar’s waveform. This is generally
difficult to implement due to the presence of other signals that occur simultaneously with
the LPI signal. Also, LPI radars employ high duty cycle waveforms that are difficult to
process in ESM systems.


         Against modern missiles, the warning and countermeasures launch time is
measured in seconds for aircraft and tens of seconds for ships and other surface vehicles.
This requires an automatic response against most missiles directed against a platform
since there is not enough time for human intervention. To respond with a countermeasure,
the type of weapon and its location and velocity must be ascertained. For RF guided
weapons, this information is provided by ESM system. IR/EO guided missiles require a
Laser Warning Receiver (LWR)/ Missile Approach Warner (MAW) system.

6. DESIGN CONSIDERATIONS OF AN ESM SYSTEM


         Refer section 3   in which the requirements of a modern ESM system are listed.
These are discussed in detail in this section.




                  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                             II -9
6.1 Handling of complex/exotic signals

       An ESM system should be able to process the following types of signals which are
expected to prevail in the environment.


                              -   Pulsed
                              -   Continuous wave (CW)
                              -   Stagger / Jitter
                              -   Chirp (linear FM)
                              -   Time Overlapped
                              -   Pulse compression
                              -   Low probability of intercept (LPI)
                              -   Ultra Low pulse width (impulse)


       The first four types can be handled by conventional ESM receivers namely Crystal
Video (wide-open) and superhet receivers (narrow band). The other types of signals are
complex in nature and can be processed by employing advanced receiver techniques.

6.2 Probability of Interception (POI)


       The probability of interception refers to the probability that an emitter signal is
detected and processed, its parameters measured and presented to the operator on the
display with minimum false alarm.


       There are three domains of interceptions viz, spatial, spectral and temporal. If the
spatial coverage in azimuth is θc and the total coverage expected is θt, the spatial POI is
given by
               Pθ = θc / θt


       Pθ can be made 100 %, if θc = θt instantaneously, i.e., at any given instance of
time, the coverage should be θt and not limited to the beam width as in Rotary systems.




                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                           II -10
       In the frequency domain, if the RF bandwidth is BR and the bandwidth covered at
any instant is BT, the POI in frequency domain is


               PBW = BR / BT


       PBW can be made 100 %, if BR = BT

       In the time domain, there should be no hindrance for receiving the signals at
anytime. If for some processing constraints or for pulse overlap in time one is not able to
detect and measure pulse parameters for a time Ta and the total time of observation is ‘T’;
the POI in time domain


               PT = Ta / T


       The overall probability of intercept is


               POI = Pθ * PBW * PT


       Even if Pθ = 1, PBW = 1 and PT = 1, still one can have a POI not equal to 1 (100
%),When the signal is beyond the dynamic range of the system. However, in the case of
the POI, one should consider signals within the dynamic range of the system.


6.3 Sensitivity and Dynamic range


       There are three types of sensitivities, which can be defined for an ESM system.


                               -   Tangential signal Sensitivity
                               -   Channel Sensitivity
                               -   System Operational Sensitivity


       There is a need to distinguish them as there is always some confusion as regards
       their definitions.




                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                           II -11
6.3.1 Tangential signal sensitivity (TSS)


       It is defined as the minimum RF pulsed signal required at the input to the receiver
channel so that the pulsed bottom portion is just tangential to the top of the noise floor. At
this RF level, the signal to noise (S/N) ratio is 8 dB as measured at the video . The
sensitivity is given by the following expression:




                TSS




Minimum Detectable Signal (MDS) = -114 + 10 log (√ (2 BrBv) ) + Front-end NF
        Where Br = RF signal bandwidth, Bv = video bandwidth


       TSS is 4 dB above the MDS
       TSS = -110 + 10 log (√ (2 BrBv) ) + Front-end NF


6.3.2 Channel sensitivity


       It is the minimum RF. level at the input of the receiver channel for which a stable
measurement of frequency, DF, Pulse Width, Pulse Amplitude and TOA are obtained.
The channel sensitivity does not include the gain of the antenna and the plumbing losses
between the antenna and the receiver input stage.


6.3.3 System Operational sensitivity


       It is the minimum RF signal level intercepted by the antenna for which a stable
pulse track is presented on the system display.




                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                           II -12
6.3.4 Dynamic range


       The Dynamic Range is defined as the difference between the operational
sensitivity level to the maximum signal level between which a stable track is presented on
the system display.


6.4 Signal density/Throughput rate


       With wide variety of Radar signals present in the environment and number of
radars operating simultaneously, an ESM system will have to encounter signal densities
of the order of 1 million pulses per second or even more. To meet this high signal
densities, ESM system should have high throughput rate (low acquisition & processing
time). The high-density signals are asynchronous and occur randomly, thus creating
simultaneous (overlapped) signal scenario at the receiver input. The ESM system must be
capable of implementing suitable algorithms and processing techniques to resolve
simultaneous signals in order to prevent the loss/corruption of the data of the intercepted
signals.


6.5 Frequency Coverage


       The frequency spectrum to be encountered by an ESM system has been already
covered under threat environment (ref. Section-5). The radar frequencies that an ESM
system should handle are mostly from 0.2GHz to 18GHz. There is not much activity in
the 18 – 40 GHz spectrum except at few frequencies around 26GHz and 36GHz. Separate
narrow band receivers are used to cover these signals.


6.6 Parameters measurement accuracy


       Segregation (De-interleaving) of radar signals is done purely based on the signal
parameters that are measured by the receiver. The efficacy of de-interleaving depends on
the measured accuracy of the parameters. The important parameters are


                              -   Frequency


                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                           II -13
                               -   Direction of Arrival (DOA)
                               -   Pulse Width (PW)
                               -   Pulse Repetition Frequency (PRF)


       Because of the high density of signals, the variation of these parameters from
radar to radar is marginal and hence demands for high measurement accuracy. The
present day requirement of these accuracies and their ranges are given below


       Parameter       Accuracy                Range


       Frequency       <1 MHz                  0.2 to 18 GHz
       DOA             <1 degree               0 to 360 degrees
       PW              <10nsec to 1%           20 nsec to few msec
       PRF             <1 %                    50 Hz to 1 MHz


6.7 Scan analysis


       Most of the radars are of Track While scan, Sectoral scan or Circular scan types.
Missile tracking radars are in general Lock-on type. The ESM system should incorporate
the appropriate algorithms to analyze and declare the above scan types. In addition to the
signal parameters, the scan analysis is essential to identify the type of radar.


6.8 Reaction time


       Reaction time of the ESM system is defined as the time taken by the system from
the instant a signal is detected to the time at which the data is presented on the system
display.


       The reaction time should be <1sec so that the intercepted emitter is immediately
observed by the operator for necessary counter measure action initiation.




                  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                            II -14
6.9 Cost, Weight & Complexity


        These factors are governed by the system specifications and requirements which
are in-turn depends on the platform constraints like the availability of space and allowable
weight on the platform. One has to choose the optimum configuration keeping in view of
all these factors.


7. DESIGN OF AN ESM SYSTEM - A CASE STUDY


        The basic configuration of an ESM System and the factors that have to be
considered while designing it are explained in sections 4 & 6 respectively. Now, let us
take an example and design the ESM System for airborne application to meet the laid
down specifications. The design of each block (Ref fig 2) of the ESM system is described
in detail in this section.

7.1 System Definition


        Design an ESM system to meet the following specifications:


        •   Frequency range                       :       2 to 18 GHz
        •   Probability of intercept              :       100%
        •   System operational sensitivity        :       -40 dBm
        •   System Dynamic range                  :       40 dB
        •   Frequency accuracy (r.m.s)            :       6 MHz
        •   DF accuracy (r.m.s)                   :       6°
        •   Pulse width handling                  :       100 ηs to 150 µs
        •   Pulse width accuracy                  :       30 ηs / 1%
        •   PRF range                             :       100 Hz to 500 KHz
        •   PRF types                             :       Fixed, Stagger, Jitter
        •   PRF accuracy                          :       < 1%
        •   Reaction time                         :       < 1 sec
        •   Scan types                            :       TWS, Sectoral, Circular
        •   Types of signals to be handled        :       Pulsed, CW


                     CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                               II -15
       •    Size and Weight                       :     To be optimum
       •    Platform                              :     Aircraft


7.2 Selection of Antennas


       The system has to measure the frequency & DF providing 100% POI in both
frequency domain and spatial domain. For frequency channel, an Omni antenna is
required to intercept the incoming RF signal and for DF channel, directional antennas are
required.


       The required frequency spectrum of 2-18 GHz has to be covered instantaneously.
Hence, wide band antennas covering 2-18 GHz have to be selected.


       The options available for Omni antennas are


               •    Log spirals
               •    Biconicals


       Both these antennas give Omni directional patterns with a gain of 0 dBi (nominal)
over 2-18 GHz. The log spirals protrude out too much and hence not suitable for aircraft
applications. For ship and vehicle based systems, this can be used. The Biconical antenna
is smaller in size and has uniform structure. Since size and weight are of concern,
Biconical Omni antenna covering 2-18 GHz is the right selection.


               The DF antenna options are:


                        •   Cavity backed spirals (CBS)
                        •   Conical log spirals
                        •   Log periodics
                        •   Parabolic reflectors
                        •   Horns




                   CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                             II -16
CBS antennas exhibit directional characteristics having 90° beam width (nominal) and
uniform gain of 0 dBi over the entire frequency range of interest. Size and weight of these
antennas are optimum compared to other types. The shape is best suited for mounting on
aircrafts providing negligible air drag. Hence, CBS antennas covering 2-18GHz
frequency range are the choice.


To meet 100% POI in spatial domain, multiple antennas are required to cover 360°.


Min No. of antennas required = 360°/Antenna Beam Width = 360°/90° = 4


        The other types of antennas protrude out or large in size or narrow band and hence
are not suitable for aircraft applications.


7.3 Frequency Receiver


        The required sensitivity in the frequency channel is –40 dBm. Taking the Omni
antenna gain deviations (0 ± 3 dB) and plumbing / cable losses from antenna to frequency
receiver as ≈ 10 dB, the required frequency receiver sensitivity is calculated as below:


        System Operational Sensitivity         :       -40 dBm
        Plumbing / Cable losses                :       -13 dB
        Frequency Rx Sensitivity               :       -40 + (-13) = -53 dBm


        To meet the 100% POI in frequency domain, wide band (2-18 GHz) Instantaneous
Frequency Measurement (IFM) receiver having –55 dBm (min) sensitivity and 55dB
(min) dynamic range is essential. Presently wide band Digital Instantaneous Frequency
Measurement (DIFM) receivers are available which gives out frequency in digital form
with an accuracy of 4 MHz. Hence, 2-18 GHz DIFM is the proper selection for our
application. Also, this is very compact and uses present day technology in meeting size
and weight requirements.


        In general, a RF amplifier is provided to compensate the plumbing/cabling losses
before the DIFM to enhance the RF signal level intercepted by the Omni antenna. For ex,


                  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                            II -17
if the DIFM sensitivity is –40 dBm, the RF amplifier gain should be at least 15 dB. The
Frequency Rx to meet the requirements is shown below (Fig 3).


    2-18 GHz Omni antenna
          (bicomical)


                                                              12 bits
                                         2-18 GHz DIFM
                                                                        Frequency Data

                      2-18 GHz RF amplifier
                         Gain = + 15 dB

                                    Fig 3: Frequency Receiver

       The limitations of the DIFM Rx are: Low sensitivity, Low frequency accuracy
and fails to handle simultaneous (time overlapped) signals.


       At this stage, it is quite appropriate to know the other types of Frequency receivers
in brief. These are mentioned below:


7.3.1 Superheterodyne Receiver


       This is used mostly in communication receivers and radar receivers. Since the
instantaneous frequency coverage is very narrow, Superhet provides high sensitivity of
the order of –90 dBm, wide dynamic range and excellent frequency selectivity &
accuracy. The POI is less. Due to more size and weight, this is not suitable for airborne
applications.


7.3.2 Homodyne Receiver


       A Homodyne receiver is a special case of Superheterodyne receiver. The main
difference is that the frequency of the Local Oscillator is the same as the frequency of the
input RF signal. In fact, the LO is often derived directly from the input RF signal.




                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                           II -18
7.3.3 Channelized Receiver


       A Channelized receiver uses a large number of contiguous filters to sort the RF
input signals by frequency. This can be considered as a parallel arrangement of a number
of tuned Superhet receivers proceeded by a chain of band pass filters. Like a Superhet
receiver, a Channelized receiver has high sensitivity, wide dynamic range and fine-
frequency resolution. Because of the parallel nature, the Channelized receiver provides
better POI than the Superhets. The large number of parallel channels makes this receiver
bulky, highly hardware intensive and expensive which has limited its use in current EW
applications. However, Surface Acoustic Wave (SAW) devices and millimetric wave
integrated circuits (MMICs) hold promise for the future of Channelized receivers which
helps in reducing the size and cost.

7.3.4 Compressive (Microscan) Receiver


       The receivers described above cannot handle signals that are transmitted by pulse
compressive radars. These signals can be effectively processed by Compressive Receiver
using pulse compression technique. These receivers handles simultaneous signals very
effectively. The frequency spectrum intercepted by the ESM receiver is scanned at very
high speed to enhance the POI.


       The name compressive receiver is given because a disperssive delay line (DDL) is
used to compress the input RF signal to a narrow pulse. It is also referred to as microscan
receiver because a fast sweeping local oscillator is used to convert the input signals to
frequency modulated (FM) signals. The detected outputs from a compressive receiver are
narrow pulses arriving in series in time domain. By measuring the positions of these
compressed pulses, the frequency of the input signals can be determined.


       The structure of this receiver is complicated and high speed logic circuits are
required as the detected pulses are very narrow and very close in time. The advances in
Surface Acoustic Wave (SAW) devices technology and high-speed logic circuits have
revitalized the interest in developing Compressive receivers.




                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                           II -19
       The sensitivity and dynamic range of these receivers are moderately high and the
input bandwidth is moderately wide.




                CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                          II -20
7.3.5 Bragg cell or Accousto-Optic Receiver


       The Bragg cell is basically an Electro-optic device. An entire Bragg cell receiver
can be constructed on a single substrate as an integrated optical circuit (IOC) by using
modern integrated circuit technology. The technology is still under development stage.


       The structure of the Bragg-cell receiver is complicated, but the size is very small.
The sensitivity is high, and the dynamic range is low. The instantaneous bandwidth of the
Bragg-cell receiver is around 2 GHz and provides fine frequency resolution.
Simultaneous signals can be processed by this receiver.


       In EW applications, optical signal processors perform their function by spatially
modulating the phase or amplitude of an optical beam with an input RF signal. The
optical or light beam in optical signal processors can be modulated by means of an
acoustic (sound) wave; hence this is termed as Accousto-optical receiver. The modulated
light beam is then passed through a lens that performs an optical Fourier transform to
display the frequency domain characteristics of the input signal as a spatial distribution of
light energy. Therefore, an optical signal processor can serve as a real-time spectrum
analyzer.


7.3.6 Digital Receivers


       Radars employing pulse compression techniques cannot be detected by normal
receivers. The advances achieved in digital hardware and signal processing techniques
made it possible to realize digital receivers for detection and processing of these signals
by applying matched filtering and correlation techniques.


        In many EW applications, a digital receiver must be able to digitize input signals
that can occupy a wide frequency range. Currently, A/D converters do not possess
sufficient bandwidth to directly digitize these input signals. Therefore, the A/D converter
is usually preceded by a Superheterodyne receiver that down-converts the signal of
interest to an IF frequency.




                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                           II -21
       The digitized samples of the input signal are usually stored in a digital memory
where they are available for analysis. There is a vast array of digital techniques for
performing detailed signal analysis.


       An extension of the basic digital receiver produces a digital RF memory (DRFM).
DRFM allows for the storage of intercepted radar signatures (RF signal) in a digital
memory and re-construct the signal waveform, which is used widely in ECM applications
for Jamming radars. Advanced DRFMs are now being configured as ASICs, enhancing
the memory size and performance.


7.3.7 Hybrid Receivers


       To accomplish some specific missions, often one kind of receiver cannot fulfil the
requirements. The present day radar threat scenario is highly dense with complex signals
and hence demands the need for selection of a combination of ESM receivers to
encounter this situation. Based on the system requirements & technical specifications and
considering the size and weight constraints, the ESM system designer has to optimize the
configuration by selecting one or more receiver technologies. Such receiver combinations
are called Hybrid receivers.

7.4 DF Receiver


       The DF Receiver computes the Direction of Arrival (DOA) of the incoming
intercepted signals. The DF technologies available are:


           •   Amplitude comparison
           •   Phase comparison
           •   Time Difference of Arrival
           •   Rotary
           •   Digital Bearing Discriminator (DBD)


       Let us consider our requirement DF Receiver sensitivity should be –40 dBm and
the accuracy is 6°.



                  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                            II -22
          Ref section 7.2, in which the selection of DF antennas are given. Four CBS
antennas are selected, each covering 90°, thus meeting the total 360° coverage.


          The DF in general is computed based on the signals received from the adjacent
antennas. Hence, each antenna should be followed by a RF Front-end receiver consisting
of RF amplifier, detector and video amplifier.


          Amplitude comparison is the simplest of all DF techniques. In this technique, the
DF accuracy depends on how closely all the channels are matched and how uniformly the
DF antennas (CBS) exhibit matched patterns over the entire frequency range of interest. It
can be shown that a channel mismatch of 1 dB causes a DF error of 3.75°.


          Our accuracy requirement is 6°. Hence, the channel mismatch should not exceed
1.5 dB. Achieving this over a multi-octane band is very difficult. However, experimental
results show that by using channel calibration and antenna pattern correction techniques,
the mismatch can be brought down to ≤ 1.5 dB. Hence, the DF accuracy requirement of
6° can be met by using Amplitude Comparison technique. In addition to the simplicity,
the other advantages of ADF are: low cost, small size and less weight. The configuration
of a 4 antenna ADF Rx is shown in fig.4, which is basically a Crystal Video Receiver.

         2-18 GHz          Front-end Receiver
      CBS antennae
                      RF Amp
                                            Video Amp
      1
                                 Detector


      2

                                                               Video                  DOA       DOA
                                                              Processor             processor
      3




  4


                               Fig.3 : Amplitude Comparision DF Receiver


                     CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                               II -23
          The DF receiver sensitivity is determined by the RF Front-end receiver. The
required DF Rx operational sensitivity is –40 dBm i.e., a –40 dBm signal present at the
antenna surface should be detected and stable DF should be provided.


          The DF is computed based on the maximum signal and next maximum signal
intercepted by the adjacent antennas. The CBS antennas exhibits gain characteristics of
0 dB at bore sight, ≈ -6dB at 45° off the bore axis and ≈ -18 dB at 90° off the bore axis.

          Hence, when a signal is received by antenna1 in the bore sight (i.e., at 0°), the
adjacent antenna (i.e., antenna 2 or 3) signal will be –18 dB down with respect to
antenna1.

          Consider the signal level received by antenna1 is –40 dBm. Then antenna 2 or 3
signal strength will be –18 dB down with respect to antenna 1 i.e., -58 dBm.


          Assume the antenna gain deviation of 0 ± 2 dB and the cable loss of 2 dB, the
total losses becomes 4 dB.


          Therefore, the required DF Rx FE Rx sensitivity should be equal to or better than
–58 + (-4) = -62 dBm.


          This figure of –62 dBm can be achieved by providing a RF amplifier with 40 dB
gain. The detector output is then amplified by the Video amplifiers, (in general, log. video
amplifier) to give reasonable video level, which can be processed by the subsequent
stages.


          The four-video signals coming from the respective RF FE Rxs are given to video
processor. The video processor digitizes these signals using flash Analog to Digital
converters. The digital output is given to DOA computation circuit. This circuit can be
realised using ASICs or FPGAs and EPROMs. The ASICs or FPGAs provide the logic
portion. The channel calibration & antenna roll of corrections and DF algorithms are
provided by the EPROMs.


          The other DF techniques are described below in brief.


                   CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                             II -24
7.4.1 Phase Comparison DF (PDF)


       Phase Comparison DF operates over octave or slightly more than octave
bandwidths. DF is computed based on the phase difference of the incoming signal given
by the two adjacent antennas. This requires an additional coarse DF subsystem to resolve
the phase ambiguities. Hence, PDF Receivers are more complex and highly hardware
intensive resulting in more size, weight and cost .The main advantage of PDF is its high
DF accuracy of the order of 1°. This is recommended for installation on ships and
vehicles.


7.4.2 Time Difference of Arrival (TDOA)


       In the TDOA method, the DOA is computed based on the difference of time of
arrival of the intercepted signal by two adjacent antennas. The difference in the arrival
time of the RF signal at each pair of antennas with respect to the baseline formed by the
two antennas is measured which is proportional to the DOA. The difference of time
arrival in turn depends on the span (distance between the antennas) of the base line.


       The TDOA approach yields high DF accuracy, but the limitations are the
requirement of large base lines and high speed processing circuits to measure the time
difference of the order of nanosecond very accurately. TDOA approach is useful for
pulsed signals only.


       The present day technology is to use high speed ECL ICs and GaAs ASICs for
time difference measurement. Hence, for platforms where large base lines are available,
TDOA DF is highly recommended.


       Experimental results show that for baselines of 14 meters span, DF accuracy less
than 2° is achievable.




                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                           II -25
         7.4.3 Rotary DF


                This approach gives high accuracy, but suffers from Low POI. This is mostly used
         where high sensitivity and high DF accuracy is required. The rotary DF uses narrow beam
         high gain antennas (parabolic reflectors).


         7.4.4 Digital Bearing Discriminator (DBD)


                This DF technique employs DBD antenna and Butler Matrix. This is too complex
         and costly but yields good DF accuracy. The hardware is too bulky and hence not suitable
         for airborne applications.


         7.4.5. Comparison of ESM receiver techniques


                The comparison of various techniques of ESM Receivers described above are
         summarized in table 1.

         TABLE 1:       COMPARISON OF ESM RECEIVER TECHNIQUES

Performance Factor    CVR         Superhet       IFM        Chanl’d      Digital      Micro     Bragg
                                                                                       scan      cell
POI                   100%         Poor         100%         100%        Poor       50to100%    100%
Dynamic. range         Low         High         High         High       Medium       Medium     Low
Freq. accuracy          --         High        Medium        High        High         High     Medium
BW(Inst’s)            Wide        Narrow        Wide         Wide        Less         Less      Less
Time overlapped       Poor         Good         Poor         Good        Good         Good      Good
signals handling
Throughput Time        Fast       Slow to        Fast         Fast      Slow to      Medium    Medium
                                  Medium                                Medium
Ability to            Good        Medium        Good         Good        Good        Medium     Good
measure Narrow
Pulse
Sensitivity           Low           High      Medium          High      Medium         High      High
Capability to        Cannot        Cannot      Cannot        Cannot     To some      To some    Cannot
detect LPI signal    Analyze      Analyze     Analyze       Analyze      extent       extent   Analyze
Complexity/          Simple       Moderate    Moderate        Very        Very         Very      Very
hardware                                                    complex     complex      complex   complex
intensity




                           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                                     II -26
7.5 Pulse Parameters Measurement Unit


        The pulse parameters of the intercepted signal i.e., Pulse Width (PW), Pulse
Amplitude (PA) and Time of Arrival (TOA) of the signal should be measured to the
desired accuracy.


        The pulse parameters are measured on the detected Video in the Frequency
receiver channel as shown in fig.5


7.5.1 Pulse Amplitude (PA) measurement


        The RF signal coming from Omni antenna is detected. The detected video is
amplified to meet the required dynamic range of 40 dB. That is, at minimum input signal
level of –40 dBm, the video amplifier output should be reasonably high (video S/N ratio ≈
12 dB minimum) with respect to threshold level and at max input signal level of 0 dBm,
the video amplifier should provide unsaturated video signal level. This video is converted
to digital by A/D converter to represent PA in digital form.


7.5.2 Pulse Width (PW) measurement


        The minimum PW to be measured is 100 ns and the accuracy at lower pulse width
is 30 ns. These two factors dictate the resolution. A 40 MHz clock gives 25 ns resolution
and this is sufficient for our application.


        The max PW to be measured is 150 µs. The counter size is decided by this factor.
A 13-bit counter represents 8192 counts. Each count being 25 ns, 13 bits represents 8192
* 25 ns = 204.80 µs. A 12 bit counter represents 102.4 µs only. Hence, to represent 150
µs, a 13-bit pulse counter is required.


        Fig 5 shows the PW measurement scheme. The detected and amplified video is
given to a threshold comparator to give a TTL signal when the input signal crosses the
threshold level. The threshold is set adaptively at 3dB down with respect to peak




                  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                            II -27
amplitude of the video signal. The TTL pulse is given to PW counter along with clock.
The counter output i.e., 13 bits digital data gives the PW of the intercepted signal.



Omni Antenna
                                        Video
                                        Amplifier    Video signal
          Rf Amplifier
                            Detector
                                                                                        Pulse Amplitude
                                                                       A/D
                                                                     converter
                                                                                        8 bits




                     Comparator
                                                                              Pulse width
                                                   PW                       Video Processor
                                                 Counters
                                                 (13 bit)          8 bits
        Threshold



                                                  TOA          Time of Arrival
    Clock 40 MHz                                 Counters
                                                 (32 bit)
                                                                    32 bits


                         Fig 5: Pulse Parameter Measurement Unit



7.5.3 Time Of Arrival (TOA) measurement


       The TOA measurement methodology is similar to that of PW measurement. The
clock resolution and bit length are set by the maximum and minimum PRFs.


       The maximum PRF to be handled is 500 KHz, i.e., the minimum time to be
represented between two successive pulses is 2 µs. This should be measured with a
resolution of 25 ns. The same clock of 40 MHz used for PW serves the purpose and meets
the TOA measurement resolution and accuracy.


       The minimum PRF to be processed is 100 Hz, i.e., the time to be measured
between the successive pulses is 10msec. A 29-bit counter caters the TOA measurements



                    CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                              II -28
up to 13.11 m sec and this is sufficient for our application. However, a 32 bit counter is
recommended to cater for future expansion. The TOA is used for PRF analysis in the
subsequent processor stage.


7.6 Pulse Formatter

         So far, the measurement of signal parameters is considered. All these parameters
i.e., Frequency, DOA, PA, PW and TOA are synchronized and formatted as Pulse
Descriptor Word as shown fig. 6.


                                         Frequency

                                          DOA
                                                                                      PDW
                                          PA                      Pulse
                                                                  Formatter
                                          PW

                                          TOA
          I/P signal

                                          End-of-process (EOP) signal
              EOP

                                                  Fig 6: Pulse Formatter



         The pulse data is synchronized with End Of Process (EOP) signal. The EOP is
generated after completion of the measurement of all the parameters. This should be less
than (te + 2) µs where te is the trailing edge of the input signal. This enables the receiver
to start processing the next signal coming immediately after (te + 2) µs of the proceeding
pulse.

         The use of VLSICs like ASICs/FPGAs greatly reduces the hardware complexity.
In fact, the whole Pulse Formatter can be realized in a single FPGA.




                       CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                                 II -29
7.7 Processor
       The processor receives the interleaved pulse descriptor words coming from pulse
formatter and performs the following functions:

                ♦ De-interleaves the PDW
                ♦ Determines PRF and performs PRF analysis in case of staggered or
                   Jittered signals
                ♦ Determines scan type and performs scan analysis
                ♦ Assembles parameters for each emitter and prepare an emitter file and
                   monitor their activity
                ♦ Performs identity search
                ♦ Transfers the emitter data to the system display


       The functional diagram is shown in fig. 7



                                        PDW




                                      Interface &
                                       Validation
                                         Logic




                                      Track Data
      De-interleaving                  Memory                       Emitter
       and Control                                                 Processor


                                                                           FIFO Port
                                                    Parallel Bus

                                                                     IO Processor      Display
                              Target Library



                           Fig 7: Block Diagram of ESM Processor




                  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                            II -30
       The processor should perform the above functions under dense signal
environment. The specification laid down is 5,00,000 pulses per second which is derived
from the max PRF of 500 kHz to be handled by the processor.


7.7.1 De-interleaving


       The de-interleaving process can be implemented by specially designed ASICs.
The ASICs does this function by comparing pulse parameters i.e., Frequency, DOA, PW
with corresponding parameters of already detected emitters in parallel. The process of
comparing parameters is completed with in 2 µs, which meets the required goal of
5,00,000 pulses per second.


7.7.2 PRF Analysis


       PRF is derived form TOA measurements. The difference in TOA of the
successive pulses (∆TOA) gives the Pulse Repetition Interval (PRI). The inverse of PRI
gives the PRF.


       PRF analysis is done by analyzing histograms of differences in time of arrivals of
successive pulses. PRF analysis is required to handle fixed, Jitter and Stagger. An emitter
having PRI variations with in 0.75% is classified as fixed or constant PRF.Emitters with
variations of PRI from 0.75% to 10% of mean are categorized as Jitter type and PRI
variations from 10% to 25% of mean are categorized as Stagger type.


7.7.3 Scan Analysis


       Scan Analysis is performed by analyzing amplitude envelope as function of time
and determining the Antenna Rotation Rate (ARR) or Antenna Scan Period (ASP).
Depending on the ARR/ASP, the scan types are declared as given below:




                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                           II -31
               ASP                             Scan Type


       0.04 sec to 0.1sec                      Track while Scan
       0.1sec to 1sec                          Sectoral Scan
       1sec to 20sec                           Circular Scan


If there is no amplitude variation in the received pulses, it is declared as Lock-on type.


7.7.4 Identity Search


       The processor performs identity search of the intercepted emitters against the
known/warned radars. The parameters of the warned/known radars are stored in the radar
library. The library capacity depends on the requirement. A radar library to store up to
300 would be reasonable.


       Identity search is carried out against the entire library for every new emitter and
when the parameter matches, the identity of the particular radar is displayed.


7.7.5 Micro-Processor Selection


       Except the de-interleaving, which is performed by hardware (ASICs), all other
processor functions have to be implemented in software. Of these functions, PRF analysis
is the most intensive computation. This is performed by histogramming difference in
TOA, a 32 bit data. Also, the incoming PDWs should be stored requiring enough
memory. This requires memory addressing of approximately 1 mega byte.


       Considering all the above factors, Intel 80386 microprocessor is selected for
building processor hardware. The required reaction time of less than 1 sec and can be
comfortably met with this.




                  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                            II -32
7.7.6 System Display Interface


          To transfer the emitter files from processor to the system display, a dedicated I/O
processor is required. The I/O processor maintains the emitter file and sends the data to
the display unit over high-speed serial link. The I/O processor also receives system
controls / commands from the system display or control panel.


          A dedicated 80386 CPU based processor can provide the I/O processing
requirements.

7.8 System Display


          The main role of the system display is to acquire parameters of the intercepted
radars signals from the processor and to present them on the screen of the display for
visualizing the radar signal scenario in easy to understand formats. It also responds to the
commands issued by the operator through the keyboard.


          For aircraft systems, the displays should be rugged, compact and lightweight. The
display panel should be flat and should provide the requisite viewing angle of at least
± 60 °.


          The size of the display mainly depends on the availability of space on the
platform. Different sizes are available ranging from 6" to 20”. Hence, display of
appropriate size should be selected such that it fits in the available space.


8. ADVANCED ESM SYSTEMS


          The radar’s performance has been greatly augmented through the application of a
variety of techniques. In particular, to improve resolution, range and anti-jamming
capability, coded or spread spectrum signals such as Barker codes, pseudo-random codes,
Linear FM or Chirp modulations are widely used. Radars employing these techniques are
called Low probability of intercept (LPI) radars. LPI signals are challenging to the EW
receiving systems attempting to detect these radars.



                   CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                             II -33
       One of the LPI techniques is the use of narrow beam antennas or antennas with
suppressed side lobes. Since these antennas emit less off axis power, the signal is more
difficult to detect. If signal duration is reduced then ESM receiver gets less time to search
the signal frequency, thus reducing the POI of the ESM receiver.


       The other technique of LPI radar is spread spectrum. This technique spreads the
signal over a wide band resulting in reduction of peak power, while maintaining required
average power level. Also, the transmitted power is optimized adaptively i.e., the power
transmitted is just sufficient to receive the reflected signal and this keeps on decreasing
for an approaching target. Reducing the peak power to minimum level makes the ESM
receivers very difficult to detect them due to inadequate S/N ratio available at the input of
the ESM receiver.


       Spreading LPI signals for long duration enables signal modulation with larger
time bandwidths. LPI modulation spreads the signal energy into frequency, so that the
frequency spectrum of the transmitted signal is many orders of magnitude wider than
required to carry the signals information. This in turn reduces the signal strength per
information bandwidth. Since noise in receiver is a function of bandwidth, the S/N ratio
in ESM receiver attempting to receive and process the signal in its full bandwidth will be
greatly reduced by the signal spreading.


       To overcome the above factors, the ESM System designer has to concentrate on
the following techniques to detect LPI radars.


   Matching the receiver to radar signal
   Increasing the receiver sensitivity
   Adopting modern detection and processing techniques
   Extracting the radar parameter details to high accuracy
   Improving the POI




                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                           II -34
        The above techniques cannot be offered by a single type of receiver. By
incorporating suitable combination of modern receivers such as Channelized, IFM,
Microscan and Digital receivers, the detection of LPI signals could be made possible.


9. CONCLUSION


        The role of ESM systems in modern Electronic Warfare has been brought out in
this paper. The factors that have to be considered in designing an ESM system are also
described.


        Based on the system requirements & technical specifications, the ESM system
designer has to carefully select the required configuration. To meet the present day threat
scenario, single receiver is not sufficient and hence the need for selection of a
combination of receivers is essential. The designer has to optimize the configuration by
selecting one or more receiver technologies considering the permissible size & weight on
the platform and cost factors.


10      REFERENCES:


1. Electronic Intelligence: The analysis of Radar Signals by Richard G. Wiley
2. Introduction to Electronic Warfare by D. Curtis Schleher
3. Microwave Receivers with Electronic Warfare Applications by James Bao –Yen Tsui
4. Electronic Warfare Receiving Systems by Dennis D. Vaccaro
5. Airborne EW system concepts by Mavrice W. Long
6. Naval Electronic Warfare (vol 5) by Dr D.G. Kiely
7. Electronic Warfare- Brassey’s Air Power: Aircraft Weapon Systems and Technology
     series
8. Electronic Warfare in the Information age by D. Curtis Schleher




                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                           II -35
            CHAPTER – III



           ELINT SYSTEMS

Smt. RENUKA CHITRAKAR, Sc-‘G’
                &
Shri. H.R. KRISHNA KUMAR, Sc-‘F’




  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                              III
                         CHAPTER - III


                        ELINT SYSTEMS


                           CONTENTS




1.         INTRODUCTION


2.         CLASSIFICATION OF EW RECEIVERS


3.         SALIENT FEATURES OF ELINT SYSTEM


4.          CONTROLLERS AND DISPLAYS


5.          CONCLUSION


6.           REFERENCES




     CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                 III
                                    ELINT SYSTEMS


      Smt Renuka Chitrakar, Sc. G , Sri H.R.Krishna Kumar Sc. F
                                                And
              Smt B. Kameshwari TO -C, DLRL, Hyderabad

1.0      INTRODUCTION:


         The extensive application of EM spectrum to communication Radar and Navigation
provided the armed forces with powerful tools. Radar Communication provided coordination
between forces and radar performs surveillance of the battle space to verify force deployments
and direct hostile forces. The EM spectrum is so extensively exploited, today without its use, the
survivability of armed forces is jeopardized.
         EW is the science of manipulation and control of EW environment for its own
survivability but denies or limit it to the adversary. EW technology extracted essential
information is then exploited to influence adversary’s capability to coordinate its activities, to
restrict its communication media, to deny the use of Radar for weapon Launching or guiding.
EW enhances the survivability of own forces by denying the use of EM spectrum by the enemy.


EW is broadly classified based on
   1) frequency spectrum;
   2) functionality;
   3) intended role;


EW is divided into three groups based on frequency
      1) Communication EW;
      2) Radar EW;
      3) EO/IR EW;
      4) Hybrid EO/IR-RFEW

               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         III - 1
RADAR EW:
         Armed forces use radar for both defensive and offensive weapon systems. Reflected RF
echoes of the target, are used to measure target range, bearing and elevation and determine target
location. Radar uses RF transmission ranging from high frequency(HF) to millimeter
waves(30MHz – 95GHz). Various alphabetical letters designate the frequencies. Both old and
new designations are given at Table1. Radar EW involves extraction of detailed information of
Radar signals emitted, use of this information either to formulate Electronic order of Battle
(EOB), or provide the information to a jammer to operate in an efficient way.


           L           S          C            X        Ku             K        Ka
           D           E     F    G    H       I            J                   K
 (GHz) 1                2     3    4       6   8   10       12     18      20             26
                            Table No 1: Radar Frequency Designations


2.0      CLASSIFICATION OF EW RECEIVERS
EW is classified based on functionality into three groups
      1) Electronic support (ES)
      2) Electronic Attack (EA)
      3) Electronic self-protection(EP).


2.1      Electronic Support (ES):
       ES also is known by ESM (Electronic Support Measure). ES involves search, intercept
locate, record and analyze radiated EM energy for the purpose of exploiting the radiation
information either for formulating EOB (Electronic Order of Battle) or to provide the real time
information to EA system. ES provides surveillance and warning information derived from
intercepted EM environment emissions.


2.2       Electronic Attack (EA):
         EA is also known by the name ECM (Electronic Counter Measure). EA involves action
taken to prevent or reduce enemy’s effective use of EM spectrum. It can be active systems like
Jammers, or can be passive like chaff.
                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                           III - 2
2.3    Electronic Self Protection(EP).
 EP is also known by old name ECCM (Electronic Counter Counter Measures). EP involves
actions taken to ensure friendly use of ECM. EP protects own platform against EA by the
adversary.


EW can be classified based on the role it has been assigned to carry out
1) Tactical
2) Strategic


Electronic Support (ES):
       The ES function is for real time use, whereas Elint Rxs are used for intelligence
collection, which can be subsequently used. Elint Rxs perform fine grain analysis of emitters of
interest. If the data cannot be analyzed it can be stored and analyzed at a later time.


The EW Rxs can be classified by their 1) Application and 2) Structure


Based on their application EW Rxs are classified as
               a)      RWR
               b)      ESM Rxs
               c)      ECM Rxs
               d)      ELINT Rx


Based on the structure EW Rxs are classified as
               a)      Crystal video receiver
               b)      Superhet Receiver
               c)      IFM receiver
               d)      Channalised receiver
               e)      Micro scan receiver
               f)      Bragg Cell Receiver


                CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                          III - 3
Comparison of various Rx at a glance is given in the table below:

                 Channel         Compressive Optical          IFM       Crystal        Superhet
Instantaneous    Good            Good        Good             Excellent Excellent      Poor
Frequency
Bandwidth
Capability    to Good            Good             Good        Poor        Poor         Poor
handle
simultaneous
signals
Frequency        Good            Good             Good        Good        Poor         Excellent
resolution
Sensitivity      Good            Good             Fair   to Fair to                    Excellent
                                                  Good      Good
Dynamic range       Good         Fair to Good     Fair      Fair to Fair               Excellent
                                                            Good

                                        TABLE 2


3.0     ELINT Rxs:
        An Elint Rx main function is collection of data and information useful for strategic
planning. It does fine-grain analysis and therefore has a very high sensitivity compared to ESM
Rx. Its instantaneous bandwidth is far less than the ESM Rx, hence it covers one or a few signals
of interest. If the data cannot be analyzed at the collection station, it can be stored and analyzed
at a later time. It normally works in peacetime to collect as much data about the adversary.
       ELINT is the abbreviation for “Electronic Intelligence”, that is gathered as a result of
observing the transmissions and there by obtain information about their capabilities. These
transmissions can be generally any non-communication emissions and specifically radar
emissions.


Salient features of ELINT systems:
       The main characteristic of an ELINT system is high sensitivity, high frequency accuracy,
high dynamic range and low probability of intercept. The importance is on measurement of all
the radar parameters very accurately.



               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         III - 4
Major signal parameters
  1) Power characteristics – Observed polarization beam patterns and the angular motion
        (scanning) of the beam.
  2) Modulation characteristics –
                 Pulsed - PW, PRI
                 FREQUENCY OR PHASE MODULATION & CW
  3) RF or carrier characteristics      Frequency band internal variations (freq. Agility)

3.1 ELINT RECEIVERS:

         A role of an Elint Receiver is SEARCH, INTERCEPT, ACQUIRE, ANALYSE and
extract the DIRECTION OF ARRIVAL of the signal. The receiver should be capable of
covering wide frequency range although instantaneous bandwidth of the receiver is only be wide
enough to cover one or few signals of intent. Also if data cannot be analyzed on line, it has to be
stored for analysis at to later time.
From the Table-2, the performance of a superhet receiver can be considered excellent in every
aspect except the low POI. Mainly superhet receiver and superhet in conjunction with IFM,
Channellsed, Micro scan and Bragg cell can be used for ELINT application. The principle of
operation of superhet and IFM receiver which are extensively used and the technology is well
developed is given below.


3.1.1 PRINCIPLE OF OPERATION FOR SUPERHET RECEIVER:

         The main components of a superhet receivers are mixer, local oscillator, IF filter, IF
amplifier and video detector. The LO generates continuous wave (CW) signal of frequency fL. If
the input signal fs, the two signal fS & fL going to be input to the mixer, which in turn generates
difference in frequency between fS and fL to give f1 the intermediate frequency. The f1 is
generally down converted frequency and some time up converted frequency. IF filter center
frequency is fixed and LO is tuned to get the frequency fL       - fS to be always in the filter pass
band.




                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                           III - 5
       Mixers are non-linear devices. Hence when two signals fS         &   fL   are present, many
intermod frequencies will be generated at the output. The IF filter plays the role of filtering out
the unwanted intermod frequency.
       IF amplifier following the IF filter provide maximum gain and increase the sensitivity of
receiver. This receiver will operate from gain limited to the noise limited condition. Amplifier
output is given to a crystal video detector. The microwave energy is converted to video signal.
The video amplifier following video detector will amplify signal for further processing. The
video generated will be LIN, LOG and FM. Each of these are used for signal characterization.




                              Block diagram of a superhet receiver

       The best feature of a superhet receiver is that it can be used to separate and isolate a
signal of interest and to perform a fine-grain analysis on the signal. It is the most commonly
used receiver in many microwave areas. Superhet receivers are most commonly used in
communication and radar applications.
       In EW applications, a superhet receiver is often used in conjunction with some other
types of wide-band receivers. The wide-band receiver can be used as a cueing receiver to
find the frequency of the signals of interest. The superhet receiver operating in the narrow
band and wide band are shown in Fig.1. and Fig.2. respectively. The superhet receiver can
then be tuned to the desired frequency to do a fine grain analysis on the signals. One such
application is to measure the AOA information of an input signal through an interferometric
system. Because of the relatively narrow bandwidth, phase matching between different
superhet receivers is easier to achieve than in a wide band receiver. In an amplitude
comparison AOA system, the superhet receivers are also commonly used.
       Recent developments in superhet receivers are in the areas of frequency synthesizers and
control circuitry. The synthesizer can improve the tuning accuracy and the tuning speed of a
               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         III - 6
superhet receiver. The control circuitry can improve the quality of the signals displayed, signal
processing, and receiver versatility. The advances in microwave components and logic circuits
will improve the receiver performance and reduce size.


3.1.2 IFM Receiver:
          An IFM Rx uses delay lines and measures the phase difference to measure
its frequency. R.F. is split into a delayed path with delay T and an undelayed path.
It θ is the phase angle between delayed and undelayed waves, θ = ωT. If phase
angle θ is measured and the delay time T is known, frequency can be computed. In
an IFM receiver the phase angle θ is measured using the relation below:
                             A       =      X Sin θ
                             B       =      X Cos θ
                                                  –1
                             θ       =      tan        A/B    =ω T
Where X is the amplitude information.          The signal is passed through a limiting
Amplifier before applying to the R.F. power divider. One of the outputs of power
divider is given to the phase correlator. Other input to the correlator is the delayed
signal by a specific time T. The correlator multiplies the undelayed signal and the
delayed signal and produces Sine and cosine Video outputs. They are digitized in a
8 bit quantizer and phase ‘θ’ is computed. The digitized data is fed to a ROM which
perform     ω =1/T tan-1 (Sin ωT/ Cos ωT). The frequency is thus directly computed.
Number of delay lines and associated correlaters are used to obtain good accuracy
and resolution.    The longest delay line determines the frequency accuracy and
resolution, whereas the shortest delay line determines the unambiguous bandwidth.


                                                             π/2
       Cost ωt

                                                              Cos θ
                                                                                Sin θ



                             BLOCK DIAGRAM OF I F M RECEIVER
               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         III - 7
The IFM will encounter many problems when there are number of emitters in space. The
problems are (i) Simultaneous signals i.e. pulse on pulse; (ii) Pulse on CW condition etc.
These problems have been solved to certain extent and lot is needed to be done with the
present high-density emitter environment.


3.2 Salient features and parameter measurements of ELINT systems :
An ELINT system has characteristically high sensitivity, high frequency accuracy and low
probability of intercept
Each received pulse potentially provides the following information concerning the emitter.
           ● Frequency
           ● Received power
           ● Pulse duration
           ● Polarization
           ● Time of arrival
           ● Angle of arrival
           ● Characteristics of intrapulse modulation


3.2.1 Noise Figure (NF) : Noise figure of a receiver is described as a measure of the noise
        produced by a practical receiver as compared with the noise of a ideal receiver.
        NF = Sin/Nin/So/No=Nout/KTBG
        This is commonly in decibels as 10 logNF.
        When no. of stages are cascaded the combined NF is given by
        F0 = F1 + F2-1/G1 + F3-1/G1G2+-------Fn-1/G1G2—Gn-1.


3.2.2   Sensitivity:
        Sensitivity refers to the minimum signal strength required. The ability of a system to
detect a weak signal is limited by the noise energy that occupies the same portion of the
frequency spectrum as does the signal, and the necessary signal power is defined in terms of the
effective noise power at the receiver input. The effective noise power is given by thermal noise
contribution multiplied by a degradation factor called the “noise figure” (NF)

               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         III - 8
               N (eff) = 114 dBm MHz+10 log 10 BW (eff) + NF (dB)


               BW (eff) = √ 2(BWrf ) (BW video)


The minimum discernable signal (MDS) is that signal level at the input for which So/ No at the
output of the receiver is 2 or 3 dB.


There are two dominant requirements establishing the required signal strength for any system
employing automatic detection:
1.      The threshold must be high enough so that false alarms are not generated at a rate, which
        overloads the signal processing equipment.
2.      The minimum required input signal must be detected with a certain specified probability.


An attempt to threshold at MDS would produce far too may false alarms, thus a typical detection
level is set 14dB above MDS. Most parameter measurements require even more signal strength.


        Several advantages are realized by a highly sensitive system. Signal can be detected at
greater range and, therefore, more signal is available for processing. Measurement accuracy is
improved as the signal-to-noise ratio improves. There is the possibility of detecting an emitter
through the back lobes or side lobes of rotation at antenna and therefore, improve the probability
of interception (POI). With the advantages come certain disadvantages. Greater signal density
imposes a greater burden on the signal processor. Extraneous multipath reflections of the
primary signal can also break threshold and clog the system with undesired spurious data. Often,
higher sensitivity is obtained by narrowing the system bandwidth, or narrowing the antenna
beam width. Both of these techniques reduce the probability of intercept and degrade the
capability to measure simultaneous or correlated signals not sharing identical frequency.


3.2.3   Probability-of- Intercept:
        Each receiving system is characterized by different limitations regarding the probability-
to-intercept a signal. Systems, which employ narrowband RF or narrow beam width, require

                CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                          III - 9
time to search for, and acquire the desired signal. There are two complementary ways to specify
the problem; Either determine the time required to obtain a given probability-of-intercept, or
calculate the probability-of-intercept, for a given scanning super heterodyne receiver. The
intercept time (probability) will be determined by the receiver bandwidth, emitter PRI, emitter
duration, and the number of pulses required. For receivers and transmitters associated with
rotational antennas, the antenna scan rates and beam widths must be included in calculation as
well as the receiver sensitivity. An intercept study must be used to determine the tradeoffs
between antenna and receiver parameters for a certain set of targeted emitters.


If a group of pulses is intercepted from a single emitter, the pulse repetition interval may be
determined as well as possible characteristics of stagger and jitter. Again many possible Inter /
intrapulse variations may be determined, such as frequency agility, PRI agility, scan
modulations.


4.0    Controllers and Displays for ELINT:
       The ELINT receiver is configured to meet the strategic or Electronic Intelligent gathering
requirements. The receiver functions in stand alone mode. The receiver is provided with very
high sensitivity for detection beyond horizon and side lobe/ back lobe detection of very weak
signals. The receiver searches in spectral and spatial domains by scanning so that only a fraction
of spectrum or space is in view for the receiver. The receiver can provide detailed signal
analysis.


       The receiver shown in Fig.3 indicates the various subsystems of a ELINT receiver. It
consists of a main controller unit along with displays for showing signal activities, Wide Band
Tuner, few Narrow Band Tuners to cover the frequency band, Analysis demodulator and Pulse
Analyzers, DF controller unit and associated Pedestal Electronic. Some times an IF PAN and
Time Domain Display (TDD) are added to the receiver system. The Wide band tuner has
around 300 to 600 MHz IF bandwidth and used for processing signals from pulse compression
and frequency agile radars. The other tuners are of super heterodyne type and are used for
scanning the frequency spectrum in the corresponding bands. The Analysis demodulator shown
in Fig. 6 provides video outputs from the IF output signal. The Pulse analyzer measures pulse

               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         III - 10
width, PRF, PRF jitter, stagger and details of intra pulse modulation. The receiver controller
provides tuning commands to the search and analysis receivers and presents data of intercepted
emitters in various modes to the operator for further action. The controller unit provides several
analysis features such as Memory lock out, IF centering, Smart analysis, Smart scan and it
collects information from search receiver, maintains monitor list for analysis receiver. It can be
programmed for smart scan, smart analysis, monitoring and frequency lock out.


The parameters measured by ELINT receivers are
                   a. Frequency
                   b. Pulse amplitude (PA)
                   c. Pulse Width (PW)
                   d. Time of arrival (TOA)
                   e. Angle of arrival (AOA)
                   f. Inter and intra Pulse Modulations
                   g. Scan parameter
        Except for the frequency measurements, all these quantities are measured with similar
schemes, even in different kinds of receivers. The frequency information is very important for
both sorting and jamming. By comparing the frequency of the pulses received, pulse trains of
various radars can be sorted out. Knowing the frequency of the victim radar, the jammer can
concentrate its energy in the desired frequency range.


4.1     Pulse Amplitude (PA):
        PA can be used to generate the scan pattern of some radar. Theoretically the PA
information may be used to estimate the distance of a radar. Accurate PA information may be
helpful in generating some effective jamming schemes, that is inverse gain jamming, where the
jammer power is inversely proportional to the PA measured.
4.1.1   Amplitude Measurements:
        A sample and hold followed by an analog to digital comparator is used for measurement
of amplitude. The amplitude of the log video gives the strength of the emitter signal. If 8 bit A/D
converter is used then we get dynamic range of 6x8 db = 48 dBm.


               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         III - 11
4.2    Pulse Width (PW):
       Pulse width can be used to provide course information on the type of radars, for example
weapon radars have short pulses.


4.3    Pulse Repetition Interval:
       Time of arrival (TOA) can be used to generate the PRF of the radar. It can be used to
predict the TOA of the next pulse for a jammer. The ELINT Analyst explores the PRI variation
and attempts to place the radar emitter into one of the categories listed in the Table titled
“TYPICAL PRIs IN CURRENT USAGE”.


4.3.1 ADVANCED PRI ANALYSIS:
       In addition to the currently employed conventional techniques in today’s complex emitter
scenarios, statistical means of PRI Analysis enables ELINT systems to automatically extract an
intercepted emitter’s PRI period and the type of pattern employed.
       This PRI Analysis method is composed of two distinct parts, the PRI period calculation
Algorithm and the PRI type Determination Algorithm. As can be deduced from the
identification, the former extracts an emitter’s PRI period and the latter detects the type of
pattern used by the source emitter.


DESCRIPTIONOF OF PRI PERIOD: : The algorithm consists of a sequence of processing
functions or steps.
         a.      Pulse train stabilization
         b.      PRI period candidate selection
         c.      Time domain sequence generation
         d.      Spectral analysis and PRI period extraction




                               Constant PRI


               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         III - 12
                                Four level staggered PRI




                              Three level Dwell –And- Switch PRI
                        Illustration of three different types of PRI sequences

A typical EW system pulse descriptor word consists of the following parameters:
a.                Frequency
b.                Angle of arrival
c.                Pulse width
d.                Amplitude
e.                Time of arrival


DESCRIPTION OF EXTRACTION OF PRI TYPE:
       The knowledge of the PRI sequence period can now be exploited to extract to individual
values making up to sequence and their repetition factors. The algorithm consists of two
processing functions or steps. They are the following
                  a) Noise free pulse sequence template generation
                  b) Processing of above template through a simple neural network.


       The first step assumes that typical pulse trains to be analyzed contain extraneous pulses
or are missing pulses from their sequence. This step attempts to correct these corruptions so that
the neural network can accurately determine the PRI pattern used by the emitter.



               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         III - 13
        In a typical scenario of multiple emitters a combination of these PRI types may be
available for the ELINT processor to de-interleave. De-interleaving of the multiple emitters is
important before characterization. De-interleaving algorithms are based on the analysis of
various parameters of the received radar pulses, such as time of arrival, angle of arrival, PRI etc;


        The histogram of the first successive difference of the TOA sequence has basic
information regarding the inherent Pulse Repetition Interval (PRI). The histogram either speeds
or concentrates around certain values of the PRI, depending on the categories of emitters present
in the data.


        Thus intuitively, one can arrive at the procedure to sort the categories based on the spread
of the histogram. In this context Spread to Mean Ratio (SMR) algorithm is applied. First it is
analyzed whether it is multi emitter or single emitter. If single emitter, PRI is analyzed as
constant, jitter, stagger, dwell and switch. For multiple emitters only constant PRI type is
analyzed.


4.4     Angle Of Arrival (AOA):
        ELINT receiver will usually measure AOA in azimuth plane. Among all parameters
AOA is the most difficult and expensive parameter to measure. It is the most valuable parameter
because hostile radar can change the first four parameters but it cannot easily change the AOA
information received by an EW receiver. Some times the polarization of the input signal is also
measured. This information can be used to select jamming schemes i.e. cross polarization
wherein the polarization of the jamming signal is 90º with respect to the input signal.
        Another parameter of interest in the ELINT receiver measurements is the range of the
emitter. This information can be used in many applications, for example for threat avoidance and
weapon delivery. Emitter position fixing is carried out by obtaining AOA for a particular emitter
by multiple ELINT receivers.


4.4.1   AOA is measured using the following techniques
        1.     Rotary DF system technique.

                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                           III - 14
          2.   Amplitude comparison technique.
          3.   Base line interferometer technique.
          4.   Digital bearing discriminator technique
          5.   Time difference of arrival technique.
          6.   AOA measured through microwave lens.
          7.   AOA measured through multiple beam arrays and beam-forming network.


      In approaches 2 and 3 most common in ELINT applications multiple number of antenna
and receivers are needed. Besides the antennas, receivers must be matched in amplitude in
approach 2 and in phase in approach 3. Thus the AOA information becomes the most costly
parameter to measure.
      The approaches 1 i.e. Rotary DF system technique is also the other most commonly used
technique for elint application. This system consists of DF controller and servo control unit. Here
the Amplitude of the signal with respect to the position of the antenna is displayed. The peak of
the feather pattern indicates the direction of the signal. The histogram mode of display I.E Angle
versus number of pulses received in a particular direction aids the operator in extracting the
information more accurately. The common display modes are
                 Polar
                 Inverse polar
                 Rectilinear
                 Rising Raster and
                 Histogram
      These graphical displays are used to aid the operator in estimating the DOA of the emitter
signal.


4.5       SCAN ALALYSIS:
          SCAN parameter forms one of the important parameter that needs to be extracted by an
elint receiver. An algorithm to extract the same is described below.
SCAN ANALYSIS ALGORITHM DESCRIPTION:
Two of the emitter’s parameters scan period & scan pattern are extracted using this algorithm.


                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                           III - 15
SCAN PERIOD: An emitter scans period in defined on the time measured for the intercepted
signal amplitude modulation pattern to repeat itself.


SCAN PATTERN: It is the trajectory employed by its transmitting antenna to cover its pre-
determined sector.


Definition Low illumination type: This is associated with search function. The ratio of
illumination time to scan period is less. Example: Circular, Bi-directional sector or Multi bar
raster.


Definition high illumination type: The ratio of illumination time to scan period is high.
Examples are conical and fixed.


Various steps involved in this algorithm is described below.


Step1: Amplitude Data sampling process:
The Scan period calculation is to generate a sampled amplitude scan envelope from the original
raw data. There were two methods considered: to use a maximum or the average of the
intercepted values within each sampling interval. Although both appeared to work well,
preference was given to the average value method since it provides a better smoothing effect on
the original amplitude data. The resulting sampled sequence offers the advantage of
Histogramming the raw data, making it more suitable for subsequent spectral analysis. Once
sampled, the mean of the sequence is removed for a reduction in computational needs during the
spectral analysis step of the algorithm.


Step2:Performing a time-domain spectral analysis
Step 2: Using the sampled amplitude sequence values generated during the first step of the
algorithm, a variant of its Auto Correlation Function(ACF) is computed. The ACF represents a
fundamental mathematical tool in the time series analysis. Given that ACF values will later be
used in a comparative analysis, their common factors are stripped out to facilitate computation


               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         III - 16
and real time performance of the algorithm. Their computation is further simplified by using the
Zero man sequence.


  Step3: Searching for a peak.:
       The search for the peak corresponding to the scan period makes use of the statistical
nature of the scan pattern amplitude data. Since emitter scan envelope tend to exhibit a highly
autoregressive behavior, the ACF decreases monotonically for a number of low lags. Using the
calculated ACF sequence, a ‘slide down’ at lag zero process is under taken where by consecutive
values are compared. This process is repeated as long as the slope of the ACF at each point is
negative (that is when ACF values are less than their predecessor). The first occurrence of a
positive ACF slope constitutes the starting point for the scan period peak search. From this
starting point, the maximum ACF value at lags beyond it is extracted. Its lag corresponds to the
emitter’s scan period. The scan period calculation rapidly performed by multiplying this lag by
the sampling interval size.


Step4: Extracting the scan period
       Prior to undertaking the scan pattern detection processing, a preliminary test aimed at
categorizing scan pattern into low & high level illumination types is performed. This test used a
statistical parameter referred to as kurtosis. This value measures the peakedness or flatness the
sampled sequence values. The higher the value, the “peaker” the sequence. Once computed, the
Kurtosis value is compared to a threshold value of unity. This threshold delineates the low
illumination scan pattern from the high illumination ones.

Step5:Extracting the scan pattern
       The peak detection process is based on the expectation that for typical low illumination
scan pattern, the main beam illuminations are clearly the above the noise level. Another expected
situation is that peaks are likely to exhibit variability in their maximum amplitude. There may
also be some missing or additional unwanted peaks in the sampled scan data. The peak detection
threshold is therefore an adoptive nature, where by it is iteratively lowered until a pre defined
minimum peaks is detected. This minimum number is based on the type of scan pattern to be
detected. A formula for this number has been devised based on the time spanned by the sampled

               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         III - 17
amplitude sequence, the scan pattern to be detected and its associated number of expected peaks
during one scan period.


Step6: Low illumination scan pattern detection
        The detection of low illumination scan patterns consists of two steps: the first detects
peaks using a predetermined threshold and the second analyzes the time spacing between pairs of
peaks. The peak detection mechanism is based on in adoptive threshold definition. The threshold
is defined according to the expected number of peaks for each possible scan pattern. This
expected number of peaks is, in turn, calculated using the total time spanned by the sampled
amplitude sequence, the previously calculated scan period and the expected number of peaks in
one scan period. (i.e., one for a circular scan two for a bi-directional sector scan etc.). The
threshold is progressively lowered from the largest sampled amplitude value of the sequence to a
value required to detect the required number of peaks. An allowance is made for side lobes and
random amplitude variations in the definition of peaks. Values exceeding the detection threshold
that are near a previously detected peak are considered to be part of the same peak. Once the
expected minimum number of peaks has been detected, an analysis of time differences between
certain pairs of peaks based on known scan period and the scan pattern under analysis is
performed. For circular scan, time difference between consecutive peaks is analyzed, for bi-
directional sector scans, the time difference between every second detected peak, for a three bar
raster scan, the time difference between every third peak is analyzed and so forth.
        The peak time difference analysis consists of a two tier test approach: one is a simple
proximity filter test and the other is a statistical test. The first can quickly determine whether or
not the scan pattern is a viable candidate and second allows the detection of the scan pattern with
missing or additional peaks in the sequence. There is order of precedence between these two
tests. The proximity filter is always given higher priority over the statistical test for scan pattern
identification.


Step7: High illumination scan pattern detection:
  The high illumination scan pattern detection algorithm is based on a template comparison
technique. The clearly identifiable peaks typical present in low illumination scan patterns are not
evident in this case. However, the shape of their ACF is mathematically definable and can be

                  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                            III - 18
exploited to determine the level of similarity between the ACF of the sampled amplitude
sequence and the template. This is performed by computing the cross correlation factor between
the two sequences and, from its value, determining whether the two sequences are sufficiently
similar to make positive identification.
To conclude, this algorithms provides ELINT systems with the capability to automatically
extract an intercepted emitter’s antenna scanning characteristics very accurate.


4.6    Auto IF Centering:
       Using this facility, measurement of the emitter frequency to an accuracy of 2 MHz is
possible. In search mode this is automatically carried out making use of the FM video, the output
of the frequency discriminator. The FM video is an analog signal generated depending on how
much the generated IF is away with respect to the actual IF, for example if the generated video is
140 MHz, –2V FM video is generated. A lookup table is generated using FM video
characteristics. This FM video table is used in IF centering process. That is, the FM video is
sensed and the required correction is applied to the tuning command. The pulse parameters are
extracted after auto IF centering of the signal in the search mode.


4.7    Wide band Tuner:
       This Processor based controller generates tuning commands to the wide band tuner.
The tuning commands are generated for a defined step size and dwell time. The wide band
tuner (WBT) generates the intermediate frequency (IF), when tuning frequency is equal to the
emitter signal frequency. The IF is processed by the Digital Instantaneous Frequency
Measurement and processing circuits to generate digital words for frequency, PW, PRI etc.
These digital words are received by a high speed data acquisition card. This data is processed by
the controller to extract frequency, pulse width, PRI and signal type information.           This
information is sent to the main controller. Main controller maintains the extracted parameters in
a file. The controller has search, analysis and attendance modes of operation. In all the modes
the extracted data is sent to main controller.


4.8    Main Controller and Display :
The controller performs the following functions

                CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                          III - 19
           1) Generates RF tuning commands to tuner
           2) Programs threshold settings.
           3) Programs step size and scan speed for search operation.
           4) Programs scan band, skip bands and lockout facility.
           5) Carries auto IF centering.
           6) Puts the receiver in search mode, Analysis (manual) mode and attendance mode
          of operation.
           7) Computes the      pulse parameters.
           8) Computes the Intra pulse parameters.
           9) Transfers all the data to Main Controller to be filed and displayed.


       This is a PC based controller shown in Fig.3 which is interfaced to all narrow band
controllers, wide band controllers, direction finding controller, pulse analyzer unit, IF PAN, time
domain display and BITE source. The interface is through RS 422 in case of all controllers and
GPIB in case of IF PAN, TDD and BITE source. The main controller has a display unit to
display the signal activity. The main controller is the heart of the ELINT system. All the
controls and equipment operations are controlled as per the instructions of the main controller
i.e., modes of operation, antenna selection, scan parameters, IF B.W selection, RF & IF
attenuation selection, lock out frequencies, attendance frequency etc. The signal data generated
by the controllers are sent to the main controller. The data consists of frequency, type of signal,
signal amplitude, and in case of pulsed emissions, additional parameters like PW, PRI, etc are
also sent. The main controller receives the signal data and does the following.
       a) Generation of commands to processors of Narrow band and wide band tuners for
           interception of signals in various modes of receiver operation.
       b) On line display of data collected from receiver operating in auto mode on the RF
           activity screen.
       c) Store the data collected from receivers operating in various modes


       d) Perform analysis and direction finding operations, on signals specified by the
           operator.
       e) Store the fine grain information of the signal.

               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         III - 20
        The main controller also provides operator interface for the following
        1) Configuration specifications for different receivers
        2) Control of modes of operations
        3) Analysis and DF operations
        4) Diagnostic operations
        5) System administration
        6) Viewing the data stored in files
The main controller has OFF line mode of operation. In this mode all the acquired system files
may be viewed in various formats. Printing of any chosen file, system administration, Finger
printing etc are carried out.


4.9     Mode of Operations :
        The receiver can function in various modes of operation such as:
        1.     Search mode
        2.     Fine grain analysis mode
        3.     Attendance mode


4.9.1   Search Mode :
        In this mode both WB tuner and NB tuner go through spectral search within the defined
sector limits at programmed dwell periods. The fig 5 and Fig.7 shows the narrowband and wide
band receiver processors. When the signal strength exceeds the preset threshold, signal is
defined as present. For the specific signal which has exceeded the threshold, auto IF centering
is done, the coarse pulse parameters and amplitude data are sent to the main controller. The
tuners will resume the scan automatically. The main controller maintains this data in the
form of receiver file which can be displayed as and when required. The receiver file
contains parameters such as frequency, PW, PRI min, PRI max, amplitude, signal type, time
first seen and time last seen. The receiver file is maintained             in increasing order of
frequencies.




               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         III - 21
4.9.2   Fine grain analysis mode :
        In this mode one of the tuners is switched to analysis mode either through RF activity
display or signal listing obtained at the main controller during       search mode. The analysis
demodulator chain shown in Fig.6 is exercised in this mode. When the tuner is fixed tuned to the
signal, from IF pan presentation, tuning can be controlled through fine tuning in steps of 100
KHz.    Required IF attenuation and BW selection can be introduced and pulse analysis is
performed. If DF data is to be acquired, required antenna selection is made and using the DF
controller display unit, DOA is estimated and the data is sent to the main controller. In fine grain
analysis mode estimation of parameters can be done any number of times for confidence
building. The parameters extracted during analysis mode is passed on to the main controller,
which    is    maintained in form of emitter file and         displayed   as and when required.
Normally      in this mode    pulse   parameter   extraction is carried out    after acquiring   256
pulses . Extraction of antenna rotation rate is also carried out. If the signal is of stagger type
it extracts the level of stagger and lists out all the PRIs in the signal.


4.9.3   Attendance Mode:
        In    this mode, the tuner is tuned to the listed    frequencies. The tuner tunes to one
frequency at a time, looks for signal‘s presence            with logical time out     consideration,
reports presence or absence of the signal to the main controller              and proceeds to tune
to the next signal frequency in the list. The main controller maintains an attendance file
which can be displayed as and when required.


5.0     Conclusion :
        Various kinds of RF front end approaches for narrow band and wide band systems
are in use in present day ELINT systems. Novel advanced algorithms using statistical
methods are being used for analysis of pulse parameters, antenna scan rates and extraction of
parameters of exotic signals. The stress is more towards high accuracy and high-resolution
systems. Digital Rxs is the Rx for future. It digitizes the RF signal to be processed in a
computer. Software can be written to simulate any type of filter or demodulator giving lot of
flexibility and optimization. The problem of Digital Rx is in the availability of fast Analog
to Digital(A/D) Converters.

                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                           III - 22
6.0 References :


             i. Electronic intelligence, The Analysis of Radar Signals, Richard G.
                   Wiley, Aertech HouseInc MA, USA 1982


             ii. Introduction to Electronic warfare, SCHLEHER, DC, Dedham,
                   Aertech House Inc, 1986.


            iii. Principles of EW, SCHLESINGER, RJ, Peninsula Publication,
                   California, 1961.


            iv. MW receivers & related components, TSUI, JB, NTIS, Spring field
                   1983


             v. Automated PRI Analysis Software – Description Document by
                   DELATA consultancy


            vi. Automated Scan Analysis Software – Description Document by
                   DELATA consultancy




      CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                III - 23
                                     TYPICAL PRIs IN CURRENT


TYPE                  TYPICAL FUNCTION                                             REMARKS




Constan         Common Search or Track radars               Variations typically less than 1% of the average PRI value
                                                            Very stable constant PRIs are associated with MTI
                                                            Doppler systems



Jittered        Reduces the effects of some types of        Large variations-up to about 30% of the average PRI
                jamming



Dwell & switch Resolve velocity (or range) ambiguities      Bursts of pulses with several stable PRIs switched
               especially in pulse Doppler radars           from


Stagger         Eliminates blind speeds in MTI systems      Several stable PRIs switched on a pulse to pulse
                                                            basis




                          CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                                    III - 24
                                                                                         DETECTOR
        RF                               MIXE                          BWIF               AT IF
                          BWRF
                                                                         IF                                                 VIDEO
                          YIG                                                                                               OUTPUT
                                                                        BPF
                                                        IF          160 MHz
                                                     AMPLIFIER                                                  VIDEO
                                                                                                               AMPLIFIER
                                          LOCAL
  Sensitivity                           OSCILLATOR
                                                                 BWRF (TYP) – 15 to 50
2-8GHz (typical values)
                                                                 BWIF (TYP) – 1 to 40 MHz
      Auto       FGA                      TUNING
NB -69dBm -82dBm                        COMMANDS
WB -63dBm -65dBm
                                                         Fig . 1. FRONT END OF NARROW BAND SUPERHETERODYNE RECEIVER

                             BWRF
                            WIDE BAND
                          FIXED


                             BPF                                                                    DETECTOR
                                                          MIXE                    BWIF              AT IF                   VIDEO
                                                                                                                            OUTPUT
                             BPF                     S                               IF
                S
                                                                                   BPF
   RF                                                                    IF      TYPICAL                         VIDEO
                                                                      AMPLIFI   500 to 1000                     AMPLIFIER
                             BPF                           LOCAL
                                                         OSCILLATOR



                                                           TUNING
                                                         COMMANDS

                                    Fig . 2. FRONT END OF WIDE BAND SUPERHETERODYNE RECEIVER


                                    CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                                              III - 25
                                                           RS 422
                        GPIB                                                DF
       BITE
                                                                            Controller/




                          RS 422       Main              GPIB
    Narrow Band                        Controller
    Tuner/                                                                       IF Pan
                                       & Display
    Controller
                          RS 422                          GPIB



    Wide Band                                                                    TDD
    Tuner                                       RS 422




                                     Analysis
                                     Demodulator &
                                     Pulse Analyzer


FIG: 3 INTERACTION OF CONTROLLERS AND PROCESSORS WITH OTHER SUBSYSTEMS


                  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                            III - 26
Log Video                                                                                Tuner Command
             SH & A/D

                                                                                     Demodulator Command
FM video
             SH & A/D
                                                         Processor
                                                         Based
                                                         Controller
                                                                                        Main Controller
            A/D Intra pulse                                                                   &
                                                                                          Display

                                                                                         RF Attn.
                                                                                         Command to RDU


                TTL Video      Pulse Width        PW
                              Measurement
                                 Circuit


                                 PRI             PRI
                              Measurement                                  Memory Lock
                                 Circuit                                      List

                      FIG: 5 NARROW BAND RECEIVER PROCESSOR

                      CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                                III - 27
                                                                     Amplifier                Filter

                                    Normal /
IF From
various                             BITE
tuners
             SP4           Switch                                                Log Video    Log
                                               Analysis
                                               Demod.
                                                                                             Amplifier
                                    I/P        Controller /
                                               Processor             Proc.
                             BITE                                    Circu
                                                                              FM Video
                                                                                             FM Disc.


                Band Code                                                             TTL
                                                                                      O/P
                                                                                             Video
                                                                                             Proc.


                             Commands                                RS 422
                                                                                             Main
              Prog.Att                                                                       Controller
                                                              Pulse Parameters               & DIsplay



          FIG: 6 BLOCK DIAGRAM OF ANALYSIS DEMODULATOR AND PULSE ANALYZER


                         CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                                   III - 28
              CHAPTER – IV


DIGITAL INSTANTANEOUS FREQUENCY
       MEASUREMENT (DIFM)
                 &
         DIGITAL RECEIVER

      Shri. R.B.R. PRASAD, Sc-‘F’
                    &
       Shri. A.K. SINGH, Sc-‘D’




       CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                   IV
                           CHAPTER - IV



DIGITAL INSTANTANEOUS FREQUENCY MEASUREMENT (DIFM)

                               &
                       DIGITAL RECEIVER



                             CONTENTS

1.   INTRODUCTION

2.   DIFM RECEIVER

3.   DIGITAL RECEIVER

4.   CONCLUSION

5.   REFERENCES




       CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                               IV - 1
DIGITAL INSTANTANEOUS FREQUENCY MEASUREMENT (DIFM)
                          &
                  DIGITAL RECEIVER

             Shri R B R Prasad, Sc’F’ & Shri A K Singh, Sc ‘D’
                           DLRL, Hyderabad

1.0    Introduction:

       Electronic support (ES) system in electronic warfare (EW) involves actions
tasked by or under direct control of an operational commander to search for, intercept,
identify and locate source of intentional and unintentional radiated electromagnetic
energy. Each instantaneous non-communication signal intercepted by the Electronic
support system must be characterized by a set of parameters. These parameters
provide information required to associate a set of signals belonging to a particular
emitter and to identify that emitter among all other emitters whose signals have been
intercepted. The parameters generally measured by the Electronic support system for
a signal are RF frequency, pulse amplitude, pulse width, time of arrival and angle of
arrival. Also, in some systems, polarization of the input signal is measured. Frequency
receiver measures the RF frequency precisely and provides the information on the
emitted signal i.e. pulse, CW, chirp.


       Present day battlefield electro magnetic environment is generally very
complex with many emitters in the form of pulse trains effectively interleaved, so that
before any use can be made of the receiver output, the pulse trains must be de-
interleaved. This process involves, identifying each pulse with its particular emitter
parameters to isolating each pulse train.


       Frequency is one of the important parameters to be intercepted for the hostile
radars. It is necessary that the Frequency Measurement Receiver must provide very
good accuracy and resolution for Pulse train identification. Also Frequency Accuracy
is essential when the receiver is used for measurements on spread spectrum signals.
Some of the important characteristics of Frequency Receivers are:


               1.      Sensitivity
               2.      Dynamic Range
               3.      Through Put Time

            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                    IV - 2
               4.        Accuracy
               5.        Probability of Intercept


1.1 Types Of Frequency Receivers:


       Various types of frequency receivers used for carrier frequency measurements
in EW system are,


       1.      Super heterodyne Receiver
       2.      Digital Instantaneous Frequency Measurement (DIFM) Receiver
       3.      Channelized Receiver
       4.      Compressive Receiver
       5.      Bragg Cell Receiver (Optical Processor)
       6.      Digital Receiver


       Electronic warfare (EW) systems use frequency receiver as the essential sub
system for the following requirements.


       a.      Carrier frequency parameter is used in de-interleaving Radar pulse
               trains.
       b.      Frequency Receiver enables detection of intra pulse signals
       c.      It is used to detect the Radar types.
       d.      Provide tuning information for ECM system


2.0 Digital Instantaneous Frequency Measurement (DIFM) Receiver:


       Of the various types of receivers listed above for frequency measurement,
Digital instantaneous frequency measurement (DIFM) receiver is the mostly used
frequency receiver in EW system because of its inherent characteristics and makes
them suitable for both ESM and ELINT applications.


2.1 Characteristics:


       1.      Wide Instantaneous RF Band Width

            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                    IV - 3
       2.      Wide Instantaneous Dynamic Range
       3.      Good Frequency Accuracy
       4.      Make measurement on Short Pulse with high Frequency Accuracy
       5.      Adequate Sensitivity for practical applications


       The only draw back of an IFM        Receiver is that when multiple signals
arrive simultaneously, only one input (strongest) only will be measured.


2.2 Principle of operation:


       An IFM receiver use delay line to compare the phase of the input signal to
measure its frequency. The principle is best explained using the figure no. 1


       A sinusoidal wave Acos (ωt) is split into two paths, one path delayed by
constant value τ with respect to the other. The phase difference between the delayed
and un-delayed waves is given by


                               θ = 2πf τ

                               f = (1/2π)/ (θ/τ)
                       Where θ = Phase shift in radians
                               f = Frequency in Hz
                               τ = Line delay in seconds


       That is by measuring the phase deference in a phase detector we get the
frequency components as


               E = A Sin θ
               F = A Cos θ


       Where E and F are phase-detected voltages proportional to the input phase
difference. From which phase angle is determined using
                               θ = tan -1 (E/F ) = ω τ


            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                    IV - 4
             Amplitude information in E and F equations over the dynamic range are made
      constant through a Limiting Amplifier used in front of the power divider. The phase
      discriminator outputs E and F are amplified and given to Analog to Digital Converter
      to get the frequency information in digital form.


             The frequency accuracy depends upon the number of digitizer bits for the
      phase measurement. It is not possible to get more than 6 bits from digitizer due to
      phase noise present in the angle information. Six bit data limits the resolution of
      DIFM. To overcome this limitation more number of channels is used to get the high
      resolution in parallel. The longest delay line will determine the frequency resolution,
      when parallel channels are used, where phase measurement > 2π are involved
      ambiguity in measurement arise. These ambiguities are resolved using shortest delay
      channels. The shortest delay line is made to measure measurements for < 2π angle
      coverage. By choosing the appropriate delay ratio and after resolving the ambiguity,
      desired frequency resolution would be obtained.


2.3   Configuration of DIFM Receiver:


             The DIFM design is based on the latest development taking place in the filed
      of microwaves and digital electronics to take the best advantage of both the fields.
      The principle of operation is same as in conventional DIFM receiver, but the DIFM is
      discussed here incorporates homodyne front end with delay lines, phase comparator
      using analog IC, phase quantizer using high speed ADC and the frequency resolving
      and processing circuits are implemented in FPGA. The homodyne based DIFM have
      advantages over the conventional direct DIFM, like phase information transferred to
      single IF signal, low frequency IF amplifiers, filters and phase detector can be used
      for frequency measurement.


      The DIFM consists of
             1.      Homodyne front end with delay line assembly
             2.      Phase detector
             3.      Phase Quantizer
             4.      Ambiguity resolver


                  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                          IV - 5
The configuration of DIFM based on homodyne is shown in fig no.2. To get the
desired frequency accuracy DIFM is designed using 5 parallel channels with 5-delay
line.


1.      Homodyne front end:


        The advantage of homodyne based front end design for IFM is that the output
is always at a single frequency i.e. at IF (160MHz) providing wide instantaneous
coverage i.e. 2–18 GHz. Phase errors in homodyne configuration are less as compared
to conventional approach.


        Fig 3 gives the principle of homodyne receiver concept. Input signal will be
first power divided (p1) into two channels. One of the outputs will be mixed with the
fixed LO signal of 160 MHz in M1 so that the signal in this channel will be “RF +
160MHz”. The other part of the signal will be power divided (p2). Finally the two
signals will be mixed in M2 giving the output IF of 160MHz as shown in fig 3.
Frequency is measured by splitting the signal components between a reference path
and a known delay path and measuring the phase difference using phase
discriminator. The RF signal from the delayed line and undelayed “RF +LO” will be
again mixed in mixer M3 giving the output as delayed IF at 160MHz. Now, Two IF’s
at 160MHz, one is delayed IF and other is undelayed IF, having phase difference
same as would have been for two RF. The phase difference measured by phase
discriminator is a linear function of frequency and is affected by the electrical length
of the delay line, a constant. The relationship between phase shift and frequency is
described by the expression:


               θ = 2πfτ
               Where θ = Phase shift in radians
                       f = Frequency in Hz
                       τ = Line delay in seconds.




            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                    IV - 6
            Power Divider splits the input signal in to two paths. The Delay Line produces
     the necessary time delay required in the equation


2.   Phase Detector:


            The phase detection between delayed and undelayed signals is done through
     commercially phase detector IC The phase detector IC detects the phase from 0 to 180
     degree. Two IC’s will be used to detect the unambiguous phase of 360 degree . For
     quantizing the phase detector voltage ADC is used whose output is 8 bit digital data.
     Thus the phase detector, digitizer and encoder units, which are there in the
     conventional design, are replaced with commercially-of-the shelf IC’s.


3.   Phase Quantizer


            The phase detector output is being digitized using the phase quantizer. The
     high speed ADC is used for quantizing the phase information. The ADC digitizes two
     independent, high performances, 8-bit analog-to-digital converters (ADCs) on a single
     monolithic IC. Combined with an optional on-board voltage reference, the ADC
     provides a cost-effective alternative for systems requiring two or more ADCs.
     Dynamic performance (SNR, ENOB) is optimized to provide up to 50 MSPS
     conversion rates. Performance can be optimized for an analog input of 2 V p-p (±1 V;
     0 V to 2 V). Using the on-board 2 V voltage reference, the ADC can be set up for
     unipolar positive operation (0 V to 2 V). This internal voltage reference can drive
     both ADCs.


4.   Ambiguity Resolver:


            The data from L1 channel is unambiguous and the phase data from remaing
     higher order delay line having ambiguity. The purpose of the ambiguity resolver
     circuit is to resolve the ambiguity that is there in the data from L2, L3, L4 and L5 delay
     line phase discriminator. The ambiguity resolving and frequency processing circuits
     are being implemented in FPGA, which provides flexibility of reprogramming for any
     modification in the design without changing physical hardware. The FPGA also has


                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         IV - 7
provision for correcting the final frequency data for systematic and non-systematic
error in the fastest possible way. Ambiguity resolving circuit is having four stage of
processing The reference data is taken from the higher delay line i.e. L5 data and
algorithm proceeds with data from the previous delay line till all the ambiguities are
resolved.


3. DIGITAL RECEIVER

3.1 Introduction

       In the past the EW receivers were built by using analog means. Analog
conventional receiver requires calibrated components, additional filters and signal
amplifiers before the signal can be given to signal processing unit. Technological
advances in analog to digital converters (ADC), Field programmable gate arrays
(FPGA’s) and digital signal processors (DSP’s) provide the possibility of developing
the digital microwave receivers. In a digital receiver the input RF signal is down
converted into intermediate frequency (IF) and digitized directly without passing
through the video detector.      Performance of digital receiver will be superior to
conventional analog receiver in two aspects


       1. The digitized data can be stored for a long period of time and detailed
            analysis of the signal is possible
       2. More flexible signal processing algorithms are available to obtain the
            desired information directly from digital data.


       Real time digital spectrum analysis is possible using special hardware
configuration and customized FFT algorithms. State of the art design of the analog to
digital converters dictates the performance capability and development evolution of
the digital receivers. With present day technology, it is possible to digitize RF input
signals up to 4 GHz directly through A/D Converter.
       The main objective of present day digital receiver is to achieve, at least 1 GHz
bandwidth with high dynamic range and good frequency resolution. The frequency
resolution of the receiver is determined by the total sampling length. Weighting
function is applied to reduce the side lobes due to sampling.



            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                    IV - 8
3.2 Advantages of digital receiver:


   Main advantages of digital receivers are:
       3. Significantly smaller in size.
       4. Provide more functionality in lesser space.
       5. Simultaneous signal handling capability.
       6. Reduced RF and analog drift and bias.
       7. Better parameter measurement and higher accuracy possible.
       8. Software re-programming possible for meeting the desired operational
           goals.



3.3 Basic configuration of digital receiver

   Basic digital receiver is shown in fig 4. Digital receiver consists of three major
   sections, which are
           •   RF/IF front end
           •   Signal sampling unit
           •   Signal processing unit


RF/IF front end:


       The RF signal intercepted by the antenna is given to the RF front end section.
RF section consists of RF amplifiers, mixers, filters and IF amplifiers. RF signal is
converted to IF by mixing the RF with the proper LO signal and then amplified by an
IF amplifier. Further a band pass filter will be part of the RF section to remove the out
of band harmonics.


Signal sampling unit:


       IF output of the RF front end is given to signal sampling section. The
sampling speed of the ADC used here will depend on the bandwidth of the IF signal.
Sampling speeds of ADC have increased tremendously in the past few years. Today
ADC’s of the speeds up to 4GHz are available. The dynamic range of ADC is also
increasing with their resolution. Dynamic range of the ADC is approximated by

           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                   IV - 9
       DR = 20 log 2^N = 6N
       Where N= number of bits.


       With an 8 bit ADC , a dynamic range of 48dB can be achieved and this may
further be increased by increasing the bit resolution of the ADC. Proper formatting of
the data is also required before the data is given to the signal processing section.
Generally the serial data is to be converted to parallel form so that it can be given to
FPGA or DSP for processing.

Signal processing unit:


       ADC data is given to the signal-processing unit, which can be a high speed
DSP or FPGA or a combination of both. Signal processing algorithms to extract the
signal features are implemented here. The pulse descriptor word (PDW) is also
generated by the signal processing system. Today very high speed and high gate
density FPGA are available so that highly complex DSP algorithms can be
implemented at much faster speeds as compared to DSPs.


3.4 Configuration of digital receiver suitable for EW application:


       The main difference in conventional digital receiver and that applicable to EW
is that in EW, the parameters measured are to be carried out on a single received RF
pulses. There are many possible configurations for digital receiver as per the specific
requirements. The following three configurations are listed, which are required for
different types of radar signal processing as applicable to EW applications.


   a. Monobit Digital Receiver
   b. Sub-sampling Digital Receiver
   c. Digital Receiver as applicable to low probability of intercept signal detection


       The Monobit digital receiver is designed using high speed 2bits ADC and one
bit kernel. The monobit FFT and frequency processing circuits are implemented in



           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                  IV - 10
FPGA. The FPGA implementation of the design makes it real time processing and
high-density simultaneous signal can be handled.


          The sub-Nyquist method allows to wide band digital receiver to be built
whereby the sampling speed does not match the instantaneous bandwidth. The sub-
Nyquist receiver is basically an IFM type receiver implemented using digital
technique using I and Q channel processing . However in contract distinction to the
IFM receiver, the digital version can handle simultaneous signal because of FFT
operation.


          The low probability of intercept digital receiver is basically meant for
detecting and analyzing different frequency and phase modulated signal, which are
being used by most of the advance radar. This signal is almost impossible to detect
using conventional analog receiver and only possible by using the real time low
probability of intercept digital receiver.


          Out of three types of digital receiver, Monobit Digital receiver is explained in
detail.


3.5 Monobit principle:


          The Monobit kernel is the key component of the design Monobit digital
receiver. The purpose monobit kernel is to eliminate multiplication and keep only
adders and subtracts in the discrete Fourier transform chip design. The DFT can be
written as


                    X(k)             x(n) exp (- j2πkn / N ) , K= 0 to N-1
                           n = 0 to N-1

          Where j = √(-1) and N is the total number of sampled input points. In this
equation the result is obtained as a result of multiplication of two functions: the input
x(n) and the kernel exp( -j*2*π*k*n / N). If either one of these two functions is one
bit i.e. + 1 or – 1, the operation requires only addition. For the EW application
Monobit kernel function is implemented in the hardware. The kernel function is


              CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     IV - 11
rounded to + 1 and –1 or +j and –j. The rounded kernel function is mapped to time
decimated radix 2 FFT algorithm. If sampling is done at 2.5 GHz and number of FFT
points taken is 256 for coarse frequency, TOA and PW measurements. Pulse width of
100ns can be analyzed. 2K point FFT points may implemented to get the fine
frequency accuracy and resolution. For 2K point FFT total time taken for
accumulation of data will be 819.2 nS. Thus TOA, PW and coarse frequency output
will be provided every 100 nS while as fine frequency output will be provided every
820 nS.


       Frequency selection logic will select the correct input frequency and avoid
picking up the spurious signal. In the Monobit receiver the maximum signals to be
processed is limited to two. Two thresholds will be set to select the signal. If both the
signals miss the high threshold level only then the lower level will become effective.

3.6 Monobit Digital Receiver Configuration

The receiver consists of three major elements as shown in fig. 5
       1. IF front end.
       2.   Signal sampling and formatting section.
       3.   Signal processor

Input front end:

       Input signal from Omni channel front end is down converted to the frequency
range of 750 to 1250 MHz and is given to the IF front end. Signal then passes through
a limiting amplifier with 60dB gain to amplify the input and limit the output at a
constant level. After the limiting amplifier a band pass filter will limit the out of band
noise. This filter will have –1 dB pass band covering from 750 to 1250 MHz. The
non-linear characteristics of the limiting amplifier will cause capture effect, which
limits the instantaneous dynamic range. Down conversion to 750 – 1250 MHz is
selected to eliminate in band harmonics interfering during measurement.



Signal Sampler and Formatting system:

       Signal from the output of band-pass filter will be input to the 2 bit ADC.
Sampling will be done at 2.5 GHz. After ADC a windowing circuit will be

            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                   IV - 12
implemented. This circuit has two key functions: converts the serial data stream to
parallel and slows down the data rate by a factor of 16 i.e. (2.5 GHz / 16 = 156.25
MHz data rate) before it will be given to FPGA for processing. The slowing of the
data rate is necessary to accommodate the speed at which the FPGA can receive the
data. This slowing of data is done by storing the output of ADC in 16 latches each of
2 bits. The latches will be of ECL logic as fast storage is the requirement here. Once
all the 16 latches are filled, data will be transferred to the FPGA i.e. 32 bits at a time
will be transferred. Thus after every 1/156.25MHz = 6.4 nS, input data will get
transferred into FPGA. In the processor board design two extra latches have been kept
just after first two latches. The purpose of these two latches is to avoid the overwriting
of data in the first two latches.

Signal processor:

        Monobit digital receiver can resolve the simultaneous signal condition in real
time. The proposed digital receiver can cover 500 MHz bandwidth which can be
extended to 1 GHz by I and Q channel approach and can correctly process two
simultaneous signals. The design uses a Fast Fourier Transform (FFT) to obtain
frequency of up to two simultaneous signals. It has better sensitivity than IFM
receiver because of FFT Channelizes the input into narrower bandwidth. It has good
frequency resolution and good frequency accuracy


        Five stages implementation of algorithm is done in FPGA , which is shown
in figure 6. The function of the input subsystem is to receive the 32 bit of parallel data
that flows in consecutively from the demultiplexer store them and finally produce 256
sets of real data for FFT subsystem that measures coarse frequency, PW and TOA. It
will also produce 2K set of real data that will be used for fine frequency measurement.
The whole FFT radix 2 algorithm will be implemented in concurrent manner in the
hardware thus making the execution very fast in comparison to any other processor
implementation. Estimated time taken by this subsystem is less than 100 nS.


        The main function of initial sorting subsystem is to locate a maximum of four
signals from output of FFT subsystem that have the highest amplitude. Address, real
and imaginary number and flag bits of these numbers will be the output. The squaring
and adding subsystem will square and add each set of data and the output from this

            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                   IV - 13
subsystem will be computed result along with the respective addresses and flags.
Final sorting system will select the two highest peaks that will correspond to the two
signal frequencies. Also the pulse width and TOA measurement will be carried out
here from 256 point FFT output.



3.6      Simulation Results:

         Monobit design flow is given at figure 7. The Monobit FFT algorithm is first
tested using Matlab program. The program             simulates all possible simultaneous
conditions like
                  Two CWs
                  One CW one Pulse
                  Two pulses for various pulse widths.
Frequency of the input signal was limited to 750 – 1250 MHz zone. Random noise
(3dB SNR)is also added with the signal and its impact on signal detection presence
also studied. Fig 8 and 9 shows the MatLab plots of Monobit FFT for various
simultaneous signal conditions. From the plots it can be seen that some floor noise is
introduced in the FFT plots due to monobit FFT, but signal presence can still be
observed and after setting proper threshold signal information can be extracted.


4.0 Conclusion


      The DIFM discussed here includes advance feature from microwave and digital
domain for designing a frequency receiver to get higher accuracy, good dynamic
range, high sensitivity, modular and lower cost. However, the problem of
simultaneous signal remain a problem of DIFM receiver which is being solved using
the digital receiver The Monobit Digital Receiver is capable of processing
simultaneous signal with good frequency accuracy along with Pulse Width (PW) and
Time–of-Arrival (TOA) measured data on pulse by pulse basis. A low SNR signals
can be processed using the digital receiver, which is not possible using the anlog
receiver. After a thorough testing the hardware for a digital receiver a Multi Chip
Module (MCM) can be created which can include IF section, Sampling, and
processing algorithms. The Monobit concept can also be extended to analyze the low
probability of intercept signals by suitably developing the different algorithms. The

             CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                    IV - 14
       algorithms can be implemented in FPGA in real time Low probability of intercept
       signals processing is possible, which is more suitable for ES receivers.


5.0 References


1. Tsui J. (1995)
   Digital Technique for Wideband Receiver,
   Norwood MA: Artech House,1995
2. Tsui J & James P. Stephen, Sr.
   Digital Microwave Receiver Technology,
   IEEE Transaction on Microwave Theory & Technique,
   Vol. 50, No. 3, March 2002
3. C. Montgomery, B.Y. Tsui, David Pok, Chien –In Henry Chen
   ASIC design for Monobit Receiver
   IEEE 1997
4. Roberto Gomez Garcia & Mateo Burgos Garcia,
   Optimization of a Monobit FFT Radar Interceiver using a Genetic Algorithm
   IEEE 2004




                    CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                           IV - 15
                                    P
                                    H
                                    A                     Differential
            P
                                    S                     Amplifiers
            O
            W                       E        Detectors
            E                       C
Limiting    R                       O                                      E = A (ω ) Sin ω
Amplifier                           R
            D
                                    R
            I
            V                       E
            I                       L
            D                       A
            E                                                              F = A (ω ) Cos ω
                                    T
            R
                                    O
                                    R




            Fig. 1 Instantaneous Frequency Measurement




            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                   IV - 16
                              Phase              Phase
                              Detector           Quantizer
                         L1


                              Phase              Phase
                              Detector           Quantizer                     14 bit
           Homodyne
                         L2                                                  Freq Data
               &
RF Input   Delay Lines
2-18GHz                       Phase              Phase           Ambiguity
            Front end
                              Detector           Quantizer       Resolver
            Assembly     L3

                              Phase              Phase
                              Detector           Quantizer
                         L4

                              Phase              Phase
                              Detector           Quantizer
                         L5




                              DLVA               Signal
                                                 Present


           Fig 2 Block diagram of Homodyne based Frequency Rx.




              CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     IV - 17
                                                     M3
                                        Delay Line
                                                     M3
                                    P                                    Phase
                              RF    D                                    Discri-   I
                                                                         minator
                          P         P                                              Q
RF 2-18 GHz                                                     160
                          D         3                     M2
                                                 P              MHz
                                                 D
                          P
                                   M1
                          1                      P    RF + LO
                                                 2
                                        LO=160
                                        MHz




                      Fig 3 Homodyne delay line discriminator




          CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                 IV - 18
Antenna




                  RF                    Signal                    Signal
                  Front End             Sampling                  Processing
                                                                               Pulse
                                                                               discriptor




                  Fig:4 Basic Digital Receiver Block diagram




          CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                 IV - 19
            Limiting
            Amp
                       BPF              2.5        Win-              FFT1
                       750 –            GHz
                                                   dow               2K         Fine
                                        2bit
                       1250
                       MHz              ADC        ckt                          Frq.
IF I/P                                                                                      Data
750 -1250                                                                                    O/P
MHz                                                                                         logic
                                                                     FFT2                              Freq.,
                                      2.5GHz       Clk/16            256pt       PW,TOA                PW,
                                      clk                                        & coarse              TOA
                                                                                   Frq

            IF Front End           Sampling and formatting          Algorithm implementation in FPGA




                                   Fig 5 Block diagram of Monobit Digital Receiver Configuration



                           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                                  IV - 20
32- b input                                                                                   Final
                                                                       Squaring               sorting   Freq Data,
              Input                                                    &
                                   FFT                Initial                                 and       PW,
              Subsystem            Subsystem          Sorting          Addition
                                                                       Subsyste               PW,       TOA
                                                      Subsystem
                                                                       m                      TOA
                                                                                              Estimat
                                                                                              es




                          Fig- 6         Block Diagram for Algorithm implementation in FPGA




                               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                                      IV - 21
                         Concept




                               MatLab        Simulation & Concept verification




                      Synthesis
   VHDL                   +                       FPGA                        Monobit Chip
   Coding               Place
                      and Route

Simulation             Simulation          Testing in H/W
    &                     &
Verification           Verification




  Fig – 7      Monobit Chip design flow




               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                      IV - 22
          0 .5                                                                                                                         A Time domain CW on CW
                                                                                                                                       signal
              0


         -0 . 5


            -1
                                        100               150             200                  250                       300
                                                                                                                                       Time samples -
              1

Amp V     0 .8

          0 .6

          0 .4
                                                                                                                                       B Monobit 256 point FFT for a A
          0 .2

              0
                      0             2           4               6         8              10             12                14
                                                                                                                              8
                                                                                                                       x 10
                                        Frequency -




                                                                                                                                        C Time domain Pulse on Pulse
          0.5
                                                                                                                                        PW = 100nSec
             0



          -0 .5



            -1
                          200             400       600         80 0   1 00 0   1 20 0         1 4 00         1 6 00
                                                                                                                                        Time samples -
             1


          0.8


          0.6
 Amp V
          0.4
                                                                                                                                        D Monobit 256 point FFT for C
          0.2


             0
                  0             2               4               6          8              10             12                       14
                                                                                                                                  8
                                                                                                                          x 10




                                                Frequency -
                                                Fig 8

                                                                       CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                                                                              IV - 23
          0.6

          0.4                                                                                                                                                              A Time domain Pulse on pulse
          0.2
                                                                                                                                                                           PW = 100 nSec
             0

         -0 . 2
                                                                                                                                                                           SNR 3dB
         -0 . 4

         -0 . 6


                          100                 200           300           400           500              600           700          800           900




          0.5                                                                                                                                                      Time samples -
          0.4


          0.3


          0.2


Amp V                                                                                                                                                                        B Monobit 256 point FFT for A
          0.1


             0
                  0                   2                     4                   6                    8                10                  12               14
                                                                                                                                                               8
                                                                                                                                                        x 10



                                                          Frequency (Hz) -



              0.6                                                                                                                                                          C Time domain Pulse on CW
              0.4                                                                                                                                                          PW = 100 nSec
 Amp V        0.2                                                                                                                                                          SNR 3dB
                  0

            -0.2

            -0.4

            -0.6


                                200                 400             600         800           1000             1200          1400          1600          1800

                                                                                                                                                                     Time samples -
              0.7

              0.6

              0.5
 Amp V
              0.4

              0.3
                                                                                                                                                                           D Monobit 256 point FFT for C
              0.2

              0.1

                  0
                      0                   2                     4                   6                8                10                  12               14
                                                                                                                                                               8
                                                                                                                                                        x 10




                                                      Frequency (Hz) -
                                                      Fig 9



                                                                                                     CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                                                                                                            IV - 24
               CHAPTER – V (1)


DIRECTION FINDING FOR EW SYSTEMS

      Shri. K.R. SUNDARAM, Sc-‘F’




      CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                V (1)
                       CHAPTER – V(1)




     DIRECTION FINDING FOR EW SYSTEMS




                           CONTENTS



1.   INTRODUCTION


2.   WHAT IS A PASSIVE DF SYSTEM


3.   ROTARY DIRECTION FINDING SYSTEMS


4.   SIDE AND BACK LOBE INHIBITION TECHNIQUES


5.   TIME DIFFERENCE OF ARRIVAL DF


6.   AMPLITUDE COMPARISON DF SYSTEMS


7.   REFERENCES




      CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                V (1)
            DIRECTION FINDING FOR EW SYSTEMS

                          KR Sundaram, Sc ‘F’

                         DLRL, HYDERABAD


1.     Introduction :
       The Direction of Arrival (DOA) of a radar signal is the most important
parameter that has to be measured for locating the Radar using an EW system.
This is because the DOA is the only parameter that cannot be camouflaged by the
enemy radar. Other parameters like frequency, pulse width, PRF etc can be varied
to incorporate ECCM features in the radar.


2.     What is a passive DF system :
       In a EOB finding the location / direction of the enemy radar without
giving away our position is of paramount importance. In such situations it will be
prudent to use a passive direction finding system which uses the enemy’s radar
emissions to locate the enemy without the requirement of a radar.


       The passive direction finding system can be implemented using various
techniques like Rotary Direction Finding (RDF), Amplitude Comparison
Direction Finding (ADF), Time Comparison DOA (TDOA), Phase Comparison
etc. Each of them Has its own specific applications and role in an electronic order
of battle (EOB).


       Each of the above systems makes use of the difference in one of the
signal parameters like phase, amplitude, time etc between two sensors.


3.     Rotary Direction Finding systems :
3.1    Principle :
       The basic principle employed in RDF is sampling the azimuth
continuously at a convenient rate. Signals picked up by the sensor are detected

         CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                 V (1) - 1
and then suitably processed and the amplitudes of each sample compared against
the positioner data to compute the DOA.
       The RDF makes use of a single sensor, which is a high gain narrow beam
antenna scanning the azimuth. The requirement of the high gain and narrow beam
width are commensurate with the DF accuracy required. In general the beam
width of the antenna varies from 2 – 10 degs. and the gain varies from 15 – 25 dB.
As a thumb rule the available accuracy from an RDF system is 1/10th the Beam
width. The sensitivity of the system depends on the instantaneous bandwidth of
the front end receiver. The typical receiver sensitivity figures for a 10 GHz
bandwidth(8-18 GHz) is around –60 dBm and hence the system sensitivity will be
around –75 to -85dBm. With a narrow band frontend receivers much better
sensitivity figures are possible.


3.2    Algorithms :
       The most commonly used processing algorithms in RDF systems is based
on centroiding.
       Centroiding is done on the basis of the amplitudes received at different
azimuthal samples of the signal by the antenna. Consider the case of a single
emitter.


                                                                    Gp




                                                                     Gs


                Gb




                                    Fig : 1




           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                   V (1) - 2
       The amplitudes of successive samples from the rotating antenna output
will be as given in fig 1. To get the actual direction of arrival of the signal the
processor looks for the peak amplitude and consider the DF antenna as it spins
continuously, intercepting, displaying, and / or encoding all received signals as
shown in a typical antenna beam width pattern in Fig1. This particular antenna
has a gain of Gp and a 3 dB beam width W, which is made as narrow as is
consistent with rotation rate and desired accuracy. (Gains of 12 dB, beam widths
of 30° and rotation rates from 15 to 150 rpm are typical.) The first sidelobe
response level exhibits a gain G, with a maximum backlobe gain of Gb width
values of 15 and 30 dB, respectively,        as typical. The usable unambiguous
dynamic range of this antenna can be seen to be
   Antenna dynamic range = (Gp - Gs) in dB


     This is the case for signals that do not overload the antenna or receiver and
are above the desired threshold of detection. It is also assumed that the signal
amplitude variation and pulse parameters are within appropriate limits to permit
detection throughout at least one full receiver antenna rotation to assure that the
receiver DF beam maximum is displayed.


       Processing was done manually in the initial stages where the operator can
position a cursor on a CRT with appropriate gain adjustment to ensure optimum
visual display. In the present day rotary DF systems the processing is done
automatically. The most common algorithm used for estimating the DF is the
Centroiding algorithm. The algorithm is required since the DOA calculated on
the basis of the peak position of the signal is likely to be affected by multipath /
reflections and is not always reliable. However the nulls of a rotary DF antenna
are much less affected by reflections and multipath. In simple terms centroiding
means that    the position of the peak of the pattern is determined at every
frequency. The positions where the signal amplitude falls by 10 dB on either side
of the peak at that frequency are noted. The arithmetic mean of the –10dB
positions for a given frequency is considered as the direction of Arrival of the

         CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                 V (1) - 3
signal at that frequency. Automatic processing enable multi signal processing
within the instantaneous bandwidth of the system.


4.0    Side and back lobe inhibition techniques:
       To reduce or inhibit the undesirable lobe responses in the single-channel
system two antennas and receivers are used in a dual channel system. Here the
first antenna is the same rotary DF type as before, but the second is omni
directional , designed to cover the same spatial elevation angle (H plane) having a
lower relative gain. When each antenna is connected to one of a set of receivers
that are amplitude matched over the desired frequency range, the relative gains of
the two channels can be adjusted to position the omni azimuthal channel gain
between the DF main and side lobe levels.


       Under these conditions, simple logic can be used to permit acceptance of
only those signals that meet the above criteria; namely, DF output will be
indicated only when the DF signal is greater than the omni azimuthal output,
which in turn must be greater than the DF side and back lobe levels by definition.


       Advantages :
       •        The RDF system gives high accuracy
       •        It also gives large range due to the high antenna gain
       •        Simple processing algorithms


       Disadvantages :
       •        Very slow processing
       •        Very low probability of Intercept
       •        PRF Vs ARR considerations will affect system performance




           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                   V (1) - 4
5.      Time difference of Arrival DF :


5.1     Principle :
        TDOA is based on the difference in time of arrival of the signals received
by two sensors due to the physical separation of the antennas. Consider two
antennas separated by distance ‘d’ receiving the same signal (planar wave front).




                                           d




                                                fig 2


The difference in time of arrival t is given by
                          t= d sin θ / c
                where d is the distance in meters
                          θ is the angle of arrival
                          c is the velocity of light
        from the above equation the angle of arrival θ can be computed as
                          θ = sin-1 (t c / d)
        If an array of two omni directional antennas is used to cover the full
azimuth, there exists a front to back ambiguity in the DOA because a signal from
the mirror image direction also gives identical DOA. To resolve this front to
back ambiguities a third antenna is introduced so that two orthogonal base lines
are available as fig 3.




          CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                  V (1) - 5
                 d ; time = t1




                                     d
                                 time = t2
                                          fig 3
       The time delays for the two orthogonal base lines are given by
                       t1 = d sin θ / c
                       t2 = d cos θ / c
       from the above two equations the DOA of the signal can be computed as
                       θ = tan-1 ( t1 / t2 )
       Though the tan –1 is ambiguous mathematically, it is possible to resolve the
ambiguity in the polarities by knowing the polarity of the time delays of the
numerator and denominator. Eg. If the t1 and t2 are positive it means that the
angle of arrival is in the first quadrant ( All Silver Tea Cups ).


5.2    Accuracy considerations :
       The accuracy obtained in the TDOA system is directly related to the
accuracy with which the time delays between the two antenna outputs can be
measured. This is again related to the clock frequency used to measure the delay.
               Tan θ = t1/t2
               θ = tan –1 (t1/t2)
if δt1, δt2 are the errors in the measurement of t1 and t2
               σ θ = tan –1 ( (t1 + δt1) / (t2+δt2) )
from the above equation it can be seen that the maximum errors occur when
t1 = t2 when δt1 and δt2 are small compared to t1 and t2.
It can also be seen that the minimum errors occur when t1= 0 or t2=0.


          CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                  V (1) - 6
5.3    Calculation of baseline distance required for a specified accuracy:
       Accuracy Required : 1° (rms) (2° pk)
       Clock used : 2 GHz
       peak error : 2 nS (let δt1 = δt2)
Substituting the above values in the error equation :
                σ θ = tan –1 ( (t1 + δt1) / (t2+δt2) )
                σ θ = (45° + 2°) = tan –1 ( (t + 2) / (t-2) )
from the above equation we find that the time required to meet the accuracy
requirement is 57 nS. Substituting for t in
                t = d sin θ / c
       we get d ≈25mts.
The TDOA system has typical sensitivity of the order of -60dBm.


5.4    Advantages :

       •        Very simple concept
       •        Simple processing algorithms
       •        System has very broad band coverage
       •        System is immune to multipath effects unlike the phase and
                amplitude comparison systems which give gross errors in
                multipath environment



5.5    Disadvantages :

       •        CW signals processing not possible
       •        Large Baseline Span required to achieve good accuracy.
       •        Not suitable for all the platforms due installation related issues
                (mostly used on airborne platforms)
       •        Very high frequency clock signals and processing circuits required




           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                   V (1) - 7
6.     Amplitude comparison DF systems

6.1    Principle of operation :

       Amplitude comparison uses multiple directional antenna receiver system,
each antenna pointing toward a different direction.         The antenna facing the
direction of the emitter will generate the strongest output from the receiver
connected to this antenna.     By comparing the different amplitudes of these
outputs, the AOA information can be obtained.


       The number of antennas to be used and the beamwidth of each antenna is
related by
               Beam width = 360 / number of antennas in the array.
       For an 8 antenna amplitude comparison system the typical beam width of
each antenna will be 45°. Practical antennas have a beam width variation over the
frequency of the order of ±15°
The typical antenna pattern are given in fig 4.


       Consider two antennas 45° apart. For a signal incident on the boresight of
one of the antennas the amplitude of the signal at the output of the adjacent
antenna is approximately 6 to 10 dB down depending on the beamwidth and the
rolloff. For a signal coming in between the two boresights the two amplitudes
will be approximately equal. This means that over a 22.5° variation in the
azimuthal DOA of the signal the variation in the amplitude is around 8 dB. This
differential amplitude is used to compute the DOA of the signal.




         CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                 V (1) - 8
                                           Fig 4


                                                       45°
             first antenna boresight


amplitude rxed by 2nd antenna                                              main lobe
6-10 dB less                                                                side lobe




6.2       DF accuracy considerations :


          The accuracy of amplitude comparison DF system is directly dependent on
the amplitude differential of the two antenna outputs as the DOA varies from the
boresight of one of them to the bisector of the two boresights. Assuming a of ±3
dB mismatch between the two RF frontend chains the DF accuracy for an 8
antenna DF system is given by
DF accuracy = 3 x 22.5 / 8 ≈ 9°.
          A more general and analytical equation for estimating the DOA accuracy
is given by
                 σθ = BW2 * ∆c / 24 * S
                 where σθ is the DF error
                 BW is the beam width of the antenna
                 ∆c is the channel mismatch in dB
          S is the angular spacing between the antennas in degs.


6.3       Calibration :
          The calibration of the amplitude comparison DF system is done in two
stages.


            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                    V (1) - 9
       RF front end calibration : The DLVA is commonly used as the RF
frontend in an amplitude comparison DF system.              The DLVA is a detector
followed by a log video amplifier. A logarithmic video amplifier is used to give a
linearly varying video output which is proportional to the input power in dBm.
Using a DLVA will enable the DF system to cover a large instantaneous dynamic
range of the order of 60 to70 dB. In general the DLVA will have an inherent
nonlinearity due to design limitations. However, the linearity error in a DLVA is
generally a systematic error for a given frequency and input power. It is therefore
possible to calibrate the DLVA by feeding the signals.              A lookup table is
generated in which the error in the amplitude to be calibrated is stored against the
frequency address. Calibration can improve the channel amplitude balance of ±3
dB to ±1.5 dB.


       Antenna roll off calibration : As already mentioned earlier the power
level difference between the outputs of the two antennas facing the signal
direction will vary as a function of frequency. This variation is also systematic
and hence can be calibrated using a lookup table against a frequency address.


       The theoretically achievable DF accuracy in an 8 antenna amplitude
comparison DF system is around 2° rms. However practically achievable results
will degrade the DF accuracy due to multipath and reflections.


6.4    Advantages :


       •        Broadband coverage
       •        Reasonably good accuracy
       •        Simple processing algorithms


6.5    Disadvantages :
       •        Prone to reflections and multi path effects
       •        Heavily dependent on calibration
           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                   V (1) - 10
7.   References :


1.   Microwave passive direction finding ; Stephen E Lipsky
2.   Microwave receivers with Electronic Warfare applications; JBY Tsui




       CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                               V (1) - 11
                    CHAPTER – V (2)


ANALYSIS OF A BUTLER MATRIX BASED
            DF SYSTEM

                       Shri. M.K. DAS, Sc-‘F’




    CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                               V(2)
                        CHAPTER – V(2)




      ANALYSIS OF A BUTLER MATRIX BASED
           DIRECTION FINDING SYSTEM




                           CONTENTS



1.    INTRODUCTION

2.    BUTLER MATRIX VS FFT

3.    DIGITAL BEARING DISCRIMINATOR

4.    CIRCULAR ARRAY

5.    BUTLER MATRIX

6.    THE DIRECTION OF ARRIVAL

7.    CONCLUSION

8.    REFERENCES




     CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                V(2)
         ANALYSIS OF A BUTLER MATRIX BASED
             DIRECTION FINDING SYSTEM
                 Shri M K DAS, Sc ‘F’, DLRL, Hyderabad



1.0      Introduction :
       Direction Finding system based on Butler Matrix and circular array such as
digital bearing discriminator (DBD) needs a careful analysis and good simulation studies.
In DBD more signal processing is involved at microwaves than other Direction Finding
(DF) systems like Phase interferometer or Rotary DF or Time Difference Of Arrival
etc.[1,2]. Thus a DBD needs as exact as possible theoretical study either before starting
the development of hardware or in the process of improving the hardware. The aim of
this paper is to model the system and obtain a general behaviour, which will lead to better
design, error budgeting and better understanding of problems and limitations.


       A Butler Matrix is a realization of Fast Fourier Transform (FFT) of aperture
domain to mode domain. It can be the conventional N-point to N-point transform or an
N-point to K-point (K<N) transform [3,4]. In a DBD the latter is used since beyond a
certain value of K the data obtained in the mode domain will not be useful. The highest
K that can be useful can be derived from the theoretical analysis to be presented.


2.      Buttler Matrix vs. FFT
       FFT computation and Butler Matrix signal processing can easily be understood in
the following manner [3,4].           FFT/BM transforms the time/aperture domain into
frequency/mode domain, representing a continuous time function/aperture distribution by
a sampled set of data/an array. Since the sampled data is in discrete form discrete Fourier
transform (DFT) is used in place of conventional Fourier transform.
       The general form of DFT can be written as

                               N -1        2π
                        F(k) = ∑ f(n) e- j N kn      ..... (1)
                               n=0



       Where F(k) are samples in frequency/mode domain and f(n) are samples in
time/aperture domain.

      CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                               V(2) - 1
For N=4 in the matrix form the above DFT equations can be written as


       F(0)        1     1     1    1       f (0)
                                                    …….. ( 2 )
       F(1)    = 1      ω     ω2   ω3       f (1)

                                                                       -j 2π / N
       F(2)        1   ω2    ω4 ω6       f (2)
                                                            where   ω=e
       F(3)        1   ω3    ω6 ω9       f (3)


       The left column matrix represents the frequency domain samples and the right
most column matrix represents the time domain samples. The square matrix is the
Twiddle Matrix, which can be factored in to different matrices as shown below (Fig.1a).
The first square matrix from left is the rearrangement matrix, which rearranges the
aperture domain samples. Rest of the two square matrices are called Butterfly matrices
which act on the rearranged data to give the Fourier transform in the mode domain. The




      CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                               V(2) - 2
Butterfly matrices give an idea of physical realisation using various fundamental
microwave components. One such realisation is shown in Fig.1b.
        By observing the factored matrices it can be seen that the first Butterfly matrix is
a combination of two 180o Hybrids and the second Butterfly matrix is a combination of a
180o Hybrid with a 90o phase shifter and 180o Hybrid. Similarly Butler matrices with
higher number of antenna ports can be designed by choosing appropriate hybrids and
phase shifters.


3.      Digital Bearing Discriminator:
        If the aperture domain samples are supplied by a circular array of N-antenna
elements, then very useful properties at the mode domain, at the output of the Butler
matrix, can be observed. If the aperture domain samples are due to an emitter at far filed
distance traversing round the antenna array then the mode domain samples will have
phase variation of 0, θ, 2θ...mθ.... where θ is the angular location of the emitter with
respect to a reference element on the antenna array. Thus by simply measuring the phase
variation at various mode ports one can obtain the bearing of the emitter.


        In order to model the Butler Matrix and the circular antenna array it is more
convenient to look from the transmitting point of view. If a mode port is excited then a
current distribution takes place on the antenna elements. This distribution will give raise
to amplitude and phase distributions at the far field point. The amplitude distribution is
omni and the phase distribution is linear. By reciprocity same amplitude and phase
distributions occur at the mode domain in a receiving situation.


        In the discussion to follow first the equation for far field of a circular array
excited by a Butler matrix is derived without specifying what sort of elements constitute
the array. The effect of array factor and the array radius only are considered. Once a
specific element is chosen for analysis the pattern factor of that element comes into
consideration.


4. Circular Array :
        Let us derive the expression for the far field by exciting the elements of a circular

      CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                               V(2) - 3
array by the current distribution produced by a Butler matrix when one of its modes is
excited [5,6] as shown in Fig.2.




Let    R = Array Radius
P = Point having polar coordinates (r,θ,φ)
Let current in the nth element is given by

                                    j
                         a n = An e α n   ....(3)




      CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                               V(2) - 4
For isotropic radiators the far field at the point P(r,θ,φ) is given by
                R Sinφ Cos (θ-θn) is neglected from the denominator for far field approximation


                                        N -1       exp[ jkr ]
                     ψ (r,θ ,φ ) = Σ a n
                                                            n
                                                                                          .....(4)
                                        n=0
                                                       rn
                                  exp[jk(r - R Sin( φ ) Cos( θ -θ ))]
                                    N -1                                                           n
                             = Σa          n
                                    n=0
                                       r - R Sin( φ ) Cos( θ -θ )                             n


                                exp(jkr)
                             N -1
                           = Σa     n    exp[-jkR Sin( φ ) Cos( θ -θ )]                                 n
                             n=0
                                   r

        since at a far field point r>> R Sinφ Cos (θ-θn).
                In the above equations exp(jkr)/r can be dropped out since it does not disturb the
        shape of the amplitude and phase distributions at the far field point.


                            exp[jα - jkR Sin( φ ) Cos( θ -θ )]
                    N -1
        ψ ( θ ,φ ) = Σ An                      n                                     n             ......(5)
                    n=0



                In deriving the above expression the type of element in the circular array was
        taken to be isotropic.             However in practice the radiating elements used will be
        directional. the pattern factor of a slot line antenna, which is the most often used in DBD
        application, can be approximated by a mathematical function given by

                                                                ML g
                   PatternFactor(PF) = E( θ ) = 10(             20
                                                                     )
                                                                         cos ( θ )
                                                                            p
                                                                                         .......(6) 2

                where p is a real number, which controls the beam width of the antenna element
        and MLg is the gain of the main lobe expressed in dB. This model may not describe the
        practical pattern perfectly, but it is mathematically convenient and the conclusions based
        on this can give a general picture, which is our aim. When p=0, it represents an isotropic
        radiator. With this type of an element in the circular array the far field will be




               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                        V(2) - 5
                    PF exp[j( α - kR Sin( φ ) Cos( θ -θ ))] ...(7)
           N -1
ψ ( θ ,φ ) = Σ An                      n                                  n
           n=0



5.      Butler Matrix :
        Having established the model for CAA let us examine the Butler matrix
modelling. Without going in to the details of the constituent components of the Butler
matrix, it can be modelled just looking at the S21 relation between mode ports and
antenna ports. The amplitude distribution should be uniform on all the antenna ports for
the case of a loss less Butler matrix, and phase distribution at nth element will be perfect
linear progression.      keeping in mind the fact that the number of antenna ports is
essentially a radix of 2, each antenna port will have an amplitude in terms of dB

                              X n = log2 N*(-3)     ......(8) 3



In terms of fraction of current/voltage wave variables

                                                       (    /20)
                  An = S 21 (at each antenna port) = 10 X n        .....(9)


The phase distribution at nth element will be



                                                  2π
                              α n = ang( S 21 ) =    * m* n
                                                  N
                                           where m is mth mode       ......(10)

        This modelling of Butler matrix is exact and is simple since it is looked as a black
box. By perturbing the Xn and αn we can create a practical Butler matrix. Similarly by
perturbing the pattern factor and the phase term in ψ(θ,φ) we can create a practical
circular array antenna.


6.      The Direction of Arrival :
        The DOA from a DBD is always obtained from the highest mode with acceptable
phase non linearity. The lower order modes are used only as ambiguity resolvers.
Depending on the algorithm chosen, the acceptable error margins of the lower order
modes is decided. The expression for DOA can be written
     CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                  V(2) - 6
                                                Φ
                                      DOA =       ....( 11 )
                                                m



      where Φ, the phase of the highest mode, is given by

            N -1               2π                2π R                                
            Σ Ar * PF * Sin( ( N .m.r + BM pe ) - λ Sin( φ ) Cos( θ - θ r ) + A pe ) 
Φ = tan -1  N=0
             r
                                                                                       ...( 12 )
            A * PF * Cos( ( 2π .m.r + BM ) - 2π R Sin( φ ) Cos( θ - θ ) + A ) 
               -1

             Σ
            r=0 r
                                N
                                            pe
                                                   λ
                                                                         r       pe
                                                                                      
                                                                                      

              The DOA will also get affected by the tracking of the RF channels, phase
      correlators and digitization errors. The first two are in general taken care of by on line
      calibration and the phase error due to digitisation is well known and can be minimised so
      that they can be neglected.


      7.0     Conclusion :
                      Simple modellig of the Butler matrix and circular array of a digital
      bearing discriminator leads to very vital clues in the design of the DBD. The effect of
      radius of the array and the type of radiator have been found to be the starting points of the
      design. Apart from these the simple model and simulation lead to error budgeting of
      various subsystems.
      8.0      References :
      1.      J.B.Y.Sui, "Digital Microwave Receivers Theory and concepts", Artech House,
              Norwood, 1989.
      2.      Stephen E. Lipsky, "Microwave passive direction finding systems", A Wiley
              Interscience Pub., New York, 1988.
      3.      Ronald N.Bracewell, "The Fourier transform and its applications", Mc Graw-
              Hill, 1978.
      4.      W.H.Nester, "The Fast Fourier Transform and the Butler Matrix", IEEE
              Trasnsactions on Antennas and Propagation, May 1968, P.360.
      5.      Wolf, E.A., "Antenna Analysis", John Wiley & Sons Inc. NewYork, 1966.
      6.      Boris Sheleg, "A matrix-fed circular array for continuous scanning", Proceedings
              of IEEE., Vol.56, No.11, Nov.1968, PP 2016-2027.

             CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                      V(2) - 7
             CHAPTER – V (3)


 INTERFEROMETRIC DIRECTION
FINDING SYSTEMS FOR RADAR EW
         APPLICATIONS


    Shri. K.R. SUNDARAM, Sc-‘F’




    CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                               V(3)
                            CHAPTER – V(3)




INTERFEROMETRIC DIRECTION FINDING SYSTEMS FOR
                    RADAR EW APPLICATIONS




                                CONTENTS




1.    INTRODUCTION
2.    THEORY OF OPERATION
3.    PHASE ERROR MARGIN
4.    ACCURACY CONSIDERATION
5.    SENSITIVITY
6.    DYNAMIC RANGE
7.    FREQUENCY & ADF REQUIREMENTS
8.    CALIBRATION
9.    SYSTEM BITE
10.   PROCESSING TIME
11.   ADVANTAGES
12.   DISADVATAGES
13.   REFERENCES




       CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                  V(3)
INTERFEROMETRIC DIRECTION FINDING SYSTEMS FOR
                      RADAR EW APPLICATIONS


                K R Sundaram, Sc ‘F’ , DLRL, HYDERABAD




1.     Introduction
       The Direction of Arrival (DOA) of a radar signal is the most important
parameter that has to be measured for locating the Radar using an EW system.
This is because the DOA is the only parameter that cannot be camouflaged by the
enemy radar. Other parameters like frequency, pulse width, PRF etc can be varied
to incorporate ECCM features in the radar.


       Several techniques are available for measuring the DOA of a radar signal,
like Rotary Direction Finding (RDF), Amplitude Comparison Direction Finding
(ADF), Time Comparison (TDOA), Phase Comparison etc. Each of them have
their own specific applications and role in an electronic order of battle (EOB).


       There are two ways of realizing a Direction Finding (DF) system
employing interferometry.     The first approach uses a circular antenna array
followed by a complex Microwave Network called “Butler Matrix” and is
commonly referred to as Digital Bearing Discriminator (DBD).


       The second approach makes use of linear antenna arrays and is often
referred to as Baseline Interferometer (BLI).


       The BLI approach gives very good performance features like good DF
accuracy of 1° RMS, sensitivity of –60 dBm, broad band coverage (octave and
multi octave), high processing speed (1.2 µS) etc. Brief details of the BLI DF
systems employing linear antenna arrays are given in this paper. Some of the

         CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                  V(3) - 1
critical processing algorithms have been explained. System parameters like
Sensitivity, Dynamic Range etc have been quantitatively derived.


1.0    PRINCIPLE OF OPERATION :



                            C

              D Sin θ
                                               θ
                                      θ
                            D




                        Fig.1 Principle of Interferometer


Consider a two element interferometer array as shown in fig 1. Consider a plane
wave signal incident on the array at an angle θ with respect to boresight of the
array. The phase delay ψ across the antenna outputs is given by
                           Sinθ
              ψ = 2π D                                         --------(1)
                             λ
       where θ is the DOA w.r.t. the bore sight axis
       λ is the wave length of the incident signal
       and D is the spacing between the two antenna elements
If the phase delay is measured and the frequency (hence wavelength) is known the
Angle of Arrival of the signal can be computed as:


              θ = Sin-1(ψλ/2πD)                                         -------(2)


From the above equations it is obvious that to get a higher DF accuracy one has to
go for a higher spacing between the antennas.            However since the phase
measurement is modulo 2π, the spacing between the antennas should be less than


         CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                  V(3) - 2
λ/2 to avoid ambiguous phase measurements. However in a practical situation it is
difficult to find antennas which can meet the λ/2 spacing requirement as well as
meet the system sensitivity requirement. This is because the aperture of the
antenna is dictated by the lowest frequency of operation and hence will be much
more than λ/2 at the highest frequency of operation.


       The main problem in an interferometer is the ambiguity in the phase
measurement because of the large base line spacing required to achieve good DF
accuracy. Several techniques used for resolving the phase ambiguity have been
reported in literature. One of them which has been developed in DLRL makes use
of an array of 4 antennas where the inter-antenna spacing bear a ratio of prime
integers. The array spacing for 8-18 GHz band are indicated in figure 2.           The
synthesis of the array spacings is a function of a number of parameters like
frequency coverage, DF accuracy required, azimuth coverage phase error margins
required, processing algorithms and dimensions of the antennas chosen. More
details regarding array synthesis are given in references [1, 2, 4, 6,].

  A1                          A2                          A3                      A4


                D1                             D2                          D3
                22mm                           44mm                        33mm
                                   D4
                                        99mm

                 Fig.2 Antenna array for 8-18 GHz BLI DF system


Consider a 4 element antenna array (using antennas A1, A2, A3 and A4) as shown
in fig.2.      The antennas generally used are Cavity back spirals which give
excellent phase tracking and are conformal.            The inter antenna spacing are
22mm(D1), 44mm(D2), 33mm(D3) and 99mm (D4). It is possible to obtain six
different phase measurements (combination of 4 antennas taken 2 at a time) from




            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     V(3) - 3
this array. However, only 4 phase measurements are required for the purpose of
phase ambiguity resolution and subsequent correct DOA estimation.


2.     Theory of Operation :
       The total phase delay and the measured phase angle for the different
spacing D1 to D4 are given by the following relationships.
ψ1 = (2 π 22 / λ) Sin θ = φ1 + 2π m1                   ----- (3)
ψ2 = (2 π 44 / λ) Sin θ = φ2 + 2π m2                   ----- (4)
ψ3 = (2 π 33 / λ) Sin θ = φ3 + 2π m3                   ----- (5)
ψ4 = (2 π 99 / λ) Sin θ = φ4 + 2π m4                   ----- (6)


       For the convenience of analysis and subsequent hardware implementation,
all phase measurements are assumed to be from 0 to 2π. It is to be noted that
phase angles greater than π will be actually negative and hence we subtract 2π
from such angles to get the actual polarity of the phase angle. In the above
equations ψi is the total phase delay for the spacing Di, φi is the measured phase
delay for ψi and mi is an integer. Since the longest spacing in the array is D4 it is
necessary to estimate m4 to compute the total phase delay for that spacing and
hence the most accurate DOA of the signal.for the array


       A variation of the Chinese Remainder Theorem (CRT) is extensively used
to resolve the ambiguity in measured phases and estimate the value of m4. The
CRT has also been used extensively in other systems like DIFM receivers (for
resolving phase ambiguity) and signal processing for unambiguous range
estimation using staggered PRI in Radar systems.


       The Chinese Remainder Theorem states that “If any positive integer N is
divided by different positive prime integer divisors, the set of remainders so
obtained will be unique as long as the number N satisfies the condition 0 < N <
(LCM of all Divisors) ”. It is to be noted that if N > (LCM of Divisors), the set


         CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                  V(3) - 4
of remainders will repeat itself. References [6,7] give a generic form of
application of the CRT for Interferometers and DIFM receivers.


       It can be seen that the ratio D1:D2=1:2 which means that the total phase

delay for the spacings D1 and D2 will also be in the ratio 1:2 From which we get


               2ψ1 = ψ2
               i.e 2 ( φ1 + 2π m1) = φ2 + 2π m2                         ----- (7)
       rearranging the terms in (7) we get
               (2 φ1 - φ2 ) / 2π =    m2 - 2 m1                         ------(8)
Since m1 and m2 are integers, the LHS of equation (8) also has to be an integer.
       Since φ1 and φ2 are measured and known quantities the LHS of equation
(8) can be evaluated. In case it results in a non-integer value, due to errors in
phase measurement, the result is rounded off to the nearest integer. Let this be k1.
Hence we get
               m2 - 2m1 = k1                                            ------(9)
       Equation (9) is a linear equation with 2 unknowns and hence will
theoretically have infinite number of solutions. However the practically valid
solutions are limited since the absolute value of Sin θ cannot exceed unity.
Solutions of m2 from equation (9) can be given in the form of an arithmetic
progression
               m2 ∈ { 0 , 2, 4, 6, … }                                  -----(10)
       if the value of k1 is an even integer and
               m2 ∈ { 1, 3, 5, 7 , … }                                  ------(11)
       if the value of k1 is an odd integer.


       However, practically, the number of solutions depend on the spacing D2
and are bound by equation (4). From equations (9), (10) and (11) we can write
               m2 = m21 + 2M2                                           -----(12)
       where m21 =0 if k1 is even in equation (9)

         CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                  V(3) - 5
       and m21 =1 if k1 is odd in equation (9).
       Substituting the value of m2 from equation (12) in equation (4) we get
                 ψ2 = φ2 + 2π m21 + 4π M2                                ----- (13)
                 ψ2 = (2π 44 / λ) Sin θ = φ21 + 4π M2                    ----- (14)
       where φ21 = φ2 + 2π m21 and M2 is any integer
       The valid solutions of M2 from are bound by equation (14)
       It can be seen from equation (14) that the valid solution of M2 for 44mm
spacing is either 0 or +1. It is also evident from equation (14) that the range of
φ21 is from 0 to 4π. Comparing equations (14) and (5) and noting that the spacing
D2 and D3 are in the ratio 4 : 3 we get
                 3 ( φ21 + 4π M2) = 4 (φ3 + 2π m3)                       ----- (15)
       Rearranging the terms in equation (15) we get
                 (3 φ21 - 4 φ3 ) / 4π =    2 m3 - 3M2                    -----(16)
       Let k3 be the integer closest to the LHS of equation (16)
       Hence 2 m3 - 3M2 = k3                                             -----(17)
       Like in the previous case the solutions of M2 will be even if k3 is even and
will be odd if k3 is odd
       i.e       M2 = M21 + M22                                          -----(18)
       where M22 is any integer and
       M21 is equal to 0 if k3 is even and equal to 1 if k3 is odd
       Hence equation (4) can be rewritten as
                 ψ2 = (2 π 44 / λ) Sin θ = φ22 + 8π M22                  -----(19)
       where φ22 = φ21 + 4π M21


       The range of values of φ22 in equation (19) will be from 0 to 8π. Applying
the boundary conditions to M22 in equation (19) we get that the value of M22 is
always zero for any angle of arrival from –48° to +48°. Thus if the field of view
of the DF system is restricted to +/- 45°, φ22 will give the totally resolved phase
data for the spacing D2. This value of φ22 can now be used to resolve ambiguity in
φ4 as follows:

          CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                   V(3) - 6
       The ratio D2:D4 is 4:9 and hence we can write
               4 (φ4 + 2π m4) = 9(φ22 )                                 -----(20)
       From equation (20) m4 can be estimated as

               m4 = ( 9 φ22 - 4 φ4 ) / 8π                               -----(21)
       As has been done earlier the RHS of equation (21) is rounded off to the
nearest integer to get m4. The value of ψ4 in equation (6) can be obtained by
substituting the value of m4 obtained from equation (21).


       The range of ψ4 so obtained will be from 0 to 18π. Out of this range 0 to
9π is recognised as positive phase delay and 9 π to 18 π is recognised as negative
phase delay.
       Since the phase delay is directly proportional to the frequency of the signal
we will have maximum ambiguity at the highest frequency. Conversely if it is
possible to resolve ambiguity satisfactorily at the highest frequency it will be
equally possible to resolve the ambiguity at any lower frequency.


3.     Phase Error Margin :
       We have already seen that in case an error occurs in phase measurement
we may get non-integer results for certain computations in the algorithm which
ideally should have been integers. Rounding off the results to the nearest integer
does not affect the algorithm provided the errors are below a specific value. Phase
Error Margin is defined as the maximum permissible error in phase measurement
below which the algorithm described will not break down. The breakdown of the
algorithm is said to take place if rounding off gives a wrong result due to excess
phase errors while estimating the modulo integer. This will result in gross errors
in DOA computation.


       The phase error margin is calculated as follows :
       From equation (8), there will be no breakdown in the algorithm if

         CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                  V(3) - 7
               (2 δφ1 - δφ2 ) / 2π < 0.5                                ------(22)
       where δφ1 is the error in φ1 and δφ2 is the error in φ2
       i.e.    (2 δφ1 - δφ2 ) <π                                        ------(23)
       The RF front end following all the antennas in the array are identical and
so are the phase measurement and digitization units. Hence we can assume that
the error statistics for phase measurements are independent of the spacing. We
can also assume that the error statistics for all the phase measurements are
identical. Let the peak error in phase measurement in any channel be δφ
       Hence we get from equation (23)
                       3 δφ < π
       Hence the available phase error margin in the algorithm is ± 60° for the
first step in the algorithm. It can be similarly shown that the available phase error
margin for the second and final steps in the algorithm are ± 51° and ± 55°
respectively. Therefore the system phase error margin is the minimum of all the
above three i.e ± 51°. In case higher phase error margins are required the array
spacing have to be suitably synthesized.


4.     Accuracy Consideration :
       The theoretical accuracy of the DF system is estimated by differentiating
the interferometer equation. i.e.
               σ θ = σφ λ / 2 πD cos θ                         ------(24)
       Practically measured values of      σφ is around 15° RMS for the receiver
front end channel including antenna, amplifiers, phase correlators and phase
digitiser. Substituting this value in equation (23) we get a DF accuracy of 1°
RMS for an IFOV of ± 45° even at the lowest frequency i.e. 8 GHz. In general
phase interferometers give better DF accuracy at higher frequencies. The spacing
D4 had been considered for calculating the DF accuracy.


       The interferometer DF system is generally configured only for FOVs of ±
45° since the accuracy will suffer drastically beyond this angle. Four separate

         CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                  V(3) - 8
sectors each covering ± 45° are used to cover the full azimuth of 360°.              The
accuracy of the interferometer will also be affected by other parameters like
frequency measurement accuracy, phase center variations in antennas, Signal to
Noise Ratio, Elevation of the emitter etc. However these contribute much less to
the overall accuracy compared to the phase measurement errors.


5.      Sensitivity :
        The sensitivity of the DF system is defined as the minimum processible
signal for which the measured DF data satisfies the RMS accuracy specification.
The processible sensitivity of a broadband receiver is given by
               Smin = -114 + 10 log (BW) + NF                            -----(25)
        Where Smin the minimum processible signal in dBm
        -114 is the noise power per MHz bandwidth in dBm
        BW is the total bandwidth in MHz
        And NF is the noise figure in dB.
        (A noise figure of 10 dB is a practically achievable in the 8-18 GHz band).


        The above equation assumes a theoretical processing capability at 0dB
SNR at the output of the RF front end. However in practical systems there will be
a degradation of upto 3dB in the processible sensitivity. For the 8-18 GHz band
the practically achievable processible sensitivity is better than –60 dBm.


6.      Dynamic Range :
        The dynamic range of the DF system is defined as the difference in dB
between the lowest processible signal power and the highest processible signal
power. The RF front end of an broad band interferometer DF system consists of a
limiting amplifier with adequate gain such that the signals from –60 to 0dBm are
compressed to an output range of +6 to +10 dBm. This is possible since the
system has to measure only the phase of the signal which is preserved by the
limiting amplifier. Thus a system dynamic range in excess of 60dB is possible
practically.

          CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                   V(3) - 9
7.     Frequency & ADF Requirements :
       The frequency of the signal is required for computing the DOA. The
frequency measurement is carried out by the DIFM receiver. In general the
frequency accuracy required for the DF system is 0.5% of the center frequency.


       If the DF system has to cover the full azimuth of 360° four separate
sectors each covering 90 ° have to be used. In such a case a coarse DOA
information (with a coarse accuracy of better than 35 degrees peak) based on
amplitude is required to resolve the sector ambiguities which are due to poor front
to back ratios of the antennas and the high instantaneous dynamic range.


       For optimum performance the sensitivity of the DIFM receiver should be
on par with the BLI DF system. However the ADF system can have a lesser
sensitivity (up to 10 dB less) without affecting the overall performance since
sector ambiguity arises only for high power signals which may be picked up
through the backlobes and sidelobes of the antennas also.


8.     Calibration :
       Calibration of the system is not required if all the front end components
are perfectly phase matched. In such an ideal situation, when a signal is radiated
at 0°. the differential phase measured across all the channels will be 0°. In a
practical situation the RF front end of the DF system will have phase matching of
± 30° only.    This is due to the practical difficulties in manufacturing      the
broadband microwave components and the antennas. By calibrating the RF front
end components it is possible to measure the residual phase errors across the
channels. This data is used to remove the phase errors from the practically
measured data. This is done by injecting a signal of known frequency and storing


         CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                 V(3) - 10
the residual phase error data in a memory (Look Up Table). Whenever signals are
intercepted the error for that frequency is algebraically subtracted to give the
corrected phase data. It may be noted that except the antennas the phase errors in
all the other components are only frequency dependent whereas the phase error in
the antenna is dependent on both frequency and the DOA of the signal. The phase
matching in the antenna array will have a much larger impact on the DF accuracy
than the phase matching in the RF front end. The second stage of calibration is
done by radiating a signal from 0° azimuth so that the phase errors including the
error in the antenna are measured at convenient frequency intervals. Calibration
by radiation is generally done only at 0° azimuth since the antenna phase
mismatch is most likely to remain static throughout the FOV. The two levels of
calibration lookup tables (LUT) are required since the antenna being a passive
component, is not likely to be replaced. However the RF front end consisting of
active components is likely to fail and hence has to be replaced. Whenever an RF
front end component is replaced, the LUT corresponding to the injected mode
calibration data only needs to be modified.


9.     System BITE :
       The Built In Test (BITE) signal is very important tool used for checking
the health of the system. BITE is a subset of the calibration scheme. When the
system is deployed in the field, the BITE signal is used to verify randomly
whether the system gives 0° DOA for any injected frequency. In case the DF
system is configured for 360 degree coverage in azimuth, the BITE signal gives
the bore axis angle data in each sector i.e. 0, 90, 180 and 270 degrees.


10.    Processing Time :
       The total processing time in a BLI DF system is the sum total of the phase
measurement time, the ADF processing time, the frequency measurement time
and the BLI processing time including sector ambiguity resolution. The




         CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                 V(3) - 11
practically measured value for processing time is 1.2 uS from the leading edge of
the RF input pulse.


11.       Advantages :
          •    High DF accuracy possible
          •    Less prone to multipath reflections
12.       Disadvantages :
          •    Requires coarse DF data for signal processing
          •    Highly calibration dependent
          •    High density of microwave components
          •    High cost
13.       References :
      1. “Ambiguity Resolution in Interferometry,” By I Jacobs and E Ralston,
          IEEE AES , Nov 1981
      2. “Monopulse Resolution of Interferometric Ambiguities “ By J Dybdal
          IEEE AES March 1985
      3. “Specifications, Calibration and Testing of Phase Interferometers” By
          MD Horner, Microwave Journal, Feb 1988
      4. “High Accuracy Direction Finding System Using a Multi-Element
          Interferometer” By KR Sundaram, Proceedings of Naval EW and Space
          Seminar, June 1994
      5   Models Simulate Butler Matrix based DBDs”By MK Das, KR Sundaram
          and UMS Murthy, Microwaves and RF, June 1996
      6   “Modulo Conversion Method for Estimating the Direction of Arrival”,
          By KR Sundaram, Ranjan Mallik and UMS Murthy, IEEE AES, October
          2000
      7   “Processing Techniques for Broad Band DIFM Receivers”,KR Sundaram
          and S. Sudha Rani, Microwave Journal, June 2002
      8   “ Microwave Receivers for EW Applications “By JBY Tsui
      9. “Passive Direction Finding Systems” By Stephen Lipsky.


              CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                      V(3) - 12
    ------------------------ D4 -------------------
                                                      Antennas
     ---- D1 ---- ---- D2 ----- ---- D3----                        BITE input


                                                               Microwave front end
                                                               super components


   φ1             φ2              φ3             φ4            Phase correlators


                 Phase digitiser
                                                               Freq. data


            DIGITAL PROCESSOR UNIT


                                       8 bit sector DOA data


Fig 3 : BLOCK DIAGRAM OF ONE SECTOR OF BLI DF SYSTEM




             CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     V(3) - 13
                           CHAPTER – VI


HARDWARE & SOFTWARE DESIGN OF
         EW DISPLAYS

  Shri. K. SIVA KRISHNA REDDY, Sc-‘F’
                    &
    Smt. U.V.V. KRISHNAVENI, Sc-‘D’




     CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                 VI
                                CHAPTER - VI




     HARDWARE & SOFTWARE DESIGN OF EW DISPLAYS




                                   CONTENTS




1.    INTRODUCTION

2.    DISPLAY SYSTEM

3.    DESIGN CRITERIA FOR DISPLAY SUBSYSTEM

4.    CURRENT TRENDS

5.    FUTURE TRENDS

6.    CONCLUSION

7.    REFERENCES




        CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                    VI
      HARDWARE & SOFTWARE DESIGN OF EW DISPLAYS

         Shri K. Siva Krishna Reddy Sc ‘F’, Smt U.V.V.Krishnaveni, Sc ‘D’
                                DLRL, Hyderabad


1.0     Introduction :

       The proliferation of modern electronically controlled, directed and commanded
weapons have caused a rapid expansion in the field of science which is generally called
“Electronic Warfare”(EW).


       The basic concept of EW is to exploit the enemy’s Electromagnetic emissions in
all parts of the EM spectrum in order to provide intelligence on the enemy’s order of
battle, intentions and capabilities and to use countermeasures to deny effective use of
communications and weapon systems while protecting one’s own effective use of the
same spectrum. A generally accepted military principle is that victory of any future war
will go to the side that can best control the EM spectrum.[2]


       Any EW system basically consists of a Data Acquisition Subsystem, Data
Processing Subsystem, Display Subsystem and optionally Counter Measure Subsystem.


       Data Acquisition Subsystem intercepts the EM energy in the environment. It
constitutes Antennas, Front End Receivers etc.


       Data Processing Subsystem processes the intercepted EM data and extracts the
parameters from it for different emitters present in the environment. Ex: ESM Processor.


       Counter Measure Subsystem activates different Electronic Counter measure
options at the command of the EW operator. Ex                 ECM Processor and Jamming
equipment.


        Display Subsystem is the basic man-machine interface for the whole EW System.

             CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                       VI - 1
It is the one through which the operator controls and configures total EW system.


2.0     Display Subsystem:


        There has been a trend over the past decade towards more comprehensive
approach in the design of EW Display Subsystems. The most salient design characteristic
of the new Display system is that they fully integrate the perceptual requirements of the
human observer, the displayed information and system hardware and software
components.


        As EW systems are constantly growing more complex with technology both
aiding and forcing the developments of new concepts, modern sensor acts as an electronic
“Vacuum Cleaners” in a target rich environments. Also there now exists new data
communication interfaces that provides data links to collect vast quantity of data and
direct it to a central location for analysis and correlation. But success in electronic
combat neither depends solely upon the sophistication of the sensors nor upon the speed
of data processing. The ultimate success of EW system is determined by how effectively
this vast array of technology is used by the EW operator, who must understand the
information and its importance; the operator must then use this information to make
decisive judgements that will affect the course of battles. We can visualize a EW Display
Subsystem as a Window through which the operator can view the electronic environment
seen by his sensors and communication links and through it the operator provides input to
control the state and dynamics of his equipment.


        In general, an EW display System is the combination of hardware, software and
human factor design techniques that transform the electronic environment into human
understanding of what the signals represent. A block diagram of a typical Display system
is depicted in fig 1.0.


The main objectives of a Display Subsystem are:
1. To improve the human decision making capability.

            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                      VI - 2
2. To present the information in such a form that it can be readily assimilated.
2. Help to improve the EW operator’s understanding and insight of the entire
      situation.
4. Insuring that the stimuli presented can be interpreted by the operator as an
      actual battle field scenario and real world images that it is supposed to represent.


In comprehension, the main objective of a Display System is to present complex
situations in such a manner that the operator can determine the significant
relationships that exists, perform correlation on the available data, and efficiently
choose among available options[1]


 POLAR DISPLAY               SITUATIONAL             TABULAR
                             DISPLAY                 DISPLAY



PRIMARY EW
                                DISPLAY PROCESSOR
HOST
PROCESSORS                    *DISPLAY GENERATION
                              *OPERATOR COMMANDS
                              *PERIPHERAL CONTROL                               DATA BASE
                                                                                STORAGE

DATA LINK




                          OPERATOR COMMAND ENTRY
                          TERMINAL(KEYBOARD)


FIG 1.0 BLOCK DIAGRAM OF A DISPLAY SUBSYSTEM


3.0      Design Criteria for Display Subsystem :


         The following are the points to be taken into consideration while designing a
Display for an EW System proposed for any platform.




             CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                       VI - 3
3.1    Man_ Machine Interface Design :


       The MMI design includes the Graphical User Interface(GUI) design, facilities,
features and utilities to be provided and design of interfaces to the Display.


3.2    Hardware design :


       The hardware design should take care of the hardware part of the system like
processor required, storage capacity required, size, weight, environmental specification
under which the system has to operate and Electromagnetic interference and
Electromagnetic compatibility the system has to possess.


3.3    Software development Environment :


       The choice of operating system and language to be used for development of
application software for Display system depends on the hardware design of the system
and the platform on which the system operates.


3.4    Ergonomics :


       Ergonomics play a major role in design of EW Display. The choice of Display
monitor, the position of monitor in the platform should be such that, maximum
information can be presented to the operator with minimum strain.

3.1    Man_Machine Interface Design :

3.1.1 Design of Man_Machine Interface for EW Displays :

       EW applications cover a wide spectrum of functional display requirements, from
simple preprogrammed threat alert indication of RWR to complete presentation of
electronic intelligence of an ELINT system to controlling and configuring a complex EW
system either in stand alone mode of operation or in integrated EW environment. But

           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     VI - 4
whatever the system may be, the presentation of huge data intercepted by the other fast,
sophisticated subsystems in an efficient, simple, user friendly way and enable the
operator to take timely tactical and strategic decisions and provide him the flexibility to
configure and control the total EW system dynamically is of paramount importance while
designing a EW Display.


         The MMI of an RWR system can be simple with features alerting the operator to
the preprogrammed threats.


         The MMI of an ELINT system apart from presenting the signal environment
intercepted in programmed bands provide various operator friendly features to control,
configure the receiver subsystems, to do fine grain Time domain and frequency domain
analysis of selected signals, tune the receiver subsystem to the required bands or
frequencies, to program frequencies to be skipped in selected bands during tuning of
Narrow Band receivers, to define sweep rates, Dwell times and to log all the intercepted,
analyzed data.


         A EW Display MMI encompass all MMI features of RWR, ELINT systems with
many more operator oriented features, facilities and utilities which enable operator to
take timely tactical and strategic decisions.


         The design of MMI of EW Display mainly depends on 1. Capacity of the EW
system    2. Complexity of the EW system and 3. Platform on which the EW system will
be installed.


         The MMI of an EW Display can be moderately complex for an EW system
where it is required to 1) present data intercepted by a Tactical receiver with quick
reaction time and capable of intercepting and identifying hostile emitters with 100% of
POI(Probability of Interception) and 2) to control and configure Tactical receiver or
ECM subsystem or both in stand alone mode.


            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                      VI - 5
       It can be highly complex for an EW system where it is required 1) to integrate
and serve both Tactical receiver system with 100% POI and strategic receiver system
with high sensitivity and analytical capabilities to provide Electronic intelligence, 2) to
control and configure both the systems dynamically, 3) to control and configure ECM
subsystem dynamically to counter the prevalent threat scenario. The MMI design
becomes more complex when the Display system is part of complex integrated EW
system where the Display is integrated not only to ESM, ECM and ELINT systems but
also has to serve different levels of control centers which gets data from multiples
sensors. In this integrated mode either it works in slave mode to Control centers or
autonomous mode. It sends data intercepted by ESM and ELINT system to Control
Centers and routes the commands of Control centers to either ESM, ELINT or ECM
systems.


       Whatever be the complexity of the design, the MMI should also cater for
multilevel skills of the operators. Ample context sensitive help should be provided for a
novice operator and shortcuts and keyboard initiations to an experienced operator. It
should also provide different levels of controls to the operators of different hierarchies.
Controlled access to the commands controlling and configuring the EW system should be
provided with levels of security protections.


       The following are the few design guidelines that should be followed for a good
design of GUI software for EW Display Systems.


Do not bury the user with data; use presentation formats that enables the rapid
assimilation of information.
Use of windows to compartmentalize different types of information.
Use of consistent labels, standard abbreviations and predictable colors.
Consider the available geography of the display screen and use it efficiently.
Categorize activities by function and organize screen geography accordingly.
Reduce amount of information that must be memorized between actions.
Minimize the number of input actions required by the user.

           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     VI - 6
Use of simple action verbs or short verb phrases to main commands.
Permit easy reversal of most actions.
Displaying only that information which is relevant to the current context.
Produce meaningful error messages.


          In next few sections the design of a typical EW Display subsystem designed for
one of the complex EW system of DLRL following almost all the guidelines given above
in design of GUI and providing efficient, user friendly and flexible MMI is discussed.


          The first two objectives discussed in Design Guidelines require proper data
organization. This is achieved by presenting the data intercepted by Receiver system and
processed by processor subsystems in different windows on the screen in appropriate
format for easy      and unambiguous understanding of the EW operator. The data is
presented in both graphical format and textual format for easy assimilation. The POLAR
and SITUATIONAL formats present intercepted data in graphical format. The
TABULAR format presents data in textual format.


          The screen of the monitor is divided into different windows like Graphics
window, status window, Emitter Report window, Detailed parameter window, Latest
emitter window and menu window.


          In graphical window, the data is presented in graphic symbol format. In status
window the operational status of system like time, date, direction of movement of aircraft
and data interface link status are presented. In Emitter Report window the data is
presented in text format with vertical and horizontal scrolling facility. The Latest Emitter
window presents the latest emitter intercepted by the system. The menu window provides
the keyboard interface to the operator.


3.1.1.1          Polar Format :


          Polar format (Fig. 2.0) presents the tactical field scenario of the outside world.

             CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                       VI - 7
For complex and high capacity EW systems the Graphics window is divided into three
zones. In each zone a unique category of signals are presented. Inner most zone presents
the hostile emitters, Central zone presents the unidentified or unknown emitters and outer
zone presents the friendly emitters. A line is drawn from center to the outer circle
indicating the orientation of platform with respect to true North.


       In this format, each intercepted emitter is indicated with a graphic symbol which
symbolizes the type of platform on which emitter is operating and emitter number. The
position of symbol is calculated using the direction of arrival(DOA) and amplitude
parameters of the emitter. The DOA gives the direction in which the emitter can be
placed on the screen and amplitude gives the radial distance from the center at which the
emitter can be drawn in the given direction.



       This   presentation format facilitates the easy identification of the active and
probable threats, their positions and signals strength. This information helps operator to
take quick and appropriate tactical decisions.


                            STATUS WINDOW


       GRAPHICS
       WINDOW

                                                 EMITTER REPORT
                                                 WINDOW


                                                 LATEST EMITTER
                                                 WINDOW


                                                 MENU WINDOW

                  DETAIL PARAMETER WINDOW


               Fig 2.0 POLAR PRESENTATION FORMAT



           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     VI - 8
      3.1.1.2      Simulational Format:


                This format uses Cartesian coordinate system, which provides a frame of
      reference for the display symbology. Looking at the standard display symbols
      representing various emitters helps the operator to quickly ascertain the nature of threats
      and their respective operational platform. It is an Direction of Arrival of emitter (X-
      axis)Vs Frequency of the emitter(Y-axis) plot. The format is shown in Fig 3.0.



                This format gives information about spectral distribution of intercepted emitters in
      bearing range.



                                                    STATUS WINDOW

                        18


                        10
                                 GRAPHIC WINDOW
                                                                          EMITTER REPORT
                                                                               WINDOW
                        8
                FREQ


                            4                                              LATEST EMITTER

                                                                          MENU WINDOW
                        1
                                180   270           0      90       180
                                               DF

                                            DETAIL PARAMETER WINDOW


                       Fig 3.0 SITUATIONAL PRESENTATION FORMAT


3.1.1.3 Tabular Format :

                It is complete alphanumeric representation of the acquired data. It is a textual
      mode presentation of the data with vertical and horizontal scrolling facility to view
      different pages of emitter data. The Tabular format presentation is shown in Fig 4.0.

                   CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                             VI - 9
                                 STATUS WINDOW



                            EMITTER REPORT WINDOW



                                 LATEST EMITTER

              DETAIL PARAMETER WINDOW
                                                      MENU WINDOW



                 FIG 4.0 TABULAR PRESENTATION FORMAT


       Another major factor to be taken into consideration is that the presentation of data
should ensure that it represents actual threat scenario. Color has proven of great benefit in
improving the readability of complex Display formats. Judicious color attributes and
Display symbols are chosen to draw immediate attention of the operator. A standard
practice is to represent hostile or threat emitters in RED, unidentified or probable threats
in YELLOW and all friendly or known emitters either in GREEN or BLUE. A different
color attribute is chosen to distinguish passive emitter from active emitter in all
categories of signals.


       Whenever a signal is detected by an ESM system, it is compared with pre-
confiigured warner library and Target library for identification of the signal. If there is a
match in Warner library, the operator is alerted for that corresponding threat by flashing
it with RED color. If there is no match in both the libraries, the emitter is considered as
unknown or unidentified emitter and presented with unique symbol in YELLOW.


       The term Display is used not only in the context of video but both audio and
speech information transfer devices are also included. An audio alarm can be provided to
inform the operator about the type of emitter. This additional control input will have a
significant effect on reducing the complexity of the operator workload. At the same time
it should be borne in mind that too many audio alarms may confuse the operator and have
to be avoided.
           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     VI - 10
        The above formats are basic modes of presentation of data. There are number of
features, facilities and utilities provided to make MMI more user friendly, efficient and
intelligent. The features and facilities to be provided depends on the capability and
complexity of the proposed EW system and the hardware design of the Display.


        For example, for a EW system designed for heli-borne application where the
capacity of system is not more than few tens of emitters and a stand alone system with no
ECM capabilities, the MMI was simple designed around 5”X4” monitor and micro
controller as Processor. There are only two modes of presentation of data Tactical and
Tabular. The features and facilities provided are also limited as the volume of data to be
presented is less.


        For a typical EW Display, GUI is provided which enables the operator to view the
intercepted data in different formats and windows, to apply various decluttering
conditions like filter, zoom, blank etc to view the huge data correctly and unambiguously.
These features are required as the limited real estate of Display screen can be optimally
used to present voluminous intercepted data. Horizontal and vertical scrolling facility is
provided to all textual windows to scan through the pages of emitter data intercepted.
Utilities like Store and Replay, Command and Error logs, print, backup facilitate operator
to perform post mission analysis correctly and efficiently and enable him to take strategic
decisions.


        The ‘Store‘ facility stores the mission-intercepted data and ‘Replay’ plays back
the stored data for post mission analysis. Facility is also provided to apply different
filters during storing process like 1) category of emitters, 2) in specific time period, 3) in
selected frequency range or sector etc. The Replay utility can playback the whole
mission or operator chosen time period mission data. It can be made fast forward or
rewind at operator’s command. The stored data is organized in the form of standard
database like MS Access and statistical analysis can be performed using MS Excel.


             CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                       VI - 11
        The command and Error logs are used to analyze all the commands executed by
the operator during mission and failures occurred in communication links between
different subsystems during mission respectively.


        The library management is another important feature in EW Displays. There are
two library files Warner and Target. During the start of mission the mission specific
warner libraries are created or modified and loaded into different subsystems like ESM
Processor and Analysis Receivers for performing search in the libraries for a match,
when a emitter is intercepted. The library management feature has flexibility for on-line
creation, modification of warner libraries, viewing and downloading of either of the
libraries. It also allows appending of the intercepted emitter data into warner library. A
protected and controlled access to the libraries features like appending, creation and
modification is provided keeping in view the highly delicate and sensitive nature of the
data.


        All the above discussed MMI features are Display oriented commands which are
local to Display and are executed by the operator to view the intercepted data in different
perceptions. But Display is not only a window through which the operator can see the
electronic environment but it is also a single point from which he can control and
configure the total system as per the mission requirements.


        GUI is provided in Display to control and configure ELINT, ESM, ECM systems
dynamically. The responses from these subsystems are collected and presented to the
operator. The health and operational status of subsystems, link status, its own operational
status are presented in a dedicated window to the operator. MMI features like ‘BITE’,
‘Health query’ facilitates the operator to monitor the health of the EW system and
perform periodic health checks apart from system provided automatic health checks.


        System Commands like INIT to any subsystem will carry out a software reset of
the corresponding subsystem. Commands are provided to remove passive emitters from
the system and store in memory. The operator is provided with facility to rebuild,

           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     VI - 12
reevaluate, reprocess the emitters intercepted, through Display console. The operator can
program the scan bands, frequencies, to which the Analysis receiver system has to be
tuned, from Display. He can perform fine grain analysis of wide open receiver system
intercepted emitter by passing the emitter details to Analysis receiver system. This option
will be of great use in case of complex emitters. The pulse pictures of intercepted signal
constructed from the analyzed parameters and presented to the operator for his analysis.


       GUI is provided to configure the receiver systems, processor systems interfaced to
the Display from the keyboard of the operator. Some of the configuration related
commands provided are commands to set the amplitude thresholds, scan bands, forbidden
bands and frequencies of receiver systems, commands to set the threat or confidence level
ordering of ESM and ECM processor for library matching, commands to set the
confidence level thresholds used during comparison of intercepted track parameters with
library entries, commands to program the tolerance windows for parameters dynamically
etc. As the configuration altering commands changes the working pattern of the total EW
system, controlled access to the commands is provided. The access to these commands is
provided to high level authorized users only. Other commands can be initiated by any
authorized user.


       The ECM system can be controlled and configured by the operator through
Display subsystem. The commands like ‘Jam Target’, ’Track Emitter’, ‘Stop Tracking’,’
Stop jamming’, ‘modify JPRO Number’, ‘Track & Jam’ enable the operator to track an
emitter and execute suitable ECM against the identified threat. The operator can
configure, control and perform regular health checks of the system through suitable MMI
provided.


       Hence the design of MMI for any EW Display depends on the complexity of the
system i.e. the number of subsystems it has to serve, capability of the EW system and the
hardware of the Display subsystem which in turn is decided by the platform on which the
EW system is proposed to be installed.


            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                      VI - 13
3.1.2   Interface Design :


        The Display subsystem of an EW system is interfaced to number of subsystems,
number being decided by the complexity and capability of the EW system. In a simple
EW system with no ECM capabilities, the Display is interfaced to a wide-open receiver
system only. Its main functionality is to present system intercepted data in different forms
and views control and configure the total EW system. In a system with integrated ESM
and ECM capabilities, the Display is interfaced to ESM Processor, ECM Processor and
controls and configures both the processor subsystems. In a complex integrated EW
system with multi level control centers, the interfaces to Display are many and
heterogeneous depending on the number of levels the Display should serve.


        The design of interface between the subsystems and Display has to take three
important points into consideration. 1) Data rate to meet the required response time 2)
Data integrity for un-corrupted data transfer. 3) The distance between the data source and
Display. The link should provide reliable data transfer by proper protocol, error check
and handshake. Keeping in view the above important requirements, proper hardware has
to be chosen supported by a good and robust software design that does proper handshakes
and efficient error checks.


        A simple asynchronous serial interface can be used if the distance is less then 20
feet, data transfer rate requirement is low and total data to be transferred is low from
Processor subsystem to Display.


        Synchronous serial interface is chosen when the distance between Display and
other processor subsystems is high, data transfer rates are high, large volumes of data has
to be transferred and data integrity is also of primary concern. The important serial
interfaces generally used to interface Data Processing System to Display are 1. MIL-
1553B Bus 2. High Level Data Link Controller protocol(HDLC) and 3. Ethernet. Mil-
1553B bus interface is mostly used in air-borne EW applications.

           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     VI - 14
       Parallel interface can also be used but its inherent limitation of signal skewing for
long distance communication cause main constraint. Mostly printer is interfaced to the
parallel port to produce the hard copy of the stored data, Error and Command logs, pulse
pictures etc for post mission analysis activity.



3.2    Hardware Design :


       Design of Hardware for Display depends on the platform on which it is proposed
to be installed and the data handling capability of the EW system. One of the most critical
factors influencing MMI design is the hardware design.


       Each of the platforms has it’s own constraints and limitations. The hardware
design has to consider all the constraints and limitations and optimize the design without
making any compromise on the functional requirements. For example, the hardware
design of EW Display for Air-borne segment puts constraints on weight and size of the
Display. It also constraints the usage of any moving part in the design. Hence hardware
design of Display systems for air-borne applications aims at reducing the size and weight
of the total system without compromising the operational performance. PC-104 form
factor motherboards, PC-104 form factor compatible peripheral cards, flash memories are
chosen to reduce the size and weight of the Display. The solid state flash disks are
preferred against conventional hard disks for their small form factor and they are free
from moving parts like read/write head of conventional hard disks. At the same time this
component becomes a bottleneck in design of MMI as the capacities available are 1 to 2
GB maximum. Customized or compact rugged keyboards and LCD monitors are chosen
to meet the size and weight constraint.


       The designed hardware has to qualify environmental tests according to JSS
55555 or Mil-810 D standards depending on the platform for which the EW system is
designed. If it is a vehicle mounted or ship-borne or sub-marine based system, the

           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     VI - 15
Display hardware has to qualify JSS 55555 standards. If it is an air-borne system,
hardware has to qualify Mil-810 D standards. The hardware has to be EMI/EMC
compatible as per Mil-461C standards.


       While designing a Display system either all mil-grade components can be chosen
and integrated or commercially available off-the-shelf items can be chosen and
ruggedized to meet environmental specifications, first option being preferable.


3.3    Software Development Environment :

       The operating system on which the S/W of Display subsystem to be developed
and language to be used depends on following factors 1. The complexity of the system
2.The memory capacity of flash disk in the system. Generally windows OS is used to
develop the S/W as the Display as seen in earlier sections is GUI oriented with intensive
user interface. Operating systems like Windows, Solaris with ‘C’ or ‘C++’ or ‘VC++’ as
developing language are chosen for development of S/W of Display. To have real time
extensions and for OS to occupy less space in secondary memory Embedded windows
with Real Time Extensions can be a good choice.


       In whole of the EW system, the Display is highly software intensive subsystem.
To develop bug free, reliable and quality software CASE tools like Rational Rose suite
are used extensively at every stage of S/W development lifecycle. A combination of
Water fall, Prototype and Incremental          process models is followed as lifecycle
methodology. Different levels of testing like unit testing, interface testing, system
testing and Acceptance testing are performed all through the development phase to
produce high quality S/W.




           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     VI - 16
3.4    Ergonomics :


       Ergonomics play a major role in Design of MMI as the operator is already
dumped with lot of Displays around him and EW Display being one of them. For this the
placement of monitor, the viewing angle of the monitor, the brightness and contrast of the
monitor and colors selected to present data should be such that less strain is put on the
operator. The monitors with considerably wide viewing angle and good and controllable
brightness and contrast should be selected. The most popular Display technologies
currently are CRT and LCD. Each has it’s own advantages and disadvantages. The
comparison table for two technologies is as given in Table 1.


         Technology                      Advantages                       Disadvantages
                                  Mature Technology                   Large Volume
         Cathode Ray              High Resolution                     High           Power
             Tube                 High Contrast                       consumption
                                  High brightness                     High voltage power
                                  Color capability                    supply




        Liquid Crystal            Low                Power            Temperature
           Display                consumption                         dependence
                                  High Resolution                     Slow switching speed
                                  Driven      from     low            Narrow viewing angles
                                  voltage
                                  High             ambient
                                  viewability


                    Table 1. Display Technologies Comparison Table




           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     VI - 17
       LCD monitors are picking up fast for there are light weight and low power
consumption and today LCD monitors with viewing angles comparable to CRT are
available with attractive price tags.

4.0    Current Trends :

4.1     Map Displays:


       The Moving Map add-on to the ESM System, display to the operator the platform
path and the location of targets detected by the ESM System on a background layer of
geographical maps which are either moving or fixed. Thus the man-machine interface of
the EW system is significantly improved.


       Map Displays facilitates tasks for tactical command, control, communications and
intelligence designed for military, paramilitary and civil operations. It provides
interactive high data storage interface to EW Display. Data from variety of sources e.g.
geographical terrain, GPS are correlated while minimizing the time between data
acquisition, interpretation and reporting. Data is analyzed using techniques like spatial
operations, visibility analysis to build up tactical database, which aids in estimating
enemy’s order of battle in the actual scenario.


       The system contains geographical data that is required for the successful
operation of the mission. The database consists of the following information.


•      Geographical Maps of the area relevant to the mission.
•      Tactical data relevant to the mission, areas and threats.
•      Platform path data


       The system displays the electromagnetic receptions of the ESM system on the
geographical maps stored in the database. The platform location is determined from the
GPS.


           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     VI - 18
The system display modes are as follows:


1.     Moving Map Mode: In this mode, the map moves with the platform so that the
platform is always located in the center of the map. The threats are displayed relative to
the location of the platform.
2.     Fixed Map Mode: In this mode, the map remains fixed on the Display while the
trajectory of the platform is drawn on the map in real time, based on GPS data.
3.     Textual Display: The system is able to display textual information regarding
threats, mission, navigation data etc.



4.2    Integrated Displays:


       With recent technological advancements in Graphic Engines, an EW Display can
be interfaced to multiple sensors like Radar, Sonar, ESM and display the intercepted data
from these sensors on screen using the PIP(Picture in Picture) feature. This is a highly
preferable Display as from single window, the multiple sensors can be monitored and
controlled. Monitoring the data intercepted by the different sensors boosts the confidence
levels of the operator on threats and enable him to initiate appropriate action quickly.

5.0    Future Trends :
        The availability of sophisticated, fast and smart graphic processors has paved
path to 3D Graphics, virtual reality, also leaving an opening for more research in
enhancing the Display capabilities and making it intelligent and user friendly.

6.0    Conclusion :


       With the availability of fast processors and sophisticated graphic cards the design
of EW Display is becoming a creative, innovative and efficient. The Display subsystem
in EW field has evolved from a simple dumb Display system presenting textual data and
driven by a host processor to threat warning unit to a smart system controlling and
configuring total EW system. Current EW Displays presents intercepted and processed

           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     VI - 19
emitter data in different views and forms but leaves the understanding and decision
making responsibilities to the operator. It only makes the inference of the electronic
environment outside. But with utilization of principles of Artificial intelligence, neural
networks the current smart EW Display can evolve into a intelligent Display which can
guide, suggest alternatives to the operator and take automatic timely tactical and strategic
decisions off-loading the responsibility of the operator in future.


       The fast and ever changing technologies and requirements put a constraint on the
designer to make his design adaptable, maintainable, modifiable and flexible. The design
of MMI which is adaptable to fast changing user requirements, flexible, quick, efficient
and intelligent and design of Hardware adaptable to the latest hardware innovations and
perfect and efficient amalgamation of the designed hardware with the designed MMI
throws a real challenge to the designer of EW Displays.


7.0    REFERENCES:


1.     Donald R.Roch, EW Information Display Systems, The International
CounterMeasures Handbook, 10th Edition,1985., pp 354-366.


2.     Schelehr, D.C. Introduction to Electronic Warfare. Dedham,Artech. House, 1986.


3.     LFE Coombs, The Aircraft Cockpit, Ist Edition 1990; PP 220-229




           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     VI - 20
                CHAPTER – VII


ELECTRONIC SUPPORT MEASURE
         (ESM) PROCESSORS


  Smt. V. SOBHA SHANKAR, Sc-‘E’




   CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                               VII
                                 CHAPTER - VII




                              ESM PROCESSORS




                                    CONTENTS




1.   INTRODUCTION

2.   FUNCTIONS

3.   DESIGN CONSIDERATIONS

4.   DE-INTERLEAVING OF RADAR PULSES

5.   TOA BASED DE-INTERLEAVING

6.   CONCLUSIONS

7.   REFERENCES




         CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     VII
           ELECTRONIC SUPPORT MEASURE (ESM) PROCESSORS

                              Smt V. Sobha Shankar, Sc-E
                                  DLRL, Hyderabad

1.0    Introduction


       The aim of the Electronic warfare systems is to optimise our usage of
electromagnetic spectrum at the same time minimise enemies usage of EM spectrum. An
EW system consists of Electronic Support, Electronic Attack and Electronic Protection
systems. Electronic support systems intercept, analyse and gives information of the usage
of EM spectrum. Electronic Attack systems provide mechanisms to avoid enemy usage of
EM spectrum. Electronic Protection systems help our usage of EM spectrum in spite of
enemies Electronic Attack systems.

       Electronic support systems consists of Wide-open receivers, Super-heterodyne
receivers to intercept the emissions present in the environment, and estimate the pulse
parameters of the emission. The digitized pulse data is given to the ESM Processor
subsystem to extract intelligence from the huge amount of pulse data. The main aim of
the ESM Processor is to extract the correct radar parameters from the pool of pulse data
in less than 0.5 sec time for the Electronic Attack system to react on this output.

       ESM Processor has to extract complete picture of the radars present in the
environment given their composite effect in a noisy environment. De-interleaving is used
to sort out the data corresponding to each emitter.

       The ESM Processor subsystem de-interleaves interleaved pulse descriptors from
Wide-open Receiver systems. It extracts secondary parameters - Pulse repetition
frequency, Scan type and antenna scan period for each emitter. Further parameters of the
intercepted emitters are compared with system library to determine the identity of the
emitter. The ESM Processor merges data from both Wide-open Receiver and Analysis
Receiver and generates integrated data file, which is transferred to the EW Display
system. The ESM Processor can be configured from EW Display.


               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         VII-1
2.0          Functions

             In general the ESM Processor is required to perform the following functions


•     Accept and process upto 500,000 interleaved pulse descriptors per second from the
      Receiver system
•     Deinterleave interleaved pulse descriptors from upto 128 emitters
•     Determine the a) Signal type - Pulsed, CW, Chirp, Frequency Agile, Prf Agile b)
      PRF, c) Scan type, d) Antenna Scan Period
•     Determine Identity of the emitter by comparing with Library parameters
•     Assemble parameters a) Frequency, b) Pulse width, c) Direction of emitter, d)
      Amplitude, e) Signal type, f) PRF, g) Scan type and Antenna scan period, h) identity
      for each emitter and prepare an emitter file and monitor their activity.
•     Merge emitter data received from Analysis Receiver and generate integrated emitter
      file
•     Send the parameters of high priority threat signals automatically to ECM systems for
      initiating Countermeasures
•     Transfer integrated emitter file to EW Display systems and process Commands
      received from Display system
•     Perform BITE checks


3.0          Design Considerations


             The ESM Processor is required to be designed to provide flexible system, which
can be upgraded to meet changing requirements.


             The ESM processor is required to track upto 128 emitters simultaneously under
pulse densities of the order of 500,000 pulses per second. In such dense electromagnetic
environment, the large number of independent emitters will cause ESM system to receive
a seemingly random pulse train consisting of interleaved pulse trains. In order to identify
individual emitters, their pulse trains must be deinterleaved. The de-interleaving process

                    CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                              VII-2
is essentially the process of determining whether the newly intercepted radar pulse
belongs to one of the already processed emitter or to a new emitter. Ideally, there will be
a one-to-one correspondence between each pulse train identified by the de-interleaving
process and the emitters in the environment. In practice, pulses from several emitters may
be combined into one chain, and pulses from one emitter may be split into several chains.
The de-interleaving process will take into account the expected characteristics of the
radar environment, the known characteristics of the measurement system and all
foreseeable sources of data corruption.


       The de-interleaving process is followed by PRI analysis for each pulse chain. This
PRI analysis will determine if more than one emitter is mapped to the chain. After PRI
analysis, Scan Analysis and Identity search are performed for each emitter pulse chain.


       Some of the design considerations are


•   Architecture of ESM Processor
•   Pulse density
•   Automatic Processing
•   PRF Analysis
•   Memory size
•   Scan Analysis
•   Identity search
•   Processing time
•   Storage of data
•   Diagnosability
•   Software issues
•   Design assumptions




               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         VII-3
3.1    Architecture of ESM Processor

       There have been, in general, three approaches, in realising ESM Processor
subsystems. They are


i)      Single Processor based system with adequate buffer for storing pulse data
ii)     Multiple processor based system
iii)    Multiple Processor based system supported by fast Front-end hardware


       The first approach is generally used in Radar Warning Receiver (RWR) systems,
particularly in airborne RWR systems. This approach results in a very compact system
and all functions are implemented in software. Typically RWR systems are programmed
to look for specific emitters only and generally do not process other signals in the
environment. However the system performance is likely to suffer under high pulse
densities and identification of fleeting emitters can not be guaranteed.


       The second approach is widely followed and all functions are implemented in
software. In these systems incoming Receiver data is either divided into time slices or
frequency intervals and fed to one of the processors for analysis. This approach leads to a
flexible system but it is not guaranteed to identify a new emitter in deterministic period.


       In the last approach, Front-end hardware performs primary de-interleaving and all
other functions are implemented in software. Typically Front-end hardware is
implemented to ensure de-interleaving operation is completed before Receiver can
process next pulse. DLRL has been using this approach.



3.2    Pulse Density


       As mentioned above, all functions of ESM Processor except de-interleaving are
implemented in software. De-interleaving is implemented in Front-end Hardware using a
set of Application Specific Integrated Circuit (ASIC). The ESM Processor is required to

               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         VII-4
process upto 500,000 pulses per second under peak load conditions and the highest PRF
required to be handled is 500,000 Hz. Front-end Hardware performs de-interleaving by
comparing Pulse parameters Direction-of-Arrival, Frequency and Pulse Width with
corresponding parameters of already detected emitters in parallel. This process of
comparing parameters is completed under 2 microseconds, which is less than the
Receiver shadow time. Computer modeling has shown that even in a simple scenario with
about 100 circularly scanning radars of 3 degrees beamwidth and the same PRF, there is
around 1% chance of receiving more than three times the average pulse rate in a 50 milli
second sample time. Although currently Receiver system do take around 2 micro seconds
to complete pulse parameter measurements, it is desirable that ESM Processor be capable
of processing more than 500,000 pulses per second for a short duration. A FIFO buffer is
provided to smoothen pulse processing during such peak load conditions. This FIFO
buffer will also provide facility for analysing pulse by pulse data variations at any
selected time.



3.3    Automatic Processing


       The ESM Processor is required to track 128 emitters. However due to factors such
as multi path effect, pulse overlapping etc, it is likely that spurious tracks are built. Also
when Frequency agile emitters are active more than one pulse chain will be created. It has
been seen from the literature that electronically scanning phased array radars may use 32
different spot frequencies. Assuming that 10% of the active emitters could be complex,
the ESM Processor should be capable of processing 512 pulse trains simultaneously. This
will enable the ESM Processor to track 128 emitters allowing for 10% spurious tracks.



3.4    PRF Analysis


      The ESM Processor uses Frequency and Direction-of-Arrival information.
Currently DLRL is designing ESM Systems where Frequency is measured to an accuracy
of 5 MHz (RMS) and Direction to 3 degrees (RMS) accuracy. Even with these parameter

                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                           VII-5
measurement accuracies, it is possible that more than one emitter is mapped into single
pulse chain by the de-interleaving process. Hence further de-interleaving is required to be
performed.    This type many-to-one mapping is likely for the radars with similar
parameters. Secondary de-interleaving is performed during PRF analysis by using Time-
of- Arrival (TOA) based de-interleaving. TOA based de-interleaving is performed by
analysing histograms of differences in time of arrivals of successive pulses. This analysis
should take into account missing pulses, corrupted parameters.

      PRF analysis should be able to handle Constant, Jittered, Dwell and Switch and
Staggered types of PRF. All pulse trains with TOA differences of successive pulses less
than 0.75% of the average PRI maybe classified as constant. PRF analysis should cater
for a maximum of 30% jitter and upto 16 levels of Stagger.



3.5    Memory Size


      The ESM Processor SW should have access to pulse by pulse data of each
deinterleaved chain. Memory sizes should be adequate for performing PRF and Scan
Analysis. In case of staggered type of emitter, upto 16 levels of stagger are required to be
processed. Assuming that data patterns are required to repeat 8 times for consistency, for
identifying 16 level stagger at least 128 pulses have to be stored. Also for scan analysis,
memory should be adequate to store data of at least 10 scan periods. In most probable
scenarios, the environment is made up of large number of scanning emitters with PRF in
the range of 250 Hz to 4 KHz and a few high PRF emitters. Number of pulses received
from each scanning emitter is limited by factors such as beam width, dwell times etc. It is
not unreasonable to assume that less than 50 pulses would be received for each scan.
However number of pulses received even from a scanning emitter would be very high
even for E and F band emitters. In such cases the ESM Processor may restrict collection
of pulse data to pulses around peak amplitude (say within 10 dB). It is seen from these
considerations pulse depth of 2048 would cater for both PRF and Scan Analysis.




               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         VII-6
3.6     Scan Analysis


      The ESM Processor should be able to identify tracking and search types of scan. It
should perform scan analysis automatically and for all tracks in parallel. The emitter data
could be reported to EW Display even in the absence of Scan information. It should be
able to distinguish between tracking (Locked-on emitter) and close by search radars.
Scan analysis is performed by analysing variation of pulse amplitude as a function of
time. If more than 512 pulses of any emitter have been received with constant amplitude,
it is investigated for Locked-on scan type.


      If amplitude is varying, peak amplitudes are determined and tracked for each chain.
The intervals between peak amplitudes are investigated for identifying scan types and
determining antenna scan periods. If peaks in amplitude occur at regular intervals, and
these intervals are more than 1 second and less than 20 seconds, Scan type is deemed to
be Circular. If the intervals are less than 1 second but greater than 100 milli seconds scan
type is identified as sector scan. If interval is more than 30 milli seconds but less than 100
milli seconds scan type is declared as Track while Scan. If the interval is less than 30
milli seconds scan type is deemed as Locked-on.

      If the interval is not constant, pattern is investigated for bi-directional sector scan.
Getting a bunch of 8 to 32 pulses with same amplitude but occurring at random intervals
indicates electronic scan. Electronic Scan information maybe corroborated by Library
data if available.



3.7     Identity Search


      Identity search is performed against 64 warned modes and 3000 target library
modes. The ESM Processor will perform identity search every time a new emitter is
identified by comparing with 64 warned modes and 300 priority modes of target library
modes. Identity search is also performed whenever an emitter changes its mode of
operation or when data is available from analysis receiver. Remaining modes of Library
                CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                          VII-7
are searched on a specific command from Display. Identity search is performed by
comparing Frequency, Pulse/CW, PRF, Scan Type and Antenna Scan period. The Library
can be down loaded into ESM Processor. The Library of radar modes will contain code
for employing suitable Jamming technique for neutralising / reducing radar effectiveness.
Parameters of new emitters meeting predefined conditions will be sent to ECM
automatically for initiation of countermeasures.


3.8      Processing Time


      ESM Processor is required to report fleeting emitters ie a radar which transmits for
a very short time. A track is built and monitored if at least 3 pulses are received within 65
milli seconds (corresponding to a PRF of 50 Hz), All the new pulse chains which do not
meet this requirement are considered to be spurious and are deleted from the system.
Ideally it is preferable to collect data of as many pulses as possible before reporting a
track to Display. Larger data sample will lead to better estimation of parameters,
particularly Direction of arrival and pulse width. However many radars have been
reported which give out as few as 8 pulses in their main lobes. The ESM Processor
should cater for such narrow beam radars and also cater for pulse drop outs. It should be
able to report a track with as few as 5 pulses in tactically important I, J and MMW bands.
Ideally the criteria for building and updating emitters should be configurable parameters.
Typically ESM processor will use 16 pulses before reporting tracks in D, E, F, G, & H
bands.

      Let us estimate time taken for reporting search radar in low bands, ESM Processor
will need around 320 milli seconds to collect data of 16 pulses (assuming PRF is 50 Hz
for lowest case). Providing 50 milli seconds for PRF analysis and 50 milli seconds for
identity search     it is seen that ESM Processor will be in a position to report new emitter
within 420 milli seconds. This will enable the system to present new emitter report to
Operator within 2 seconds taking care of ESM processor to Display communication and
Display management.



                  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                            VII-8
       Let us also consider the case of a tracking emitter. For a tracking radar it is quite
reasonable to assume PRF will be more than 2000 Hz. It takes less than 5 milli seconds to
collect the data and further 100 milli seconds to complete processing and identification.
Hence the process of identification for tracking radars will take less 200 milli seconds.
Lock-on identification will come after receiving 512 pulses ie around 260 milli seconds.
Hence for a high priority emitter such as warned and Locked-on emitter, the ESM
processor will be in a position to initiate action for ECM under 300 milli seconds from
time of the receipt of the first pulse.


3.9     Storage of data


       ESM Processor and EW Display together should be able to store emitter data and
log all operator commands. If the system is working on surface or ground platforms, the
EW Display should be able to record data in a secondary storage medium such as hard
disk orCD Writer. If the system is exploited continuously for over 2 to 3 weeks, it is
estimated that 2GB hard disk space will be required to record the data. This data can be
used for Post Mission Analysis by the Operator and for verifying that system performed
as per design by the maintainer / designer. For airborne systems, where EW Display may
not be equipped with secondary storage, the ESM Processor will need to provide the
facility of recording emitter data and operator commands in local Flash memory. The
ESM Processors of DLRL have been provided with such a facility.


3.10    Diagnosability


       ESM Processor will need to provide both on-line and off-line diagnostics. The
ESM processor will have facility for extensive BITE facility including Power On Self
Test (POST). All modules will be designed to be tested, localising the faults to a logic
function of a PCB module. The ESM Processor will provide support for RFBITE tests.




                CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                          VII-9
3.11       Software issues



       The application software for ESM Processor should be in an industry standard high
level language. DLRL has been developing software using a mix of C and ASM for
developing application software. Assembly language will be used for performance
optimisation of critical sections of the program and interrupt service handlers.



3.12       Design Assumptions


       Following assumptions are made about the electromagnetic environment.


       •     It is assumed that electro magnetic environment is quasi static(ie there are very
             few simultaneous changes) and most of the signals received by the system will
             be from already known emitters, and only few will belong to new emitters.
       •     There is no need to process every pulse of a known emitter continuously. It is
             sufficient to sample it from time to time. Hence, parameters of a search radar
             are updated during main lobe reception, and a tracking radar by sampling every
             one second.
       •     For computing an emitter's antenna scan period, it is not necessary to deal with
             all of the emitter's pulses. It is only necessary to deal with its main lobe.
       •     Priority is given to processing data of a new emitter.
       •     If an emitter has not been updated for either 20 seconds or twice the antenna
             scan period, it will be treated as passive.
       •     It is assumed that Scan type is not required for MMW band


       Following types of emitters are considered as threats


       •     All emitters whose parameters have matched with warned library data
       •     All pulsed unknown emitters with Locked-on scan type provided Frequency is
             above a programmed limit

                  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                            VII-10
       •     All emitters in MMW band.
       •     Unknown CW emitters falling outside known frequency bands of Cellular
             signals, Pager signals and Radar Altimeter.



4.0        De-interleaving of Radar pulses


           De-interleaving of received radar pulses based on the analysis of various
parameters such as time of arrival, angle of arrival, pulse amplitude, pulse width and
carrier frequency. Among all these parameters angle of arrival is the one which can’t be
altered by the radar itself. Angle of arrival can be used as a primary parameter in de-
interleaving the interleaved radar signals.

           Radars are either continuous wave (CW) type or pulsed type. Under pulsed it
could be fixed PRI or PRI agile, fixed frequency or Frequency agile. Under PRI agile it
could be either staggered type or Jitter type PRI. Under Frequency agile it could be scan
to scan frequency agile or pulse to pulse frequency agile or Chirp type. From the
operational point of view radars could be classified into tracking radars, scanning radars
and track while scanning type radars.

           Under these different radars operating simultaneously in the environment, the
ESM processor receives a composite picture of the radar pulses. The received pulses are
first separated into sets of similar pulses based on their angle of arrival. Continuous wave
type radar is indicated by the receiver hence they don’t require any further processing
except correct determination of their amplitude and angle of arrival. Pulsed type radar
pulses have to be analysed for there PRI, agility and scanning characteristics for radar
identification and to enable jamming.


 5.0       TOA based de-interleaving


           PRI (pulse repetition interval) determination is necessary to separate pulses of a
given radar from a background of pulses for radar identification. The parameters

                  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                            VII-11
associated with the extracted pulses can be analysed to determine parameter variation
with time. E.g. scan pattern analysis. Time of arrival (TOA) de-interleaving can be used
for determining PRI.

       Each radar can be characterized by a pattern of pulse intervals that repeats from a
given start time (phase). In the simple case of stable PRI signal, only one PRI value is
repeated. The range of PRI values is several µs to several ms . The PRI can be staggered,
where several PRIs form a frame which is repeated, or jittered, where the PRI deviates
around a nominal value.

       The received pulses are first separated into sets of similar pulses based on
direction of arrival (DOA) before attempting TOA de-interleaving. However, the
environment contains similar radar types with agility which may not be resolved by
grouping instantaneous parameters and thus TOA analysis is necessary. The TOA de-
interleaving algorithm is used to extract all radar patterns that can be reliably recognized.
Arithmetic computations are performed on a sample of TOA words by the de-interleaving
algorithm. A sufficiently large sample must be taken such that the signal patterns are
apparent.

       Thus the sample consists of a sequence of events. We define the sample interval k
as the TOA measurement resolution and the sample length as N sampling intervals (SI).
Thus k is a scaling factor between the stored integer values and the actual time in
seconds. The TOA of each pulse can be represented by a delta function, i.e. value 1 at the
appropriate sampling interval; otherwise the sampling intervals are assigned value 0.
Each TOA is measured as an integral multiple of the sampling interval. Thus, the ith
stable PRI sequence αi with a PRI of m i SI, a start time of q i SI and n i pulses in the
sample (the function ‘int’ is greater integer less than the operand) can be written as
             N
       αi = ∑ fi (rk )
            r =0

                                                N - qi 
             1 where r = ami + qi, 0 ≤ a ≤ int          = ni
        fi =                                    mi 
             0 otherwise
             

                   CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                             VII-12
       and r , a , m i , q i and n i are positive integers.

       The sample of pulses to be sorted consists of several interleaved signals. Where
events coincide, only one event is indicated, and thus the resultant sample P is the logical
OR of the x individual sequences and is thus represented by the ‘max’ function:
             x
        P = ∑αi
             1
             N
          = ∑ max{ f 1(rk ), f 2(rk ),..., fx (rk )}
            r =0
             N
          = ∑ p (rk )
            r =0


       Thus the de-interleaving algorithm will analyse the sample and attempt to extract
the individual sequences. We will now discuss the performance of various de-interleaving
algorithms in the literature under high pulse densities and complex signal conditions and
identify areas of improvement.

5.1    TOA Difference Histogram

         A sample signal deinterleaver is the TOA difference histogram. Each TOA is
subtracted from every subsequent TOA and a count is accumulated at each TOA
difference. Applying this to a stable PRI sequence will result in a count at an integer
multiples of the PRI. The number of computations required for a sample of E events is of
the order of O(E 2 )

       The difference histogram is an autocorrelation of the sample as can be seen by
applying a delay of d SI to the sample and correlating:
                    N
        y (d ) = ∑ p (rk ) p{( r − d )k}
                   r =0


       Thus for each delay d, i.e. for each PRI entry in the histogram, y(d) contains a
count equal to the number of solutions of the eqn. The full count of events in the ith
stable sequence occurs at the appropriate PRI and multiples the PRI. Each stable PRI
sequence is therefore identified by the correct count at each multiple of the PRI. A
threshold above which the sequence is said to be present must be defined. This allows for
                   CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                             VII-13
missing pulses and interfering pulses. If the counts at the PRI hormonics are less than the
thresholds, then the PRI is only a subinterval and does not form a sequence.

       However, when several signals, missed pulses, erroneous pulses occur the count
will vary and decision thresholds will be critical. Either signal will not be detected or the
false identification rate will be unacceptable. The histogram does not use sequential
information, it simply counts the number of event pairs separated by given PRI. Neither
does the histogram identify the sequences. However, as it is based on subtractions offers
fast processing.



5.2    Sequence search algorithms


        Sequence search provides identification of sequences and gives greater accuracy
and reliability than the difference histogram, but at the expense of processing speed.
Sequences of identical intervals are to be extracted from the sample; however the PRIs
and phases are unknown.

       As a starting point we can postulate sequence with any of the possible PRIs at all
phases and attempt to match with the sample. The signal postulated is
              N
       αp = ∑ fp (rk )
             r =0

       This can be correlated with the sample to give a count
             N
        y = ∑ [ fp (rk ) p (rk )]
            r =0

       This yields a value of y equal to the number of solutions to


        qp + amp = qi + bmi           i = 1,..., x
       Thus a maximum count of np occurs when the entire sequence is matched in the
sample. For a PRI of m there are m possible phases and the number of TOAs to be
correlated is N/m. assuming the PRI of signals can be from 1 to N, then to search for all
possible sequences would require of the order of O(N 2 ) computations.

                    CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                              VII-14
       As N is much greater than E, a far greater number of computations are required
than for the histogram method. Two points must be noted. First, it can be seen that the
multiples PRIs in the sample will be extracted unless the smallest intervals are examined
first. Secondly, signals will not have PRIs that are exact multiples of the sample interval.
For example a PRI of (m+0.5)SI would be measured as successive TOAs separated by m,
(m+1), m, … due to the quantisation. This sequence rapidly diverges from a sequence
with PRI of either m or (m+1)SI. Thus noninteger-PRIs must also be examined and a PRI
variation allowed for.


       Now, as the actual PRIs and phases in the sample are a small subset of the
possible values, the search should be limited to these.
       This can be achieved by selecting a pair of adjacent events from the sample and
projecting the TOA difference. An event is selected and the interval with the adjacent
event is projected. The postulated sequence is compared against the sample, with a PRI
tolerance on each TOA. If the match is insufficient, the interval between the event and
every subsequent event is projected and compared. This process is then repeated, starting
from different events. When a sequence is found the events are removed, thereby
simplifying further processing. These algorithms are efficient for only a small number of
PRIs with high-quality data. The extrapolation of a single PRI may rapidly diverge from
the actual sequence; this will prevent detection even if a large PRI tolerance is used. In
dense environments a significant proportion of measurement errors and missing pulses
will occur. This type of algorithm is prone to extracting multiples of PRIs.



5.3    Cumulative difference Histogram with weighted sequence search:


       This algorithm combines histogram technique with sequence search techniques to
obtain optimum de-interleaving. Initially a histogram is formed of TOA differences only
between adjacent events. This is the first difference. The count at each interval, and at
double the interval, is compared to a threshold. If both counts exceed the threshold then a

               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         VII-15
sequence search is performed at that PRI. If a sequence is not identified, the second
difference, i.e. the TOA difference between each event and the next but one event, is
calculated and the count is accumulated. The difference level is increased until detection
occurs, or until a particular difference level is reached. By requiring second hormonic to
be present searches are limited to cases where sequences of three events occur, rather
than only pairs. If higher hormonics were used then the search would be limited to longer
sequences; however the number of difference levels required is multiplied, thus reducing
the efficiency.

       When the sequence search identifies a sequence, the pulses are removed from the
sample and the histogram reset, thus simplifying subsequent processing. The smallest
PRI radars will thus be removed quickly and only the optimum number of TOA
differences calculated. For example, if five similar PRI signals are interleaved then
detection would occur on or before the tenth difference level.

       Sequence search techniques can now be restricted to the PRIs that are repeated a
sufficient number of times to form a sequence within the sample. The CDIF algorithm
quickly determines these PRIs. The conventional sequence search algorithms start from
any selected pulse pair. When the sample contains several signals or a staggered radar,
several unsequenced pairs exist, causing unsuccessful searches. A sequence of three
pulses has been chosen as the starting point for projection. This will reject random pulse
pairs and sub-intervals of a staggered PRI and provide a more accurate PRI.

       The conventional techniques simply counted the number of events matching the
expected positions. Use of weighting schemes enhances detection of uninterrupted
sequences. A simple scheme is to count the number of events fitting the sequence and add
number of these events separated by the correct interval. An alternative weighting
function is based on the probability of a sequence of events occurring without being
blocked. If average pulse density and pulse width are monitored, then the multiple of two
gives an estimate of the average probability of blocking a TOA measurement (R). Thus
the probability of a sequence of n intervals with PRI m is being measured is
        p(n) ≈ (1 − R) n     m >> 1 R < 1

                  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                            VII-16
       Thus, for each unbroken sequence of events found in the sample, the reciprocal of
this probability is added to the count. This automatically gives enhancement proportional
to pulse density.

       The significant drawback of this algorithm is the high number of difference levels
is required even in very simple cases, such as two-level staggered radar. Unintentional
PRI variation due to transmitter circuit imperfections can cause serious problems in the
algorithm, because histogram peaks decreases and could not exceed the threshold. Even if
it exceeds greater tolerances are necessary in the sequence search for the detection of
corresponding PRI sequences.

5.4    Sequential Difference Histogram with Weighted sequence search


       This algorithm solves the draw back of the above algorithm which requires high
number of differences even in very simple cases, such as two-level staggered radar. By
requiring the second hormonic to be present, the PRI calculation is limited to cases where
the sequence of three events occur, rather than only the pairs. However, the same thing is
can be achieved in the second part of the algorithm, in the sequence search, using
appropriate conditions for detection and extraction of PRI sequences. Therefore
discarding the condition that the second hormonics have to be present, the accumulation
in the difference histogram is no longer necessary. This is the main reason for using the
sequential difference histogram, which does not include accumulation in the successive
difference levels.


       In the SDIF histogram only current differences exist, and the histogram is much
clearer than the corresponding CDIF histogram. To extract the true PRI, it is sufficient to
compute second difference and to compare the histogram value with the threshold. There
is no need to compare the double the PRI with the threshold; hence the computation time
is more than halved.




               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         VII-17
       For every difference level c, the SDIF histogram is formed, the threshold is
calculated and all histogram values exceeding the threshold after the subhormonic check
become the potential PRI for sequence search. If the extraction of PRI sequence is
successful, the process is repeated until extraction of pulse trains, or until five pulses
remain in the entry buffer. If the sequence search cannot extract a PRI sequences, the
next difference is calculated, the new threshold is set and the whole process is repeated.
At the end of the PRI analysis, the staggered recognition is performed.

5.5    Multiple-parameter de-interleaving algorithm:


       Here the PRI analysis would be performed on groups of pulses previously sorted
by azimuth (DOA), frequency (RF), pulse width (PW), or other parameters. By using
other pulse parameters besides the time of arrival, new possibilities for efficient de-
interleaving are opened up.

       The analysis of other pulse parameters in this algorithm is predominantly
performed by using histograms. Azimuth is chosen as the most important de-interleaving
parameter because of its stability and reliability, and the azimuth histogram is formed
from all the pulses in the input buffer. The pulses grouped about every histogram
maximum correspond to the different emitters, and it is necessary to separate these pulses
according to their azimuth value. This presents the problem of determining the
boundaries of histogram groups. One possible solution is that the local minima of the
averaged histogram values are declared as group boundaries. Then, the pulses from the
input buffer with the same, or similar, azimuth are separated, and sorted into azimuth
groups. Inside every azimuth group it is necessary to analyse the next parameter.

       If we choose frequency as the second de-interleaving parameter then we have
pulse clustering in the azimuth frequency space. In the case of frequency-agile radars,
frequency clustering can cause the appearance of a large number of pulse groups, and,
consequently, ‘false’ emitters. By classifying the pulses into frequency groups inside one
azimuth group, the correlation between pulses from the same emitter could be last, and so


               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         VII-18
the application of PRI analysis could be almost impossible. Thus the time of arrival
(TOA) is chosen as the second de-interleaving parameter.

       As a result of PRI analysis, inside each azimuth group, pulses with the same or
similar PRI values are sorted into PRI groups. For better classification of detected
emitters, the carrier frequency is taken as the third parameter for de-interleaving, which
means that inside each PRI group the histogram of frequency is formed and analysed.

       For radars with fixed carrier frequencies, the frequency histogram will produce a
single significant peak. However, if multiple significant peaks appear in the frequency
histogram, we can conclude that the emitter is frequency-agile. Frequency clustering is
not carried out and the frequency histogram only tells us whether the pulses from the PRI
group belong to a frequency-agile radar or not.

       Staggered analysis is performed last. An important improvement on the overall
multiple parameter algorithm can be achieved if we introduce the two-dimensional
sequence search into the TOA de-interleaving algorithm. This consists of analysing the
pulse width as well as the TOA, ensuring reliable PRI extraction even in the case of very
high interleaved pulse density.



5.6    Our Implementation of De-interleaving:



       A variation of multiple-parameter de-interleaving algorithm is used in various
ESM processor designed by DLRL. The complete de-interleaving is done in the
hardware, using high speed Application specific integrated chips (ASICs), programmable
logic devices, FIFOs and static RAMs. FIFOs are used to suppress the peak incoming
pulse data rate with the processing rate. Data stored in the FIFO are read using
programmable logic devices and fed to the de-interleaving logic designed using
Windowed content addressable memory ASICs (WCAM). Pulse data is deinterleaved
based on frequency, pulse width and DOA of the incoming pulse with the predefined
tolerance limits for the band of pulses. The grouping of incoming pulses is done based on

               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         VII-19
their match with any of the previous pulses. In any pulse group the latest pulse is used for
comparison with the next incoming pulse.



        Digitized
        Radar
        Pulse data              DOA
                                                      Group 1
                                                                                 Sequence
                     Validate            Extract                   Extract       of pulses
                      Each      Freq
                                       Pulse groups   Group 2
                                                                     PRI
                      pulse             (WCAM)        Group n      (SDIF)
                                 PW




                                       Estimation
                                                                                    Send to
                     Scan and              Of                   Classification      display
                     Stagger/            DOA,                     Of radar
                       Jitter            AMP                      Based on
                     analysis            FREQ                      Library
                                          PW



       Figure: Sequence of operations in ESM Processor

       SDIF histogram is used on pulse group to extract PRI. If a single PRI is crossing
the threshold then a fixed PRI emitter is declared after a successful sequence search. If
more than one PRI is crossing the threshold then, stagger analysis and jitter analysis is
done based on the PRIs crossing the threshold value. Scan analysis is done based on gaps
between groups of pulses observed during sequence search.

       ESM Processor has to correctly classify emitters from a given radar library.

       Radar library consists of radar parameters and their tolerances. Extracted emitter
parameters are sequentially compared with the radar parameters to establish identity.

6.0    Conclusions:

       New techniques for de-interleaving based on Neural networks and Radar
identification using nearest neighbor pattern classification perceptrons are under study.

7.0    References:

       1.       MARDIA. H.K. : ‘New techniques for the de-interleaving of repetitive
sequences’, IEEE Proceedings, Vol 136 August 1989, pp. 149-154
                     CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                               VII-20
       2.   Milojevic. D.J. and Popovic. B.M.: ‘Improved algorithm for de-interleaving
of radar pulses’, IEEE Proceedings, Vol 139 No.1, Feb 1992, pp. 98-104




              CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                        VII-21
           CHAPTER – VIII



ELECTRONIC ATTACK SYSTEMS


Shri. P. PRATULANANDA PAL, Sc-‘F’




   CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                              VIII
                         CHAPTER - VIII


           ELECTRONIC ATTACK SYSTEMS


                            CONTENTS


1. INTRODUCTION


2. ROLE OF ECM SYSTEM


3. TYPES OF ECM TECHNIQUES


4. JAMMER EQUATION & SELF SCREENING RANGE


5. DEVELOPMENT OF ECM SYSTEM DESIGN
  CONSIDERATIONS


6. DESCRIPTION OF ECM SYSTEM


7. JAMMER EFFECTIVENESS


8. CONCLUSION


9. REFERENCES




       CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                  VIII
                          ELECTRONIC ATTACK SYSTEMS
                                   Shri Pratulananda Pal, Sc’F’
                                         DLRL, Hyderabad


1.        Introduction :


          A defensive measure system is developed to counter a military system, just as the
use of gas resulted in gas mask, use of heavy bomber Aircraft established the use of
radar. Thus, the countermeasure is developed against the radar itself. A military radar
operating in a hostile environment may be subjected to deliberate interferences which
appear on the display as real responses. These responses may be resembling the real
targets or may be a large number covering a significant portion or even full display. The
various measures (techniques) that interfere with the enemy’s electronic systems are
called Electronic Counter Measures ( ECM ).                The present days ECM is called as
Electronic Attack (EA). Use of Electro Magnetics ( EM ) are directed energy to attack
personal, facilities or equipments is called Electronic Attack which is divided in two
types :
                 1.        Destructive EA.
                 2.        Non destructive EA.


          Destructive EA can be treated as
                 Anti-Radiation Missile
                 Directed Energy weapons
                           High energy Laser
                           High Energy Microwave Transmitter


          Non-destructive EA is the radius measures that interference with the enemy's
electronic systems are called ECM.




                      CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                               VIII - 1
        The basic purpose of ECM system is to degrade the performance of enemy’s
electronic system disabling it to perform its intended mission by introducing deliberate
interference signals into it.


        Deliberate interference is potentially successful because of certain design,
inherent design weakness of the system. They are mainly
        (i)     Sensitivity.
        (ii)    The visual nature of these result on the radar display
        (iii)   The inability of the radar to distinguish and identify the precise nature or
                number of relatively small targets.


2.0     Role of ECM System:


        The ECM system is again classified in accordance to its function / jamming
strategies


        (1)     Self Protection Jammer
        (2)     Escort Jammer
        (3)     Stand-Off-Jammer


        An ECM System when carried by an aircraft or ship to provide its own protection
is called Self Protection Jammer. The platform has ability to protect itself against any
attack by meeting ESM, ECM integrated on platform. It is likely to be vulnerable to
“home on jam”.


        Escort ECM implies a dedicated aircraft carrying high power jammer which
accompanies the friendly strike forces and provides a protective jamming shield in
support of the entire force in the formation. It requires proper coordination in the entire
force. The escort jammer becomes a high priority target for the enemy defense system
and when rendered inoperable, leaves the entire force vulnerable to enemy action.



                  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                           VIII - 2
       Stand Off ECM implies a jamming vehicle which stand off (not accompanying
like Escort) at a distance beyond the effective range (Lethal range) of target defences. It
has the advantage of carrying large dedicated ECM pay load but has the disadvantage that
large ERP is required due to long jamming distance.


3.0     Types of ECM Techniques: [1]


       Electronic Counter measures Techniques depending on whether it generates or
reflected electromagnetic energies is classified as:
                 (i)       Active ECM Techniques
                 (ii)      Passive ECM Techniques


       Active Counter measure techniques are those which internally generates/amplify
the electromagnetic energy which is then radiated towards the radar. The Noise Jammer
and Repeater Jammers are examples of active ECM.


       Passive countermeasures do not generate/ amplify the signals but act in passive
manner by reflecting the energy             back to the radar. Examples of passive ECM are
Chaff, decoys and radar cross-section reduction which is also known as Stealth.


3.1    Active Counter Measure Techniques


       Deliberate interference can be introduced into the radar by injecting external noise
to raise the receiver noise level or by injecting spurious signals through radars main lobe
or side lobes.


       Active Electronic Counter Measures can be divided into two classes, depending
upon whether they are intended primarily for confusion or for deception, as


                 a.        Confusion Counter Measures / Noise Jamming Technique
                 b.        Deception jamming Technique


                      CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                               VIII - 3
3.1.1   Noise Jamming Technique


        The purpose of a confusion countermeasure is to mask or hide real targets by
cluttering the radar display. Its effects are similar to ground or sea clutter. The effective
confusion countermeasure should completely obliterate the radar screen. This is also
known as Noise Jamming and is shown in Fig.1(a) & 1(b).


3.1.1.1 Noise Jamming – Spot :


        A convenient classification of noise jamming is by the bandwidth of the noise
spectra added in jamming signal. A jammer whose Noise energy is concentrated in
relatively narrow band but wide enough to cover the receiver pass band is called a spot
jammer. It concentrates large amount of power per MHz. in receiver pass band. The
noise power density generated by spot jammer of 100 watts is 100/10 watts/MHz. where
BW = 10 MHz. This is in compare to receive signal of some mwatts. Once this noise is
allowed to enter the receiver, it is impossible to detect any target out of it. Concentration
of large power in narrow BW, which is its main contribution is its main disadvantage
also. Spot jammer is ineffective to frequency agile radar.


3.1.1.2 Noise Barrage Jamming:


        The disadvantage of Spot jammer is overcome in Barrage Jammer which has
much higher band width to cover the entire tuning range of radar. This is done at the
cost of its noise power density which is of the order of 100/1000 = 0.1 Watts/MHz.


3.1.1.3 Sweepthrough:


        A compromise between above two is swept through jammer. In this system the
jammer frequency is swept at a very high rate across the complete frequency band of
radar sweep rate should be such that the carrier to sweep the receiver pass band should
equal to the receiver response time i.e. time equal to reciprocal to bandwidth. This


                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                          VIII - 4
method allows high power density of spot jammer to appear for short period on all of the
possible radar channels. So it can jam simultaneously many radars operating at different
frequencies.


3.1.2   Deception Jamming Technique:


        The purpose of deception countermeasure is to present the false signals which
appear as though they were echoes from the real targets. If a sufficiently large number of
false targets were present on the radar display, the operator might not be able to process
them all. Some real targets might be lost, or else radar operator might direct a weapon to
a non existent target. The deception countermeasure is also known as Repeater Jammer
as repeats a replica of radar signal, but delayed in time, so that it appears displayed in
range and /or angle from the true target. These cause misdistance in missile system.


        The deception jamming technique is classified into


               a.        Deception in Range
               b.        Deception in Velocity


        A range gate stealer is a deception jammer whose function is to break the lock in
range of the tracking radar on the target. It operates initially transmitting a pulse of high
amplitude in synchronism with the echo return, then slowly shifts the timing of its own
pulse transmission is causing an apparent change in the target range, thus pulling the
range tracking away from the original target. This technique is known as Range Gate Pull
Off (RGPO) and is shown in Fig.1(c).


        The velocity deception technique is employed against a pulse Doppler radar or
CW Radar which employ tunable filter ( called speed gates ) to track the Doppler
frequency shift to calculate the velocity of the moving target. Velocity deception is
employed to deceive this type of radar by replacing a frequency shifted replica of the



                    CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                             VIII - 5
radar receive signal. This types of velocity deception is called Velocity Gate Pull Off
(VGPO).


3.1.3    Comparison between Noise and Deception Technique


         Noise jammer are usually of high power as they must be designed to obscure the
real target echoes. On the other hand, deception jammer is of moderate power since the
false echoes can be of comparable strength to the true echoes. The primary advantage of
noise jammer is that only minimal details about the enemy equipment need be known.
For better effectiveness, the deception countermeasure should be designed so that the
signals produced by them can not be distinguished from the real target echoes. If the
false target echoes do not match with the characteristics of the real echoes (the return of
the radar transmitted pulse by the target), then they can be rejected altogether by the
radar.


         Designing the deception-repeater jammer is obviously more complex than the
noise jammer. The deception jammer is most effective (as compared to noise jammer )
against modern radars which employ coherent integration techniques such as the pulsed
Doppler and pulse-compression types. This is because those radar have large processing
gain ( in the order of 20 to 60 dB) and hence attenuates the noise jammer signal by that
amount, while necessarily accepting any target like returns (Repeater jammer’s false
target) unattenuated.


3.2      Passive Countermeasure:


         Passive counter measures involve no radiation of Electro Magnetic Energy but
involve in actions taken to launch Chaffs Flares which deceive / degrade the performance
of the threat radar.


3.2.1    Chaff: It is also known as window. It consists of thousands of thin strips of
reflecting material like thin aluminum foil. The length and width of the strip depends on


                  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                           VIII - 6
the frequency of operation of radar to be jammed. Bundles of present chaff are dropped
by aircraft or they may fired by a small rocket. They scattered bythe wind and form a
highly reflecting cloud.


        Since the dipoles are resonant, they cover only limited band width of 10 to 15%.
Different sizes of Chaff are used to cover a larger band.


3.2.2   A decoy is a small aircraft like vehicle made to appear to the radar as a realistic
target. A small decoy can be designed to have radar cross section comparable with that
of large aircraft by fitting it with corner reflector, Luneburg reflectors or active repeaters.
This can be towed to target so that in SPJ role the ARM locks to decoy thus saving the
aircraft.


3.2.3   Stealth


        Another type of passive countermeasure is the reduction of radar cross section
(RCS) to escape detection. This falls into a general class of techniques called low
observable or Stealth technology [3]. Low observable technology and other
countermeasure techniques ( both active & passive ) are complement each other in
reducing the detectability of targets. When target RCS is reduced, the free space detection
range of convetional radar is reduced in proportion to б1/4 where б is radar RCS. On the
other hand the countermeasure syste becomes more effective for a given jammer effective
radiated power (ERP) by the figure of merrit (= ERP/ б ).




4.      Jammer Equation & Self Screening Range:


        The Radar detection capability is determined by the target S/N and the Jammer
capability depends on J/S ratio at the input of radar receiver. Where
               J = Jammer Signal Receive at the Radar
               S = Screen return of the target received at the Radar


                  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                           VIII - 7
                N = Noise


       The Jammer Signal (J) received at the radar varies inversely as the square of the
distance R between them (one way transmission ) and is represented by:


           J = Pj x Gj x Ar            … … … …. … …. (2)
                   4 π R² x Bj                  Where
                                                Pj =    Jammer Tx Power
                                                Gj =    Jammer Antenna Gain
                                                Bj =    Jammer Rx Bandwidth
                                                Ar =    Radar Antenne Aperture


       The radar echo signal (S) received at the radar varies inversely as fourth power of
the distance between them (two way transmission ) and is represented by:


                  Pt x Gt x σ x Ar              … … … ….            ….    (3)
           S=
                  4 π R² x 4 π R² x Br          Where
                                                Pt =    Radar Tx Power
                                                Gt =    Radar Antenna Gain
                                                Br =    Radar Rx Bandwidth
                                                Ar =    Radar Antenne Aperture
                                                σ =     Radar Cross section


       It is, therefore,


                   Pj x Gj x4π x Br x R²         … … … …. … … (4)
         J/S=
                   Pt x Gt x σ x Bj


       Although radar transmitted power and antenna gain is much higher than those of
jammer, but there will always be range where both jammer signal and radar echo signal
will be equal. This distance is called Self Screening Range (Rss) or Cross-over Range or


                  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                           VIII - 8
Burn Through range below which echo signal is stronger and above this range, the
jammer signal is stronger. Equating J and S is equations (1) and (2) above, we get, Self
Screening Range as


        R²ss = Pt x Gt x σ x Bj                 … … … ….           ….    (5)
                     4 π x Pj x Gj x Br


5.     Development of ECM System Design Considerations: [4] [5] [6]


       A principal objective of an ECM system is to achieve an adequate ratio of
jamming power to the received echo signal.             For effective jamming the required
jamming-to-signal (J/S) ratio depends on several factors. These factors include the nature
of the ECM mission (escort or self protection), the geometry of the engagement,
parameters of the radar, back-scattering cross section of the target, and the types of ECM
techniques planned, as well as on-radar electronic counter-countermeasures (ECCM)
techniques.


       In addition to these parameters, the actual J/S ratio achieved depends on the
effective radiated power (ERP) directed toward the radar by the ECM system. ERP is the
product of antenna power gain and the RF power applied to the antenna by the
transmitter. Depending on the application and jamming transmitter used, the gain of an
ECM transmit antenna will range from 6 to 30 dBi with respect to a matched polarized
isotrope.


       While adequate ERP is an essential requirement of the ECM system, other factors
must also be addressed. The system must be able to


               (i)       direct the ERP in the threat direction
               (ii)      change that direction rapidly to jam other possible threats and
               (iii)     cover a wide sector in azimuth and elevation by the jammer
                         antenna.


                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                          VIII - 9
       Furthermore, the antenna must also be compatible with the transmitter in
frequency range and power handling capability, and permit jamming of radar having a
variety of polarisation.


5.1    Optimun utilization of jammer resources :


       Early EW systems of the first generation permitted optimum utilization of the
jammer resources only in frequency domain by transmitting the jamming signal of correct
frequency and bandwidth. They did not permit the optimum utilization of jammer
resources in time and space domain as the jammer signal was continuously transmitted
over a wide angular sector, once switched on. Similarly the EW systems of the second
generation made optimum use of the jammer resources in frequency and space domain
only by transmitting the jammer signal of correct frequency and bandwidth over a narrow
angular sector but did not do so in time domain because as in the case of first generation
systems, they continuously transmitted jamming signal, once switched on.


       Application of phased array antenna techniques to EW permits the optimum
utilisation of the jammer resources in space, time and frequency domain. The ability of
the phased array antenna to switch the beam in different directions in microseconds time
interval makes it possible to jam many threats in a time shared mode on a pulse by pulse
basis., using cover pulse technique. In this technique the jamming signal of correct
frequency is transmitted for a duration little more than the pulse width of the threat radar
so as to synchronise with the pulse received from it. Each jamming cover pulse is
modulated with an optimum wave form for most effective jamming. Thus each threat
receives a jamming signal of optimum wave form at the correct time and frequency for an
optimum duration and having an optimum frequency band. This ensures optimum
utilisation of jamming resources in time and frequency domains. By designing the array
for a sufficiently narrow beam width, the jammer power is concentrated over a narrow
angular sector for increased jamming effectiveness which contributes to optimum
utilisation of the jammer resources in space domain.



                  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                           VIII - 10
5.2    Power Management


       Generally, it is not possible to counter simultaneously          all enemy electronic
systems, and hence it is necessary to manage ECM resources so as to counter those
systems which pose the greatest threats to a particular mission. This dictates an overall
strategy of ECM deployment which is known as Power Management of ECM system.
Size and weight are limited in strike aircraft, presenting a problem in carrying ECM
jamming transmitters covering the full radar band ( eg. 0.5 – 18 GHz ). One solution is to
use external jammer pods. The stand- off jammer has the disadvantage is that large ERP
is required due to the potentially long jamming range and the need to jam into the side
lobe pattern of the victim antenna.


6.     Description of ECM System:


       Present day ECM systems are designed specifically to counter the pulsed, CW
and Pulse Doppler Radars. In a highly dense threat environment with these types of
emitter, ECM system should able to meet the stringent requirement of handling
simultaneous multiple threats effectively by providing efficient power management
coupled with intelligent jamming techniques. The detailed study of blocks in Fig.5
provides the insight of present day ECM systems used in the battlefield. [3]


       The ECM system is designed as per specifications to detect, classify and identify
hostile radar threats and to direct jamming responses automatically against these threats
on a priority basis. The system should be provided with future growth potential. The
functional part of ECM system can be divided into (i) ESM system which intercepts and
develops attributes (Eg. Pulse descriptors) of the threat (ii) a receiver processor which
each threat (iii) computer which compares the threat data against the pre-stored threat
library and established the prioritized response i.e. JPRO number (Jammer Program
Number) to each detected threat. (iv) a technique generator which translates the JPRO
number into appropriate RF modulation technique suitable for application to the jamming
transmitter (v) the ECM processor which acts as a controller of the whole ECM


                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                          VIII - 11
transmitter (vi) Repeater where the input signal is stores for deception technique (vii)
Repeater channel which generates the jamming signal synthetically at the threat
frequency. (viii) jamming transmitter and antennas covering the band of interest.


       The computer is the heart of the modern ECM system. It makes the deception on
the relative priority of the each threat after comparing the collected threat data against the
stores data in the library. Its ability to program on a mission engagement basis is a
invaluable in optimising the use of the available jamming resources.               Also, the
computer’s reprogramming capability makes it ever useful by the emergence of new
threat or intelligence data. The computer allows a real time solution to the complex
problem of allocating the jamming resources of the ECM system in spatial, time, power
and spectral domain. This is called power management and is dependent on the mission
requirement, the jamming assets available and the characteristic of the platform to be
protected. The architecture of the computer can be either centralized or distributed where
a network of microprocessors is spread throughout the ECM system.


       The effective radiated power (ERP) is transmitted through the transmitter which
include high power amplifiers and logic matrix to point the steerable beam (i.e. phased
array) at the threat or select the proper fixed antenna transmitter used to be provided with
good cooling system to dissipate the heat generated by the high power amplifier. For
ship borne system, a stabilized platform to take care of roll and pitch movement is used.


       Modern ECM systems employ a look-through (LT) mode to provide periodic
monitoring of the threat environment while simultaneously jamming multiple emitter.
The look-through mode is generally interfaced with the jamming transmission so that the
intercept receiver can operate at full sensitivity for short duration while jammer is turned
off.


       The advance of radar threat pose a significant challenge to the ECM system
design and may cause a revision in architecture of future ECM system due to (i)
extension of radar frequency into millimeter wave frequency band (ii) the use of agile


                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                          VIII - 12
frequency agile PRF radar (iii) the proliferation of coherent intra-pulse (pulse-
compression) and inter pulse (i.e. pulse doppler) radar requiring the use of coherent
jamming ECM waveform to combat the radar high processing gain (iv) the development
of low probability intercept (LPI) radar which use noise like spread spectrum wave form
to escape detection by ESM system.


7.      Jammer Effectiveness:


        The effectiveness is a quantitative measure of performance of the system to meet
the required objectives in the specified mission.


        To have better effect of counter measure on the threat emitter it should be ensured
that ECM transmitter radiates at:


               (i)     the correct frequency
               (ii)    the correct direction
               (iii)   the correct time
               (iv)    the correct jammer program (JPRO)


        Transmitted frequency can be correctly tune with the emitter frequency by using
Automatic Frequency Control (AFC) while generating the signal synthetically for noise
jamming. In the process higher spectral power (Power/MHz.) with higher J/S ratio at the
victim radar is achieved with spot noise jamming of narrow band width.


        The proper direction of transmission is determined by measuring the direction of
arrival (DOA) of the radar signal with the help of high accurate DF system.


        Transmission with correct JPRO is most important for the effectiveness of ECM
system. It is not only the selection of technique but also its parameters play important
role.



                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                          VIII - 13
There are a few factors which cause the effectiveness of jammers:


(a)    Polarisation:      Need to accommodate victim radar with various
       polarisation force the jammer radiation either 45º         slant or circularly
       polarised.   This causes 3 dB loss of jammer and radar antennas are
       orthogonally polarised.


(b)    Quality of Noise: Based on information theory it is reasoned that white
       (uniform spectral density) Gaussian noise is the best noise-jamming
       wavwform. This follows because white Gaussian noise has the maximum
       entropy, or uncertainty, of any random waveform for a specified average
       power. It is statistically impossible to distinguish between receiver noise
       generated in the victim systems and the externally injected Jammer signal.
       But in practice it is difficult to generate high part Gaussian noise since
       final amplifying tubes (i.e. Travelling Wave Tubes) are limited in peak
       power they can handle. If the jammer noise bandwidth is much larger than
       radar bandwidth, even though the noise input distribution is highly non-
       Gaussian due to clipping of noise peaks, the output of the victim emitters.
       If amplifiers tends to approach a Gaussian distribution because of central
       Limit Theorem. This status that the probability density of the sum of a
       larger number of independently distributed quantities approaches
       Gaussion     probability   density   function    irrespective    of   individual
       distribution provided that the contribution of any one quantity is not
       comparable with the resultant of all others.


(c)    FM Signal       : The saturation of power amplifier has little effect on
       frequency modulated signal. Uniform spectral density than Gaussian is
       desirable for FM and one means to achieve the uniform spectrum is to
       frequency modulate the jammer transmitter with saw tooth wave.
       Randomness can be added by combining the noise with saw tooth wave.



         CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                  VIII - 14
                       For Wide Band FM if the receiver impulse response duration
             (1.1/Rx BW) is much greater than jammer deviation rate, then a number of
             randomly spaced (and, hence, independent), overlapping impulse
             responses will be added together to form the victim receiver’s output
             waveform. This waveforms meet the condition for Central limit thereon,
             and hence, has an amplitude probability distribution which approaches a
             Gaussian distribution.


                       The response to jamming with narrow band FM waveform in the
             victim receiver produces the effect of many of many randomly spaced
             spurious target. This creates a confusion effect making it difficult to
             locate the real target.


      (d)    Repeater jammer must have very short turn around time (50 nSec to
             100 nSec) to minimise the probability of leading edge range trackers
             rejecting the deceptive pulses.


      (e)    Parameters of the counter measure techniques should be critically chosen
             eg. Dwell time, slope of RGPO/VGPO curve, max pull of deception
             techniques and bandwidth of noise jamming.             These parameters are
             different for different but of same class radars.                Parameters of
             countermeasure techniques can be selected by using “Finger Printing”
             technology which is a new field to provide the signatures of all enemy
             radars.
      (f)    Combination of active & passive counter measure will provide much
             better effect than any single form.


8.    Conclusion :


      Design & development of effective and affordable EA systems is becoming
challenging day - by - day. The past history of EW industry has demonstrated cost,


               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                        VIII - 15
effectiveness and reliability factor driven to the EA systems to the edge of operational
viability. The compact efficient transmitter using microwave power module will be the
building block of present day ECM system. Co-operative jamming using both active and
passive counter measures will make the modern ECM more effective. The expendable
jammers and response jammers will pay a major role against missile defence.
Development of high power microwave sources leads to the Direct Energy weapons
which can destruct the front end electronics of the radar missile sensors thus making them
blink. EA is the key element in present day wars. It acts as a decisive force multiplier for
the success. The recent experience shows a significant result in favour of a country who
uses them most.


9.     References:


        1.   Skolnik, MI, Introduction to Radar System , Mc-GRAW Hill.
        2.   Robert J. Schlesinger - Principles of Electronic Warfare
        3.   Schlcher, DC, Norwood, MA - Introduction to Electronic Warfare
             Artech House, 1986
        4.   RL Moynihan, Phased Arrays for Airborne ECM
             Microwave Journel, Jan ’87
        5.   Murray Simpson- High ERP Phased Array ECM Systems
              Journel of Electronic Defense, Mar ’82.
        6.   Leonard Mansky – Broad band Phased array antennas, Microwave Journal,
        Sep ’84.
        7.   D. Curtis Schleher, Norwood, MA-Electronic Warfare in the Information
        Age. Artech House, 1986.




                   CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                            VIII - 16
                               Radar screen
                                                              PPI without jamming        PPI with Jamming
Clear radar screen          with Noise jamming

    Fig. 1(a) Noise Jamming                                             Fig. 1 (b) Noise Jamming


       COVER JAMMING




                                    Fig. 1 (c) Deception Jamming


                                     Fig.1 Jamming Techniques
                     CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                              VIII - 17
GUIDED OR
HOMING
MISSILE



                                        DECOY

                      DECOYS




                Chaff
               Corridor




                                   CHAFF
                    Fig. 2 Passive Jamming
            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     VIII - 18
                                            STANDOFF



                                                  LETHAL
       JAMMING                                    RANGE
                                                                              SELF
                                                                           PROTECTION




ESCORT SUPPORT
                                                                     JAMMING




  JAMMING




                  Fig. 3 JAMMING STRATEGIES
            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     VIII - 19
   Si   +20                                        Sj = Radar signal power
                                                   received by repeater
   g
   n      0
                              J = Repeater power
   al                         received by radar ( Pj = 1 KW )
   le   -20
               Cross over
   v            points
   el   -40
                     J/S ratio = 10 dB
   s                 at 1.5 miles
   in   -60
                                                                S = Reflected signal
   d                                                            received by radar
   B    -80
   m
        -100


                              1.0            10                     100       ( log scale )

                                         Range in miles



Fig. 4 SIGNAL POWER VARIATION WITH DISTANCE BETWEEN TARGET AND RADAR


                            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                                     VIII - 20
                                  SYNTH.
       FRONT END
                                 CHANNEL




                    FM                            TECHNIQUE                TRANSMITTER
                                                  GENERATOR
OMNI

        DF ANT.




         DF & IFM                                                ECM
                                                               PROCESSOR



           Rx
        PROCESS                COMPUTER

                                                                                         DISPLAY




                              Fig 5 ECM BLOCK DIAGRAM


                    CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                             VIII - 21
              CHAPTER –IX


DSP BASED SERVO SYSTEMS


   Shri. SIVA RAJU DADI, Sc-‘F’
                &
   Shri. M.P. NARENDER, Sc-‘C’




CEP Course on Radar EW held at DLRL on 25th Oct to 29th Oct 2004
                              IX
                         CHAPTER - IX


              DSP BASED SERVO SYSTEMS


                              CONTENTS


1.0     Introduction
2.0     Overview of control systems
3.0     Servos used in EW systems (General description)
4.0     Case Study
5.0     References




      CEP Course on Radar EW held at DLRL - 25th Oct to 29th Oct 2004
                                    IX-1
                          DSP BASED SERVO SYSTEMS


                   Siva Raju Dadi, Sc’F’ and M.P.Narender, Sc’C’


1.0     Introduction:


       Servos are being used in both ESM and ECM sub - systems of EW systems. In ESM, the
servos are used to rotate / position the rotary DF antennae. These are hybrid servos, which
operate in both velocity mode and position mode. These operate in velocity mode to rotate the
antenna in high speed, to find the direction of emitter and operate in position mode to position
the ESM antenna at required position for detailed analysis of a particular emitter. Servos are also
used to test and calibrate the direction finding sub–systems of ESM.


       In ECM, servos are being used to move the jammer antenna in the direction of hostile
emitter for effective jamming. These servos in conjunction with Auto Track Receivers, track the
moving hostile emitters accurately for effective jamming. Servos are being used for stabilization
of multi-beam jammer to stabilize against pitch and roll motion of a ship / aircraft. Servos are
also being used for testing the multi beam jammer.



2.0     Overview of control systems:

      There are different control systems to control different parameters in different areas. The
examples control systems are:
       1) Motion control systems
       2) Process control systems
       3) Biological control systems
       4) Financial control systems and etc.


Motion control systems are also called as servo control systems.


              CEP Course on Radar EW held at DLRL on 25th Oct to 29th Oct 2004
                                           IX-1
Servo Control Systems: Servo control systems are used to control the motion of a body in
one or more axes of three longitudinal axes (X, Y, and Z) and three rotational axes (pitch, roll,
and yaw). The rotational motion can also be represented in terms of azimuth and elevation.


       The servos can be classified based on its actuator as:

          1.   Electromechanical (uses electrical motors).
          2.   Hydraulic.
          3.   Pneumatic.


       The servos used in EW systems, developed by DLRL are electromechanical servos. An
electromechanical servo uses DC motors or DC brush less motors or AC motors or stepper
motors.


       These servos operate in different modes, depending upon application as given below:


          1.     Position mode servos
                 a.   Set motion: The destination is known in advance.
                 b.   Tracking: The destination is continuously updated.
          2.     Velocity mode servos.
          3.     Torque servos.
          4.     Hybrid servos.


3.0    Servos used in EW systems (General description):


       The servos in EW systems are electro-mechanical servos operate in position mode or
velocity mode or hybrid mode.




               CEP Course on Radar EW held at DLRL - 25th Oct to 29th Oct 2004
                                             IX-2
       A typical high performance servo will have multiple loops as shown in fig.- 1. This also
can work as a hybrid servo to control the selected parameter as per the input command. It is a
position loop servo, which moves the load as per the position command. It is having inner
velocity loop. The velocity loop is having inner current loop. This also can work as torque
(current) servo or as velocity servo by accepting external current or velocity commands. The
function and features of each block of servo are briefly explained below.


       1. Profile generator: The position loop operates in two ways (a) after receiving a
       position command; it moves the load to the commanded position in a predetermined
       velocity and acceleration. After reaching the destination with a specified accuracy it is
       ready to accept further position command. It is called set motion command. The “Profile
       generator” generates the incremental position command with a predetermined velocity
       and acceleration and gives to the input of position loop. (b) Another mode of position
       loop operation is “track” mode of operation. In this mode the position commands are
       continuously updated / accepted and moves the load as per the command with minimum
       possible error. The velocity and acceleration of the movement of the load are as per the
       velocity and acceleration of input position command. In a typical target tracking
       application, the servo operates in both ways. It initially   operates in set motion control
       mode to catch up the target and then it changes to track mode to track the target.


       2. Position comparator: It generates position loop error. It is simple subtractor or
       subtraction instruction in case of digital loop.


       3. Position loop compensator: This block is designed to maintain the balance
       between stability margins and gain of the position loop. The gain of the loop determines
       the bandwidth of the loop. The bandwidth indicates the performance of the loop. The
       compensator is generally designed by drawing Bode plots of the mathematical model of
       the system. Using control system toolbox of MATLAB software, we can draw the Bode
       plots of the modeled system. The compensator generally consists of lag / lead / lag-lead



            CEP Course on Radar EW held at DLRL - 25th Oct to 29th Oct 2004
                                          IX-3
 networks or PID (Proportional , Integral, Differential) circuits. These can be
 implemented by using hardware (Op-Amps, etc) or by software in digital loops.


 4. Velocity feed forward: To improve the transient (dynamic) performance of the
 position loop, the velocity of position command is calculated in on line and given
 directly at the input of velocity loop   by bypassing the position loop compensator as
 shown in fig.2. It will avoid delay in position loop compensator and improves the
 transient performance. The velocity feed forward is implemented by using hardware or
 software differentiators.


 5. Velocity command adder: It adds the output of position loop compensator and
 velocity feed forward signal. This adder is implemented by using OP-Amps or a simple
 add instruction in case of digital loop implementation.


6. Velocity comparator: It accepts the velocity command from outer position loop
 or an external velocity command when the system is operating as velocity servo. The
 velocity command is compared with feedback velocity and velocity loop error is
 generated. It is implemented by Op-Amps or by using simple subtraction instruction in
 software.


 7. Velocity loop compensator: This is similar to position loop compensator, to
 maintain the balance between stability margins and bandwidth of velocity loop. This is
 designed and implemented similar to position loop compensator. Generally, the velocity
 loop compensator is bypassed when the system is operated as position servo.


 8. Acceleration feed forward: The acceleration in the position command is
 calculated in on line and directly applied to the input of the current loop. The
 acceleration command demands the corresponding force / torque and hence the motor




     CEP Course on Radar EW held at DLRL - 25th Oct to 29th Oct 2004
                                   IX-4
current. In servomotors, generally, torque is proportional to the armature current. This
will improve the dynamic performance of the system.


9. Current command adder: It adds the output of velocity loop error and
acceleration feed forward signal.


10. Current comparator: It selects the output of velocity loop or external current
command. It selects external current command when the system is being used as a torque
servo. It compares the current command with current feed back to generate the current
error.


11. Current loop compensator: The current loop compensator is designed to have
the required bandwidth with the required average and peak current capabilities. It is a
high power electronics circuit which should have fool proof overload protections.


12. Power stage: It develops the required voltage across the windings of the motor to
counter the armature back emf and supply the required armature current to generate the
required torque. The power stage can be a linear amplifier for low and medium power
servos but modern servos generally use PWM switching power amplifiers. The switching
amplifier uses high power MOSFET switches or IGBT (Insulated Gate Bipolar
Transistor) switches.


13. Prime power supply: It is a raw DC supply that converts the prime power
supply to the required DC supply to the power amplifier. A bulk holding capacitor is
connected across the raw DC supply to maintain the stiffness during transient high
current demands.




     CEP Course on Radar EW held at DLRL - 25th Oct to 29th Oct 2004
                                   IX-5
14. Current sensor: Current sensor            senses the armature current. The output of
current sensor is generally isolated from high power circuit for reliable operation of low
power control circuit. Precision resistor with a analog isolation amplifier or hall effect
current sensor can be used as current sensor. The current sensor output is digitized using
A to D converter in the case of digital current loop.


15. Motor: It is an electromechanical device to convert the electrical energy to the
mechanical energy. The motor can be DC brush type motor or DC brush less type or AC
motor or stepper motor.


16.       Velocity sensor: It is an electromechanical device, which senses the velocity
of motor shaft. Tacho meter, resolver, Hall effect sensor, or shaft encoders can be used as
velocity sensors. The accuracy of velocity directly depends upon the accuracy of velocity
sensor.


17. Gearbox / Pedestal: The reduction mechanism gear is used to match the
optimum speed range of the motor to the required speed range of the load. The gearbox
should have minimum backlash to minimise the non–linearity. There are different types
of gear mechanisms like spur gears, harmonic gears, and etc.


18. Position sensor: The position loop accuracy and resolution depends upon the
position sensors. Resistors, synchros, resolvers, shaft encoders are generally used as
position sensor. A to D, synchro / resolver to digital converter can be used in digital
loops and digital display of position information. Incremental shaft encoder gives digital
pulses that are used to evaluate the shaft position. Absolute     shaft   encoder directly
gives the digital code of shaft position.




      CEP Course on Radar EW held at DLRL - 25th Oct to 29th Oct 2004
                                    IX-6
4.0    CASE STUDY: A DSP Based Servo System


4.1    Introduction


       DSP based servo system is developed in servo systems division of DLRL. The block
diagram of the servo system is given in figure below. The actuator is DC motor driving a gear
unit with “incremental encoder” mounted on motor shaft as a feedback sensor. From the
incremental encoder outputs, both position and velocity of the load is evaluated by DSP. A
reference pulse is generated when the load is at reference position and it is used by DSP on
power on, to initiate the load position accurately.


       This is a hybrid servo, which can operate either in velocity mode or in position mode. A
PWM power amplifier (SA 01) from M/s APEX is used as voltage amplifier. A hand held PC
acts as host device from which we can operate in different modes and record the selected
parameters. The recorded data can be used for off-line evaluation.

       A DSP, model number TMS320F240 from M/s Texas instruments is selected. This DSP
is designed for motor control applications. The DSP contains almost all peripherals required for a
typical servo system. The functional blocks in figure below, which are implemented by DSP are
shown inside the dotted lines. “Event Manager”, which includes incremental encoder interface,
high speed inputs and etc is used for velocity and position measurement. Asynchronous serial
link peripheral is used to communicate with host PC. It is also having dual 10-bit ADC’s, which
are used for current sensing for digital current loop.




             CEP Course on Radar EW held at DLRL - 25th Oct to 29th Oct 2004
                                           IX-7
                                     BLOCK DIAGRAM OF DSP BASED SERVO
 |---------------------------------------------------------------------------------------- |
 |         VELOCITY COMMAND
                                                                                             |
                                                      VOLTAGE
 | VELOCITY                                           FEED                                   |
 | FEED
     FORWAR           COMMAND VELOCITY                FORWAR
                                                      D              COMMAND VOLTAGE         |
     D                    ESTIMATION
 |                                                                      ESTIMATION
                                                                                             |
 |                                                                                           |                      Vss
 | +                                          +                                             +|
 |                                                                                           |         ISOLATION   PWM VOLTAGE MOTOR
                      COMPENSATION                                       COMPENSATION            DAC                                    PEDESTAL
 |                                      +                 +                            +     |         AMPLIFIER    AMPLIFIER & GEA
                                                                                                                                R
 |PO          -                                                    -
  S                                              Position/velocity                           |
 |COMMAND                                        mode                                        |
                                                 select
 |                                                                    FEEDBACK     VELOCIT   |                            INCREMENTAL
                                                                       FILTE       Y
                                                                                  ESTIMATIO                                 ENCODE
 |                                                                     R          N          |                              R
 |                                                                                           |
                                                                             POSITION
 |                                                                         ACCUMULATO        |
 |                                                                         R
                                                                                             |
 |                  DSP BASED CONTROLLER                                                     |
 |                                                                                           |
 |                                                                                           |
    ---------------------------------------------------------------------------------------
                                                                          POSITION REFERENCE PULSE
                          RS232 INTERFACE
                  PERSONA
                  L
                  COMPUTER




          The on-chip clock generator with watchdog timer will avoid spurious hangings. On chip
interrupt controller is used to synchronize the DSP power stage. The program is running from on
chip flash EEPROM. Any modification in parameters or modes can be done by directly
downloading software from PC to the flash memory in DSP.

4.2       Design of velocity loop:

          Incremental encoder mounted on the motor shaft is used to sense feedback velocity as
well as feedback position. A digital velocity loop is designed to have a bandwidth of
1200rad/sec. To improve dynamic response voltage feed forward is applied. To have better
steady state accuracy a digital integral network is implemented. To compensate the effect of
static friction a “signum” (signed offset) is applied in forward path. The sign of the signum is
same as the sign of velocity error. The velocity is measured by using the incremental encoder
pulses. The time difference between two encoder pulses is measured, from which the velocity is
calculated. The on chip capture unit peripheral in DSP is used to measure the time difference
between two edges. Block diagram of the velocity loop is shown in figure below.




                   CEP Course on Radar EW held at DLRL - 25th Oct to 29th Oct 2004
                                                 IX-8
                                  BLOCK DIAGRAM OF VELOCITY LOOP

                                                              8000H
                              SIGNUM


                           VOLTAGE FEED
                             FORWARD
                             CONSTAN                                  Exp (- jT/2
                                                                               )
                             T
                                                +         +     +
         +
                      VEL        INTEGRA
                                                                          Z                   ISOLATION    POWER          MOTOR
                      .
                      LOOP       L MPEN-
                                  CO                                                DAC                                                      PEDESTAL
                                                                          O                   AMPLIFIER   AMPLIFIER       & GEAR
   MILL               GAIN         SATIO
   I                                        +       +         +           H
   DEG./ SEC.                      N                     2’
                  -    K                                 s
                                                         COMPLEMENT
                                 1 + 50/s                                           5/6553        1          48/5     (11.3 / (1+S/153)) * (1 / 678.96)
                       v                                 TO                         6                                 6
                                                                                    V/COUNT
                                                         OFFSET
                                                         BINARY
                      MILLI DEG. PER SEC.                                                                 RAD. /
                                                              (360 *           Π                          SEC.
                                                              1000)/2




Torque constant of motor (Kt) = 0.088 Nw-m/Amp.
Estimated Mechanical time constant of motor with gear and load = 6.5 m Sec.
Approximate motor transfer function Tf = (1/ Kt)/(1+s/τm) = (1/0.088)/(1+s*6.5*10 –3)
                                                        = 11.36/(1+s/153) rad /sec./volt.
Motor transfer function in low frequency format = (11.36 * 153)/(s (1 + 153/s))
(refer reference1 for the details on formats)
Loop transfer function (without voltage feed forward) =
Kv*(5 / 65536)*1*48/5*((11.36 * 153)/(s (1+153/s)))*1/678.96*((360*1000)/(2 * Π))
= Kv * 107.43 * (1 / (s * (1+153/s)))
The gain crossover frequency is ≈ Kv * 107.43
The gain crossover frequency is approximately equal to closed loop Bandwidth of the system.
(Ref: Reference1)
Required bandwidth = 1200 rad/sec.
Kv * 107.43 = 1200
Kv = 1200 / 107.43 = 11.17


4.2.1        Design of voltage feed forward:


Let voltage feed forward constant = VFFC




                 CEP Course on Radar EW held at DLRL - 25th Oct to 29th Oct 2004
                                               IX-9
The feedback speed should be equal to command speed. The feed forward voltage should
generate the required speed.
Let command velocity = Vc
In steady state, the velocity due to voltage feed forward
Vc = Vc * VFFC *(5 / 65536)*1*48/5*11.36 *1/678.96*((360*1000)/(2 * Π))
VFFC = 1.4 * 10 9 / (9.28 * 10 8) = 1.43


4.2.2   Implementation of integral network:


To improve steady state accuracy, an integral network is implemented.
Transfer function of integral network = 1+ 50/s
Applying bilinear transformation, with velocity loop sample period of 1millisecond, we get y(n)
= (1/40){ 41 * x(n) – 39 * x(n-1) } + y(z-1)
The above difference equation is implemented to realize the integral function. While operating
the system, as position servo with velocity loop as inner loop, the integral function is disabled for
stable position loop operation


4.2.3   Signum:


        The value of the signum is experimentally measured. This is the count required at the
input of DAC to just overcome the static friction. The sign of the signum changes as per the
direction of rotation.


4.2.4   Closed loop transfer function (without integral function):


Forward gain (A) = 11.17 * (5 / 65536) * 1 * 48/5 * (11.36 / (1+s/153)) * 1/678.96
                    = 1.37 * 10 –4 * 1 / (1+s/153)
                    = 153 * 1.37 * 10 –4 * 1 / (153 + s)
                    = 0.0209 / (s+153)
Feedback gain (B) = 360 * 1000 / (2 * Π) = 57295



             CEP Course on Radar EW held at DLRL - 25th Oct to 29th Oct 2004
                                          IX-10
  Closed loop gain = A / (1+AB)
                             = (0.0209 / (s+153)) / (1+ 0.0209*57295/(s+153))
                             = 0.0209 / (s + 153 + 1197) = 0.0209 / (s + 1350)
  Note that closed loop band width is slightly more than the open loop crossover frequency.


  4.3       Design of position loop:


         Refer block diagram of position loop in figure below. Position loop is designed to have
  crossover frequency (closed loop bandwidth) of 12 rad/sec. To improve the steady state response
  a lag network is used. To improve the dynamic response velocity feed forward is used. The block
  diagram of the position loop is shown in figure below




                                   BLOCK DIAGRAM OF POSITION LOOP


                                       COMMAND VELOCITY
                                         CALCULATIO
                                         N


                                                                                  +
POSITION COMMAND
    (MILLI DEG.) +                           Lag
                                                                                           0.0209 / ( s +   1/S
                                             network
                                    Ks * (1 + 10 / S) / (1 + 2 /    +                      1350)
                                    S)
                         -




                                    MILLI
                                    DEG.                           360 * 1000 / (2 * Π )




                     CEP Course on Radar EW held at DLRL - 25th Oct to 29th Oct 2004
                                                  IX-11
Loop transfer function = KI * (1+s/5) / (1+s/1) * 0.0209 / (s+1350) * 1/s * (360 *
                           1000 / (2 * Π))
Loop transfer function in low frequency and high frequency format
= KI * 1/5 * (1+5/s) / (1+1/s) * 0.0209 / (1350 *(1+ s/1350)) * 1/s
  * (360 * 1000 / (2 * Π))
The gain crossover frequency is ≈ KI * 1/5 * (0.0209 / 1350) * (360 * 1000 / (2 * Π))
Required bandwidth = 12 rad/sec.
KI * 1/5 * 0.0209/1350 * (360 *1000 / (2 * Π)) = 12
KI = 67.64


4.3.1     Velocity feed forward:


        The velocity of the position command is calculated at Position loop sample rate and used in
generating velocity feed forward.


4.4       Real Time Kernel:


A typical position loop will have inner velocity loop and a velocity loop will have inner current
loop. The DSP is required to execute these loop algorithms concurrently. Apart from
implementation of these servo loops, the DSP has to implement the host interface
communication. The required sample rates for inner loop is typically 10 times compared with its
next outer loop. For convenient and flexible implementation the sample rate of the inner loop is
selected as harmonic of outer loop sample rate. The sample rate of position loop is selected as
100 samples/sec and velocity loop sample rate is selected as 1000 samples/sec and current loop
sample rate is selected as 10,000 samples/sec. The velocity loop algorithm is divided in 10 parts
(10 procedures) and position loop algorithm is divided in 100 parts (100 procedures). There is
only one interrupt at the rate of inner most current loop to implement all three loops. For each



               CEP Course on Radar EW held at DLRL - 25th Oct to 29th Oct 2004
                                            IX-12
interrupt, complete current loop algorithm, 1/10th of velocity loop algorithm, and 1/100th of
position loop algorithm are executed. Continuous current limits and position limits are
implemented as part of position loop algorithm. This will avoid the unpredictable latency
problem of multiple interrupts.


The status of host communication receiver is polled at the rate of velocity loop sample rate. The
transmission to the host is done in the background. Single interrupt eliminates the saving and
restoring time of multiple interrupts and free from interrupt latency problems. The ‘C’ language
facilities of “pointers to functions” are used to implement the kernel.


4.5     Initialization:


       Since incremental encoder on motor shaft is being used as position sensor, initialization is
required on power on. A cam limit switch is used for marking the approximate location of zero.
On power the drive motor is given a constant voltage to move in CW direction, approximately at
maximum speed. While drive is moving the controller looks for operation of limit switch. After
the operation of limit switch the motor voltage is reduced such that it move slowly, while it is
moving slowly the DSP looks for reference pulse of incremental encoder on motor shaft. On the
occurrence of reference pulse the quadrature encoder interface unit is initialized to the set value
and it come to position mode of operation with the reference position as position command.


4.6     PWM Amplifier:


       A PWM amplifier from M/s APEX, USA (model no-SA 01) is used as voltage amplifier
to get the required voltage and power. For full details of the amplifier please refer Reference-6.


4.7     Man Machine Interface:


       A handheld personal computer is used to provide the man machine interface during the
development phase and in regular operation. The PC communicates with the servo processor


            CEP Course on Radar EW held at DLRL - 25th Oct to 29th Oct 2004
                                         IX-13
through RS-232 COM port. For receiving the data, the DSP polls the status at the sample rate of
velocity loop (1khz). The pre-determined data is stored in RAM once for every ‘A’ samples of
position loop where A is a parameter called “among values”. This way the data is stored in RAM
at the time intervals of A*10 msecs. After giving any command the storing of pre-determined
data continues until RAM allotted for this storage is available. The communication to host PC is
not in real time. The communication will be done at the rate determined by baud rate of
communication link. If rate of RAM storage, which is determined by the parameter “among
values”(A) is slower, the communication rate is slower. After transferring the complete data in
RAM, the communication is stopped. By changing the parameter “among values” the time of
recording can be changed.


        In the host side, the data is recorded in PC files. By using any graphical package like
MATLAB the graphical data can be displayed/ printed. The following commands are
implemented to operate the servo in different modes.


4.8     Modes of operation:


4.8.1   Position mode: In this mode the servo operates as position servo and moves to the
commanded position. In this mode the command position and feedback values are transmitted to
the host.


4.8.2 Velocity Mode In this mode the servo operates as velocity servo and moves with
commanded velocity. Commanded velocity and feedback velocity are displayed / stored.


4.8.3   Ramp Mode: In this mode, the servo operates as position servo and follows the
commanded ramp. The instantaneous values of position commands are calculated by DSP, to get
the required slope of commanded ramp. In this mode instantaneous command values and
feedback values are stored/communicated at the rate determined by “among values”.




            CEP Course on Radar EW held at DLRL - 25th Oct to 29th Oct 2004
                                         IX-14
4.8.4    Sinusoidal Mode: In this mode the servo moves in sinusoidal fashion. The period of
sinusoidal function is a configurable software parameter. The peak amplitude of sinusoidal
function is given along with command. In this mode also instantaneous position command and
feedback values are stored / communicated / displayed.


4.9      Testing and Test Results:


         To evaluate the servo performance, the following tests are conducted.


         1.    Position mode tests.
         2.    Ramp mode tests
         3.    Velocity mode tests.
         4.    Sinusoidal (acceleration) mode tests.
         5.    Backlash tests


         For evaluating the system performance, the required servo parameters like command
position or feed back velocity, are stored in the RAM in real time. The stored parameters are
communicated to host PC through RS 232 COM port. Observing PC data files and corresponding
graphs evaluate the performance. Depending upon the test the corresponding program should be
executed on the host PC.


4.9.1 Position mode tests


         Position mode tests are conducted for different step inputs and following parameters are
noted:


   1. Steady state position loop error.
   2. Max. Overshoot. Settling time to reach and settle within the accuracy of 0.1°.




              CEP Course on Radar EW held at DLRL - 25th Oct to 29th Oct 2004
                                           IX-15
One of position mode test results is given in figure below.



                                                                                                                P O S ITIO N S TE P C O M M A N D ( 10 D E G )

                                                                  12



                                                                  10
       COMMAND, FEEDBACK AND ERROR POS. (DEG)




                                                                                8



                                                                                6



                                                                                4



                                                                                2



                                                                                0
                                                                                    0              1000         2000                    3000                      4000          5000      6000


                                                                         -2
                                                                                                                            T IM E ( M IL LI S E C O N D S )




4.9.2 Ramp mode tests


       In this test, successive position commands are generated, to move in a ramp of defined
velocity and given at the input of position loop. The corresponding program “ramp test” will run
on host PC. Operator selects different ramps by executing the corresponding program in the PC.



                                                                                                                       Ram p Com m and (10 deg/sec)


                                                                                    25




                                                                                    20
                                       Command, Feedback, and Error*100 (deg)




                                                                                    15




                                                                                    10




                                                                                        5




                                                                                        0
                                                                                            0             500           1000                               1500          2000          2500


                                                                                     -5




                                                                                    -10
                                                                                                                            Tim e (m illi seconds)




                                                                                            CEP Course on Radar EW held at DLRL - 25th Oct to 29th Oct 2004
                                                                                                                         IX-16
4.9.3 Velocity mode tests


                                                     In this test a step velocity command is generated and given at the input of velocity loop.
The corresponding program “velocity test” program will run on host PC. Operator selects
different velocity inputs through the host PC. One of the velocity mode test results are given in
figure below.

                                                                           V E LO C ITY C O M M A N D (5 D E G /S E C )


                                                6




                                                5
   COMMAND, FEEDBACK, AND ERROR VEL.(DEG/SEC)




                                                4




                                                3




                                                2




                                                1




                                                0
                                                     0         200         400                      600                   800   1000   1200


                                                -1
                                                                                       T IM E (M ILL I S E C O N D S )




4.9.4 Acceleration tests:


                                                     A sinusoidal command is generated by DSP in real time and given at the input of position
loop. The amplitude and period of sinusoidal input are selected such that the maximum
acceleration of position command is10°/sec2 and the maximum velocity is 10°/sec.


                                                     The amplitude and period of sinusoidal input are selected such that the maximum
acceleration of position command is10°/sec2 and the maximum velocity is 10°/sec.


                                                     The period and amplitude of test input is selected by using the following equations to
generate the required maximum velocity and maximum acceleration.



                                                          CEP Course on Radar EW held at DLRL - 25th Oct to 29th Oct 2004
                                                                                       IX-17
Position command = A sin (wt)
Velocity = A w Cos(wt)


Acceleration = -A w2 sin (wt)

                                                                    S in u s o id a l T e s t (P e a k A m p litu d e 5 d e g , P e r io d 1 0 S e c )


                                               8



                                               6
     Command, Feedback, and Error*100 (Deg.)




                                               4



                                               2



                                               0
                                                    0        2000                 4000                        6000                     8000              10000   12000


                                               -2



                                               -4



                                               -6



                                               -8
                                                                                                   T im e (M illi S e c o n d s )



Therefore, Max. Velocity = Aw = 10º/sec
Max. Acceleration = -A w2 = 10º/ sec2
Amplitude of 10º and period of 1/(2*phi) (i.e., w=1) is one possible solution.
The results of sinusoidal test is given in figure below.


4.9.5 Backlash Test:


                                                The position sensor (incremental encoder) is fitted to motor shaft, to get very high
position resolution and good dynamic performance by avoiding non-linear backlash in the
closed loop. So the backlash should be added to the loop accuracy to get actual physical
accuracy of the system.


                                                To measure backlash, a laser pointer is fitted on load and its light spot is made to fall on
wall which is at a distance of 25 meters form the Pedestal. A position command is given to
move the pedestal by 0.5 deg in counter clockwise direction. Then give successive position
commands, in steps of 0.01 degrees, to move in clockwise direction and observe the movement



                                                        CEP Course on Radar EW held at DLRL - 25th Oct to 29th Oct 2004
                                                                                     IX-18
of light spot. Initially light spot will not move in clockwise direction due to the backlash in the
gear unit. Continue to increment the position command in steps of            0.01 degrees until a
movement of light spot is observed. The total of the increments of CW command will give the
backlash in the gear unit. Repeat the experiment to find out the backlash in CCW direction and
different positions.


       The maximum backlash observed = 0.12 degrees.


4.9.6 Summary of Test Results:
       1    Position mode tests:

       1.1 Steady position loop accuracy                     < 0.001deg
                                                             (See Note below )

       1.2 Maximum overshoot                                 < 0.75deg


       2.   Ramp tests:

       2.1 Steady state follow-up error                      < 0.01deg
         for ramp input of 10 deg / sec                      (See Note below )

       3    Velocity mode tests:

       3.1 Steady state velocity loop error for step input
         of 10 deg/sec                                     < 0.05deg/sec

       3.2 Maximum overshoot for velocity input              < 3.11deg/sec
          of 10 deg/sec

       4 Sinusoidal tests:

       4.1 Maximum loop error for sinusoidal input           < 0.01deg
         with max. velocity of 10 deg/sec and                (See Note below)
          max. acceleration of 10 deg/sec2

       5 Maximum Backlash                                    < 0.12 deg

Note: Backlash of the gear unit should be added to the actual physical accuracy of the system.



             CEP Course on Radar EW held at DLRL - 25th Oct to 29th Oct 2004
                                          IX-19
5.0   References:
      1. Principles of “ FEEDBACK CONTROL “ Vol.1 and Vol.2 by George Biernson.
      2. Servo Motor and Motion Control using DSP by Y. Dote.
      3. TMS320C24X DSP Manuals Vol. 1 and 2.
      4. Isolation and Control Comp. Designer’s Catalog by M/s. H.P
      5. APEX MICROTECHNOLOGY CORP. Product data book.




          CEP Course on Radar EW held at DLRL - 25th Oct to 29th Oct 2004
                                       IX-20
  CURRENT
  COMMAND
   VELOCITY
   COMMAND


                VEL.
                FEED
              FORWARD
                  4
                                    ACCLN.
                                 FEED FORWARD
                                       8

Set                            VEL.                  CUR.
                        +                     +                                                                 GEAR
Pos.   Prof        POS.        COMP. VEL.            COMP.            CUR.           POWER
                                           +        ++                                                MOTO    PEDESTAL
Cmd    Gen        LOOP              5 LOOP                 10         LOOP           STAGE
                                                                                                       R         &
       1       - COMPEN +         - COMPEN             - -           COMPEN            12
                                             CUR. CMD.                                                 15       LOAD
                    2   VEL. CMD.      7.                              11
                                             ADDER                                                               17
                        ADDER
                             5                   9                          CURRENT
                                                                            SENSOR
                                                                              14
                                                                                           CURRENT
  Track                                                                                     LOOP
  Pos. Cmd                                                                      VEL.
                                                                               SENSOR
                                                                                 16
                                                                                              VELOCITY LOOP
                                                        POSITION
                                                         SENSOR
                                                           18                                    POSITION
                                                                                                   LOOP

                                     Fig. 1 High performance servo with multiple loops




                         CEP Course on Radar EW held at DLRL - 25th Oct to 29th Oct 2004
                                                      IX-21
             CHAPTER – X


TECHNIQUES GENERATOR

Shri. N SRINIVAS RAO, SC‘E’
             &
   Shri. R ANAND, SC ‘D’




CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                            X
                          CHAPTER - X




                TECHNIQUES GENERATOR




                            CONTENTS




1.   INTRODUCTION


2.   ECM TECHNIQUES OVERVIEW


3.   TECHNIQUES GENERATOR


4.   ARCHITECTURE OF TECHNIQUES GENERATOR


5.   CONCLUSION


6.   REFERENCES




       CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                   X
                    TECHNIQUES GENERATOR


              Shri N Srinivas Rao, Sc‘E’ & Shri R Anand, Sc ‘D’
                                 DLRL, Hyderabd


1.0    Introduction
       This paper presents Techniques Generator unit of Electronic Counter
Measures (ECM) system. In present day battles,           ECM plays an important and
decisive role. Techniques Generator is heart of ECM system. The basic function of
Techniques Generator is to generate low power rf signal with suitable modulations to
counter enemy radars or sensors. This chapter gives brief overview of active ECM
techniques and requirements of          Techniques Generator for each technique,
implementation scheme and finally general architecture of Techniques Generator.


2.0    ECM Techniques Overview
       In literature we find number of ECM techniques and there are different
approaches to classify them in different ways. One way is to classify them from the
perspective of intended threat i.e., search radar or track radar ECM techniques.
Another approach is to classify them as active or passive. In this approach sub-
classification depends upon the jammer signal is radiated or not. Based on this
approach the techniques can be classified into Active ECM techniques and passive
ECM techniques.


       Active ECM techniques are those which involves radiation of electromagnetic
energy to deny the enemy full use of his radar or weapon system. Passive ECM
techniques involve no radiation of electromagnetic energy, but some action taken to
degrade the performance of enemy’s radar or weapon system.


       Active Jamming techniques can be broadly classified further into two
categories i.e., noise jamming and deception jamming. Noise jamming aims at
masking the signal at victim radar whereas the deception jamming technique tries to
fool the victim radar receiver processing circuits.        There are number of active




           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                      X-1
       jamming techniques described in the literature. However we have made an attempt to
       cover few important active ECM techniques in this paper as shown below.

                 Active ECM Techniques




Noise Jamming                                   Deception Jamming
  •   Spot
  •   Barrage
  •   Swept Spot
                             Range Deception                             Angle Deception
  •   Doppler Noise
                                • RGPO/RGPI                                •     Inverse Gain
                                • VGPO/VGPI                                •     Scan Rate
                                • False Targets                            •     Swept Scan Rate
                                                                           •     Blink
                                                                           •     Cross Eye




       The effect of noise and deception jamming techniques on Radar is shown below.


             Noise                                                         Deception
             Jamming                                                       Jamming

                                              Radar


         Increase in False Alarm rate                   Increase in Range Errors
         Data Saturation                                Missile Miss distance due to Range
         Loss of Target                                 Error
                                                        Reacquisition delay in Range
                                                        Increase in Angular Errors
                                                        Break Track and reacquisition delay
                                                        Missile Miss distance due to Angular
                                                        Error




                  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                             X-2
3.0     Techniques Generator
        Primary role of Techniques Generator is to generate modulating signals
required for the techniques. In following sections, active ECM techniques mentioned
in previous section are discussed and requirements of generation of signals are
brought out.


3.1     Noise Jamming Techniques
        The basic principle of noise jamming techniques is to transmit noise like
signal which will mask wanted target echo signal from a radar. Aim of jammer is to
maximize the noise power in radar receiver bandwidth. Effectiveness of noise
jamming in degrading performance of receiver depends upon relative strength of
wanted signal to jamming signal and also on the quality of the jamming noise. Ideally
jammer noise should have uniform spectral density and guassian amplitude
distribution. This shall make jammer noise signal to look like receiver thermal noise
and it will be difficult for radar to reject jammer signal. The aspects to be considered
in noise jamming techniques are tuning accuracy of jammer signal, jammer
bandwidths and quality of the noise.


        Based on jammer bandwidth and receiver bandwidth, noise jamming
techniques can broadly divided into following types.


3.1.1   Spot Noise Jamming Technique
        In this technique jammer concentrates it’s power into narrow band centered on
the frequency of victim radar, typically here jammer noise bandwidth equals or
exceeds the radar bandwidth. See figure 1. This is the most effective technique against
fixed frequency radars provided jammer is capable of radiating correctly at victim
radar frequency.


3.1.2   Barrage Noise Jamming Technique
        When jammer spreads it’s power over a wide bandwidth, it is called as
Barrage jamming technique. See figure 2. As a result jammer will be able to handle
frequency agile radars. This technique is simple and does not require accurate
frequency tuning of jammer signal. The penalty paid by jammer is in terms of



            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                       X-3
         dilution of jammer signal power density that can be injected into victim radar
         receiver.


         3.1.3   Swept Spot Noise Jamming Technique
                 Swept Spot jamming technique attempts to provide jamming power density of
         Spot jamming and wide bandwidth coverage of Barrage jammer. In this technique a
         spot noise signal is swept rapidly to and fro across wide band of frequencies of
         interest. This technique will be useful to counter set of Radars, present in the same
         direction (with respect to Jammer) and are operating in the same frequency band. If
         bandwidths and sweep rates are correctly chosen, this technique can be very effective.


         3.1.4 Doppler Band Noise Jamming Technique
                 Noise can be used against a doppler radar, with noise bandwidth just covering
         all the doppler cells. Noise bandwidth required for this purpose is in order of tens of
         KHz. However in most cases, noise can have bandwidth equal to doppler radar’s IF
         bandwidth. Doppler Noise jamming disrupts doppler processing circuits of radar.




                                                                                    Radar
                             Receiver                            f1     f2   f3     Frequencies
Power                        Bandwidth            Power
Level                                             Level

                              Frequency                                             Frequency

                                Jammer
                                                                      Jammer
                                Bandwidth
                                                  Power               Bandwidth
Power
Level                                             Level

                              Frequency                                             Frequency

        Figure 1. Spot Noise Jamming                      Figure 2. Barrage Noise Jamming



         3.1.5   Noise Jamming Techniques Implementation
                 The most important consideration in noise jamming techniques is quality of
         noise waveform. As already mentioned, jammer noise signal should have white
         guassian distribution so that it resembles receiver thermal noise. This can be

                     CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                                X-4
generated by direct noise amplification (DINA) or frequency modulation of carrier. In
direct noise amplification (DINA) approach, thermal noise over bandwidth of interest
is amplified to the required power level and transmitted. But due to peak power
limitations of wide band amplifiers, quality of noise is degraded.


        Other elegant approach is to frequency modulation of the carrier signal by
ramp signal added with noise. The ramp signal produces discrete spectrum. To get
continuous spectrum, which is desirable, noise is added to the ramp. The bandwidth of
noise added should be greater than half of the repetition rate of the ramp so that
continuous spectrum will be obtained.


        Certain principles can be applied for generating good quality of noise.
Frequency modulation deviation (i.e. bandwidth of the jammer signal) performs dual
role of creating amplitude noise within Radar Receiver and allowing a tolerance for
the error in tuning of ECM carrier frequency to that of Radar. To create large
amplitude variations as viewed by the Radar receiver, the bandwidth of jammer signal
must exceed the instantaneous bandwidth of the Radar. Bandwidth of two to three
times of Radar receiver’s instantaneous bandwidth is adequate because of steep filter
skirts/role off in the Radar receiver’s front end filter.


        Each time jammer signal sweeps across the Radar receiver’s instantaneous
bandwidth an impulse will be produced. Repetition rate of Ramp signal determines
the number of impulses. Too high repetition rate (when compared instantaneous
bandwidth of Radar Receiver) will produce more impulses, whose energy content will
be very low. Too low rate will produce less number of impulses, but their energy
content will be high. Normally frequency of the Ramp signal shall be of the order of
the instantaneous bandwidth of the Radar receiver.


        Block diagram of Techniques Generator hardware for Spot, Barrage and
Swept Spot noise jamming techniques is given figure 3. A tunable RF source such as
Voltage Controlled (VCO) or Digital Tuned Oscillator (DTO) is used to generate
carrier signal of required frequency and is frequency modulated with ramp signal
added with noise. By controlling amplitude and frequency of ramp, noise amplitude
and noise bandwidth it is possible to get required noise bandwidth. For swept spot

            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                       X-5
      noise jamming, modulating signal is applied to DTO and in addition to this frequency
      of DTO is varied in ramp fashion to get required swept spot jammer bandwidths.
      Output of DTO is amplified to suitable level and then radiated. Parameters required
      for the techniques can be loaded from the controller.


             For spot noise jamming it is required to tune rf source accurately to radar
      frequency. So for this purpose a suitable receiver is used to measure the frequency
      difference between the jammer rf        source and radar transmission and apply this
      correction to jammer tunable rf source.




                                                                Modulating Signal

 Ramp Frequency        &
 Amplitude
                                                                                             Jammer
                                                                                             Signal
                                                        Frequency Data
       Ramp
      Generator

                                  Adder                  DTO                     Amplifier
       Noise
      Generator




Noise Amplitude & Bandwidth



           Figure 3. Block Diagram of Noise Jamming Techniques Generator




                  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                             X-6
3.2     Deception Jamming Techniques
        Deception Jamming Techniques tries to reproduce replica of target’s echo at
radar receiver with suitable modulations to deceive the radar in range or in angle or in
doppler frequency or combinations of these. These techniques require less
jammer to signal (J/S) ratios when compared with noise jamming techniques.
Deception    jamming techniques can be broadly classified into Range Deception
techniques and Angle Deception techniques.


3.2.1   Range Deception Jamming Techniques
        Range deception jamming techniques aim at inducing false range information
into radar circuits. Under this category, Range Gate Pull Off (RGPO), Range Gate
Pull In (RGPI) and False Targets jamming techniques are discussed. Velocity Gate
Pull Off (VGPO) is also grouped under this category.


3.2.1.1 Range Gate Pull Off (RGPO) Jamming Technique
        RGPO technique is self protection jammer technique. RGPO aims at pulling
off the range tracking gates of radar receiver from return skin echo. In this technique
Jammer transmits false pulse which is in sync target skin return initially and then
false pulse is delayed gradually. Since jammer’s signal is stronger than skin echo,
the radar receiver’s range gates move away from echo position to center of jammer
signal and follow the jammer signal. Refer to figure 4. With the result radar starts
getting false range information. Jammer pulls off range gates to a suitable range and
then either switches off the transmission or starts a fresh cycle or follows up with
angle modulation technique. When          jammer switches off transmission, the radar
receiver is forced to go in for range search mode. The most important considerations
in RGPO technique are minimum throughput response time, range pull off rates and
pull off range. Ideally speaking the minimum throughput response time should be
zero or less than few tens of nano seconds to prevent radar receiver from
distinguishing jammer signal from the skin echo. Pull-off rates should fall with
bandwidth of range tracking circuits and shall support both linear and non linear
profiles. Pull off range should be atleast 3 to 4 times radar pulse width so that at the
end of pull off program, jammer signal falls outside radar receiver’s range gates.


            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                       X-7
     Radar Signal

                                                                            Target
                Skin Echo

                                                     Receiver
           Jammer Signal                             Antenna
                                     T1

                                      T2
                                                      Transmit              Jammer
                                                      Antenna
                                      T3



  Figure 4. Skin Echo & Jammer Signal Timing at Jammer for RGPO Technique


          Range Gate Pull In (RGPI) technique has evolved due to Electronic Counter
Counter Measures (ECCM) applied against RGPO technique using lead edge range
tracking by radar. In this jammer pulse precedes the skin echo and it can usually
employed against radars operating with fixed PRIs.


3.2.1.2          False Targets Jamming Technique
          alse Targets jamming technique is similar to RGPO/RGPI and is used against
Search Radar to create number of false targets. In this technique, jammer receives the
radar transmission and triggers a jammer transmission with frequency and pulse width
same as radar transmission but delayed in time. Generally a burst of pulses spaced at
different intervals is released in response to radar transmission whose position may
changing to simulate moving targets. This techniques presents false targets on PPI and
gives illusion that there are number of false targets. Consideration for this technique
are number of false targets, maximum range and separation in angle.


3.2.1.3          Velocity Gate Pull Off (VGPO) :
          elocity Gate Pull Off is deception jamming technique used against radars
using doppler principle to separate signals from moving targets and the stationary
targets and also doppler frequency shift of return signals to process signals using
doppler filters or velocity gates. Doppler frequency shift of return echo depends upon
the relative velocity of the target and radar frequency. Figure 5. gives frequency

             CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                        X-8
      contents of return signal. Radar uses bandpass frequency filter to process the target
      returns. This frequency filter is called velocity gate which is tunable to track the
      doppler shift changes due to target accelerations.




                         Ground Returns                     Target Returns
Received
Signal Power



                                           Frequency

           Figure 5. Frequency Spectrum of Radar Received Signal

               The concept of Velocity Gate Pull Off (VGPO) jamming technique is similar
      to Range Gate Pull Off but it is in frequency domain. Jammer receives the radar
      signal, amplifies and retransmits it, or an exact replica with little or no change in
      frequency to ensure that jammer signal is accepted by doppler filter of radar receiver
      processing the skin echo. Since jammer signal is stronger than skin echo, it captures
      the velocity gate. Jammer then progressively shifts frequency of it’s signal pulling off
      velocity gate from the target’s skin echo. Figure 6 gives jammer responses during
      VGPO technique. When doppler frequency shift is positive it is called VGPO and
      when it is negative it is referred as Velocity Gate Pull In (VGPI).




                 Velocity Gate
                                      Initial Jammer Signal

    Jammer                            t1       t2      t3
    Signal
    Amplitude




                                                                         Frequency
                                   Skin Echo

                                 Figure 6. Velocity Gate Pull Off



                  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                             X-9
          The issues concerned with VGPO technique are initial frequency shift, rate of
frequency shift and profile used for shift. Once frequency shift reaches the maximum
doppler shift accepted by radar, jammer does one of following


i)        Simply restart the cycle causing the radar to return to reacquisition mode with
          consequent loss of tracking data.
ii)       Radiate a hook signal to hold radar velocity gate at certain doppler shift.
iii)      Follow with angle jamming techniques to break angle tracking loop.
          In general RGPO and VGPO are used in combination. Doppler frequency shift
is made proportional to RGPO pulse velocity.


3.2.1.4          Range Deception Jamming Techniques Implementation
          Block diagram of deception jamming techniques generation is given in figure
7. For RGPO and VGPO techniques incoming signal is stored in RF memory. RF
storage unit can be digital RF memory or analog delay line based memory or fiber
optic based delay line memory. Trigger from RF storage is used to start the delay
pulse generator which generates RGPO pulse delayed with respect the trigger and
having same pulse width. Both pulse width and delay are programmable parameters
and can be loaded from controller/processor. Issues to be taken in design are
propagation delay, resolution of delay, maximum delay and minimum pulse width.


          Doppler frequency shift can be achieved by either AM-SSB modulation with
tone, or serodyning of TWT amplifier or using digital phase translator. Digital phase
translator is a good candidate for realising doppler frequency shift. Here phase of
incoming signal is varied over 0 to 360 degrees and slope or rate of phase variation
will decided frequency shift induced in the rf signal. Refer to figure 7. Phase
translator controls waveform can be easily generated with help of digital counters.
Phase Controls Generator consists of two blocks, i.e., programmable clock generation
module based on counters whose output is given to free running counter. The free
running counter outputs form controls for digital phase translator. When counter is
incrementing, it results in a positive frequency shift and it decrements it results in
negative frequency shift. By programming the clock frequency one can get the
required doppler frequency shift.


              CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                        X - 10
                Data for range delay, frequency shift for different velocity profiles can be
        computed by controller and data is loaded


        3.2.2             Angle Deception Jamming Techniques
                Angle deception jamming techniques induce false angle information into angle
        tracking circuits of radar. Angle tracking techniques can be divided into two broad
        classes i.e., sequential lobing/conscan and monopulse tracking.
    Receiver                                                                                      Transmitter
    Antenna                                                                                       Antenna



                     RF                     Phase                   RF                   Amplifier
                   Storage                Translator               Switch


                          Trigger


                 Delay Pulse                        RGPO Pulse
                  Generator
                                                 Digital Phase Ramp
.
      Pulse Width,
      Range Delay                          Phase
                                          Controls
                                          Generator

                                                       Sweep Time &
                                                       Slope (Sign)
                Figure 7. Block Diagram of Range Deception Techniques Generator


                       3600
                  Phase
                  (Deg)

                          00
                                         Ts                    Time
                                                                                Ts : Sweep Time
                  Doppler                                                       Fs : Frequency Shift = 1 / Ts
                  Shift




                  Figure 8. Phase Translator Controls & Doppler Shift
                     CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                               X - 11
3.2.2.1          Conical Scan Angle Tracking Radar Jamming Techniques
          A sequential lobing radar derives it’s angular tracking error of target by
switching antenna beam in certain positions and processes amplitude and phase of
received signals. A conical scanning radar is extension of sequential lobing radar
wherein radar antenna beam is rotated at certain rate. It derives target angle error
information from the amplitude and phase modulation caused by its offset mutating
beam as it scans around the radar target boresight. A jammer signal for this radar
should have modulations at Conscan or lobing rate but with a false amplitude and
phase information to produce errors in radar tracking error loop. In addition to this
these radars are susceptible to changes in received signal strength, so jammer can
degrade tracking performance by doing on/off modulation at slow rate effecting the
AGC of receiver.


          Jamming techniques employed against the Conscan radar are Inverse Gain,
Scan Rate Modulation, Swept Scan Modulation,. On/Off modulation and so on.
These techniques differ based on type of modulating signal, synchronization with
victim radar’s scan frequency.


          In Inverse Gain jamming technique, jammer detects the radar scanning pattern
and use this waveform information to amplitude modulate it’s own pulse train with
1800 out of phase with Conscan envelope. The 1800 phase shift produces maximum
error in radar tracking error loop. Conscan AM waveform is nominally sinusoidal but
using a square wave is easier and advantageous.


          Scan Rate Modulation technique based on scan frequency measurement or
prior information from Electronic Support Measures (ESM) system, modulates the
jammer pulses with square waveform with frequency variation at scan frequency and
rate of scan frequency variation is in order of servo bandwidth. Since tracking errors
are centered around scan frequency, this technique induces the noisy errors in radar
angle tracking circuits.


          Swept Scan Rate Jamming technique is used in situations where scan
frequency information is not available i.e., when jammer is handling Conscan Receive
Only (COSRO) radars, In this case jammer sweeps frequency of on/off modulation

              CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                        X - 12
       over jammer pulses, over the likely range of victim radar’s scan rate. Jammer sweeps
       the frequency of squarewave slowly and look for variations for victim radar’s signals
       received during off periods of squarewave modulation. As jammer’s scan modulation
       frequency approaches victim radar’s scan frequency, jammer signal effects the
       tracking circuits of radar. Jammer senses this then stops slow frequency sweep and
       switches over to scan rate modulation at this squarewave frequency.


              Conscan Trackers are susceptible to variations in received signal strength.
       On/Off modulation at very slow rates i.e., in order of AGC bandwidths will effect the
       AGC functioning of victim receiver and this in turn effects the recovered conscan
       modulation. This degrades the angle tracking performance of radar. Figure 9 gives
       signals at radar end for Inverse gain square waveform jamming technique. Figure 10
       give scan frequency variations for SRM and SSRM techniques. Difference of
       parameters for scan jamming techniques is given in table 1.

                                                         Conical Scan Radar Response


Amplitude




                                                  Time
                                                                Jammer Responses

 Amplitude



                                                  Time


                                                           Combined Result At Radar


Amplitude




                                               Time

             Figure 9. Inverse Square Wave Jamming of Conical Scanning Radar
                  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                            X - 13
                                                                  Scan Step Dwell Time
                           Scan Time


              Scan
                                                                        Scan
              Freq
                                                                        Dev



                                                             Time

                        Figure 10. SRM/SSRM Scan Waveforms




                                            Table - 1


                                 Scan Modulation Parameters


Parameter               Inverse Gain        SRM                             SSRM
Phase w.r.t Conscan 180 degrees             No Synchronization              No Synchronization
Frequency
Scan Deviation          Not Applicable      10%         of          Scan Possible        range     of
                                            Frequency                       Conscan frequencies
                                                                            for class of radars
Scan Frequency Step     Not Applicable      10%         of          Scan 4 to 5 Hz
                                            Deviation
Scan   Step      Dwell Not Applicable       Tens of milli seconds           Hundreds      of     milli
Time                                                                        seconds
Scan Cycle Time         Not Applicable      Hundred          of     milli Seconds and depends
                                            seconds.                        upon scan band to be
                                                                            covered.




                   CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                             X - 14
       3.2.2.1.2        Conical Scan Radar Jamming Technique Implementation Scheme
              The figure 11 gives implementation scheme for deception jamming techniques
       against the Conscan Radar. The incoming signal is stored in RF memory and
       retransmitted. The sensor output is processed by processor to measure the conscan
       frequency and phase to          generate modulating signal required for Inverse Gain
       jamming technique. In case of SRM and SSRM techniques, processor does the
       necessary frequency sweep of square waveform. In case of SSRM technique
       processor does necessary jog detection and identifies the scan frequency of victim
       radar. If square wave is used as modulating signal then, RF modulator can be a simple
       rf switch.




Receiver                                                                            Transmitter
Antenna                                                                             Antenna




                      RF                      RF                         Amplifier
                    Storage                 Modulator




                     Sensor




                    Processor              Waveform
                                           Generator



   Figure 11. Block Diagram of ConScan Angle Deception Techniques Generator




                     CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                               X - 15
3.2.2.2          Monopulse Angle Tracking Radar Jamming Techniques
          In a mono pulse tracking radar system, an angular error estimate is made on
each return pulse. The amplitude comparison type involves the formation of multiple
antenna beams simultaneously in space where each beam is squinted in angle from an
adjacent beam. The angular measurement is made, by comparing the power received
in overlapping adjacent beams. An angular error is indicated by the difference in
amplitude of the signals returned in the adjacent beams. Monopulse trackers
maximize the information rate on the target and are immune to pulse to pulse
amplitude variations. So techniques employed against conscan radars are ineffective
against monopulse trackers.


          A number of deception jamming techniques are possible and can be grouped
into two major categories depending upon whether they are exploiting weakness of
implementation or they are exploiting the fundamental weaknesses of monopulse
technique itself. Techniques such as skirt jamming, image jamming and cross
polarization techniques tries to exploit weakness of implementation may not be very
effective against all monopulse trackers. Cross Eye, Terrain or ground bounce,
blinking or formation jamming techniques attack the basic weakness of monopulse
tracking systems. Hence we have taken up Cross-eye and blink jamming techniques
for the discussion.


3.2.2.2.1        Cross Eye Jamming Technique
          Cross-eye jamming technique generates artificial glint (an apparent change of
target direction as viewed from the radar) into the mono-pulse tracking loop. This
technique is implemented by using pair of repeater chains which are coherently
related. The signals received in each antenna are repeated in the other antenna, except
for a 180 degree phase shift introduced in one line. Antennas should be as far as
possible. The signals from the two repeaters will be 180 degrees out off phase when
they reach radar’s tracking antennas, independent of the direction to the radar. This
causes null in combined response of the radar’s sensors just where the radar tracking
circuit would expect a peak. This results in distortion of wave front at the radar




              CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                        X - 16
     resulting in generation of false tracking error signal. Figure 11 shows the
     implementation and wavefront for this technique.


              There are two     important requirements to be met by             jammer for this
     technique. One is that the electrical lengths of two repeater paths have to be matched
     very close i.e., in other terms the phase error between channels should be less than 5
     degrees. Second requirement is that it needs very high jamming to signal ratio (in the
     order of 20 dB).

                           To Radar


                                       Wave Front




                              Receiver                       Transmitter
Transmitter                   Antenna                        Antenna                       Receiver
Antenna                                                                                    Antenna
                                     1800                   Amplifier
                                  Phase Shift



                                            Amplifier




                 Figure 11. Block Diagram of Cross Eye Jamming Techniques Generation



               3.2.2.2.2     Blink Jamming Technique
              Blink jamming attempts to attack the tracking dynamics of the mono-pulse
     angle tracking radar. This method employs multiple spatially dispersed jammers and
     turns on one jammer at a time at a rate which is within the pass band of the angle
     tracking servo of radar. The angle servo will tend to average the signals from the
     various jamming sources, resulting in the radar antenna to track a displaced centroid,
     which corresponds to the angular center of mass of the jammers within its main lobe
     antenna response. Each jammer can a repeater based working in co-ordination with
     other jammer to synchronize on/off periods.


                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                           X - 17
            The critical parameter with blink jamming is the rate at which the jamming
   resources are commutated. Too high rate causes tracking radar to average the data and
   tends to minimize the tracking error. Too low rate allows the tracker to determine
   accurately the angular position of each jammer. Rate should be in order of tracking
   bandwidth of victim radar.


   4.0      Architecture of Techniques Generator
            The figure 12. gives general architecture of Techniques Generator. This
   architecture is based on combination of processor, digital, analog and RF modules.
   Processor interacts ECM processor to get radar parameters and decides the techniques
   to be employed against the radar. It controls the modules to generate modulating
   signals with required profiles and jam cycle times.           Waveform generation block
   generates waveforms required for different techniques. This block can realised with
   combination of analog and digital hardware. Today we have high density
   programmable logic devices available, with which one can realize a compact
   waveform generator with flexibility to incorporate modifications. Digital Signal
   Processor (DSP) can also be used for this purpose. Control & Timing block generates
   signals required for handling multi threats and jammer and lookthrough periods and
   can be realised with digital logic.
                                              RF From
                                              Repeater




               RF                Tunable RF                Modulators
                                 Sources &                  & Level              To Amplifier
               Input
                                 Receiver $                 Control



Processor                        Waveform
                                 Generation



                                                         $ Note : Receiver for frequency
                                 Control &               correction for spot noise jamming
                                  Timing
ECM Processor

            Figure 12. Architecture of Techniques Generator
                CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                          X - 18
         Modulators shall support amplitude and phase modulations with provision to
control power level. It has necessary rf routing network to select signals from repeater
for deception jamming techniques and different modulators.


         Actual configuration of Techniques Generator depends upon ECM system
requirements and needs good understanding of radar working and design skills in field
of microprocessors, digital and analog logic design, rf components and modulators.


5.0      Conclusion
         Techniques Generator requirements for ECM active techniques are discussed
in this paper. It can seen from the discussion, development of techniques generators
needs good knowledge about radar functioning. In addition to this present day ECM
system requirements puts high demands on Techniques Generator to be compact,
easily configurable and support for different jamming techniques. Today ECM
designer has excellent options available to him in terms of high speed processors,
programmable logic devices and high speed analog/digital circuits to develop
techniques generator which meets the ECM system requirements.


6.0      References
1. Air Vice-Marshal J P R Bowne, Wing Commander M.T. Thurbon, “Electronic
      Warfare”, Brassey’s U.K., 1998.
2. D. Curtis Schleher, “Introduction To Electronic Warfare”, Artech House, USA,
      1986.
3. D. Curtis Schleher, “Electronic Warfare in the Information Age”, Artech House,
      USA, 1999.
4. Richard J. Wiegand, “Radar Electronic Countermeasures System Design”, Artech
      House, USA, 1991
5. August Golden JR., “Radar Electronic Warfare”, American Institute of
      Aeronautics and Astronautic Inc., USA, 1987.
6. Lessons 7 to 12 of EW101 Series, Journal of Electronic Defense.




              CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                        X - 19
                  CHAPTER – XI


    HIGH POWER MICROWAVE
       TRANSMITTERS FOR
ELECTRONIC ATTACK (EA) SYSTEMS
   Shri. Y. GOPALA KRISHNA SC ‘E’,
                 &
    Shri. NIRANJAN PRASAD, SC’D’




  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                              XI
                         CHAPTER - XI


HIGH POWER MICROWAVE TRANSMITTERS FOR
     ELECTRONIC ATTACK (EA) SYSTEMS

                          CONTENTS


1. INTRODUCTION


2. ROLE OF NON DESTRUCTIVE EA SYSTEMS


3. THE ESSENTIAL FEATURES OF HIGH POWER
  TRANSMITTER (HPT)


4. CONFIGURATION OF HIGH POWER AMPLIFIER (HPA)


5. ELECTRONICALLY BEAM STEERING EA TRANSMITTER


6. CURRENT TRENDS IN EA TRANSMITTERS


7. CONCLUSIONS


8. REFERENCES




    CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                XI
                High Power Microwave Transmitters for
                    Electronic Attack (EA) Systems
                  (Smt. G. Jayasri, Sc ‘E’, Y. Gopala Krishna Sc ‘E’,
                             and Niranjan Prasad, Sc’D’)
                                 DLRL, Hyderabad


1.0    Introduction:

       The high power microwave transmitter is one of the major sub-system of
Electronic Attack (EA) System used for the Electronic Warfare (EW) applications. The
EA System is used to degrade, deny or neutralize the effective use of electromagnetic
signals by adversaries with intention of destroying the combating capability. These are
soft-kill and hard-kill. Soft kill involves non-destructive action, while hard kill is a
destructive action. Jamming is a soft kill action that attempts to dilute the effectiveness of
enemy weapon system through confusion or deception techniques. Soft kill EA Systems
are generally economical, have inherent multi-threat capability and non-lethal. The
destructive capability of EA is a new concept in EW. Modern EW systems use lasers or
Directed Energy Weapons (DEW) to either destroy or disable enemy’s electronic
equipments. The various types of EA systems used in Radar-EW are on-board microwave
jammers, IR and electro-optical spectral bands and off-board jammers such as passive
and active decoys. The presently used high power microwave ECM transmitters and their
future trends in EW applications are explored in this paper.


2.0    Role of Non Destructive EA Systems :


       There are three types of non-destructive EA System in worldwide use. They are
stand-off, escort and self-protection jammers.


       Stand-off jammer (SOJ) is positioned away from the lethal range of the threat
weapon system and is capable of jamming the threat through its side lobes while
protecting its mission platform. Since SOJ jams from longer ranges and through side



            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                      XI - 1
lobes, there will be a need of high Effective Radiated Power (ERP) in order to produce
the required jammer power to signal power (J/S) ratio.


       Escort-jammer (EJ) is positioned on the platform going on mission along with
the fleet/convoy and it protects them by jamming the threat through its main lobe. The
ERP requirement in this case is moderate as it jams through the main lobe of the threat.


       Self Protection Jammer (SPJ) is positioned on the mission platform and protects
its own platform by jamming through main lobe of the victim radar. The required ERP is
less as compared to SOJ case.


       With the increase in the complexities of the threats scenario, the EA system
should be capable of handling multiple threats operating in various frequency bands
covering 360° in azimuth and engages them in short time. In order to meet the objectives
of EA systems, transmitters play crucial role and some of present day systems are
described below.


       The modern radars with high ERP force the ECM transmitters of EA Systems to
generate large jamming powers to counter victim radars with an appropriate J/S ratio.
This calls for high power output TWT amplifiers with smaller antenna beamwidths (for
more directive gain).


3.0    The Essential Features of High Power Transmitter (HPT):


       The transmitter consists of a high power amplifier, an antenna and a servo-driven
pedestal. The essential features of HPT are as follows,
               a)       High ERP
               b)       Broad Bandwidth
               c)       Multi-threat engagement capability and
               d)       Less reaction time


           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     XI - 2
3.1 High ERP:


       The Effective Radiate Power (ERP) requirement of an ECM system mainly
depends upon the role of the jammer (Stand off or Self Protection) engaged in the
mission. The ERP requirement is also dependent on the ERP of the victim radar, RCS of
the target, required J/S ratio, distance between the jammer and the radar and finally on
the distance between the radar and the target. ERP is the product of the TWT output
power (Pj) and the jammer antenna Gain (Gj). The ERP required for a typical Stand Off
Jammer is given by the equation

                 ( Pj + Gj )dBm = -71 + Pt + 2*Gtr – Grj + 20 log (Dj) + 10 log (σ) –40 log (Dt)
                                                                                             (1)


       Where, Gtr is the gain of the radar transmitting antenna towards target, Grj is the
       gain of the radar antenna (side lobe gain) towards jammer antenna, Dj is the
       jammer to radar distance and Dt is the distance between the radar and the target.


Depending on the jammer platform conditions and the system requirements, the
transmitters are basically configured into following three types,


       (i)        The transmitter configured with high power amplifier and a broad beam
                  antenna is less in weight, provides moderate ERP. It is used on airborne
                  platform.


       (ii)       The transmitter configured with high power amplifier, a narrow beam
                  antenna and a servo driven pedestal provides high ERP. The limitation of
                  servo driven transmitters is that these can handle the threats lying within
                  the beamwidth with high reaction time (approximately 5 seconds). Used
                  on naval and ground-based platforms.


       (iii)      The transmitters configured with lens fed array meets the requirements of
                  high ERP and handles simultaneous multi-threats and engages them in a

               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         XI - 3
               short time. This is the latest trend in the development of EA transmitters.
               Used on airborne and ship borne platforms.

3.2 Broad Bandwidth:


       The EA System needs to counter various narrowband or wideband radars
covering an octave or multi-octave bands. Helix TWT is a non-resonant device and is
available with more than one octave bandwidth. These are suitable for EA transmitters.
A high power CW TWT with grid/focus electrode control can be used for both CW and
Pulsed applications.

3.3 Multi-threat Handling Capability:


   With the increasing complexity of the threat scenario, there is a need to design an
ECM transmitter that provides high ERP, handles simultaneous multi-threats, operates in
broadband and engages the threat in a short time (less reaction time). Lens fed multi-
beam transmitter can provide the multi-threat handling capability with less reaction time.


3.4 Reaction Time:


   Reaction time is defined as the time duration from the interception of the threat signal
to the engagement of jammer on the threat. In multi-threat environment, it is required to
switch the beam between multiple threats spread over 360° in azimuth on pulse by pulse
basis or within Pulse Repetition Interval (PRI). The servo-driven pedestal used for
positioning the jammer beam on the threat has limitations of handling multiple threats
instantaneously due to the mechanical motion of the servo pedestal. Therefore, inertia-
less beam scanners like electronically steerable lens fed multi-beam transmitter is better
suited for engaging the faster moving threats at a lesser reaction time.




           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     XI - 4
4.0 Configuration of High Power Amplifier (HPA)


         The block diagram of conventional high power amplifier is shown in Fig.1. The
amplifier is associated with High Voltage Power Supply and modulator. The functional
description of these sub-systems is given below.

4.1 Solid State driver Amplifier (SSA):


         The low power RF signal from Synthetic Channel/ Repeater Channel is
modulated with appropriate jamming technique and fed to the input of solid-state driver
amplifier. The SSA amplifies this signal to drive the TWT.


4.2 High Power TWT:


         Helix is used a slow wave structure. Since, it is a non-resonant device, it provides
broadband operation. For EA applications, high power helix TWT is used as a high
power amplifier. The main elements of the helix TWT are Electron Gun, Slow Wave
Structure and Collector. RF amplification in the TWT takes place due to continuous
interaction between the RF wave and the electron beam when they are near
synchronization. When the input RF wave travels along the helix slow wave structure, its
velocity slows down to slightly less than that of the electron beam passing through the
center of slow wave structure. The axial alternating RF field, caused by the propagation
of RF wave along the slow wave structure, modulates the velocity of electron beam. The
alternating nature of RF field generates alternative forces on electrons, so that some
electrons are accelerated and others are retarded. The velocity modulation of the electrons
causes density modulation, which gives rise to bunching of electrons. The retarded
electrons transfer some of its kinetic energy to the RF field and causes amplification of
the RF wave. This process continues along the length of the helix, where RF signal builds
up, thus achieving RF amplification in the TWT. The block diagram of TWT is shown in
Fig.2.



             CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                       XI - 5
4.3 High voltage Power Supply (HVPS):


         The High Voltage Power Supply (HVPS) generates the required TWT electrode
voltages and also providing necessary protection and control. All the electrode voltages
are generated using high frequency DC-DC converters. The HVPS is normally built using
two separate sections, Low Voltage (LV) and High Voltage (HV) section.


         The Low Voltage section generates regulated DC voltages for the heater inverter,
the focus inverter and isolated pulsewidth modulated square wave drive to HV inverter. It
also provides the regulated voltages to the control, protection, interlock and status
indication circuits. The functional block diagram of HVPS unit is shown in Fig. 3.


         The HV section is isolated from LV section by means of isolation transformer and
HV inverter transformer. The High voltage section generates heater and focus electrode
voltages by means of push-pull inverter and collector and cathode voltages by means of
half bridge resonant inverter. The purpose of this inverter is to convert unregulated DC
into high frequency AC signals, which drives the step-up transformer. The outputs of
step-up transformer are then rectified, filtered and regulated to provide the required high
voltage DC supply to the TWT electrodes.


         Since, HVPS plays very crucial role in the design of a HPA, the indigenous
development support of this unit has made an important achievement in the development
of EA.

4.4 Modulator:


         Electron Beam in the TWT is controlled with a control electrode, such as an
anode, a grid or a focus electrode. Modulators are developed with fast switching HV
FETs. With this type of modulator, it has overcome the drawbacks of pulse drooping and
duty cycle limitations. Present modulators are able to produce control voltages with




            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                      XI - 6
100nsec pulse width to CW with fast rise and fall times, providing less throughput delays.
Because of this, a CW TWT can be used for both CW and pulsed applications.

4.5 HPA Operation:


       The High Power TWT is mounted on a cooling system to maintain the TWT base
plate temperature at 100°C. The TWT is integrated with the High Voltage Power Supply
(HVPS) and modulator. The transmitting beam is controlled, based on the command from
the ECM Processor. Solid State Amplifier is used to amplify low power jamming signal
to drive the TWT into saturation for achieving maximum power. High Power TWT
output is fed to the antenna through Dual Directional Coupler (DDC) for RF power
radiation into space in the desired direction. The DDC forward coupled port is used for
power monitoring and reverse coupled port is used to sense the reflected power and
protect the transmitter from the mismatched loads. The block diagram of interfacing
between TWT and HVPS and Modulator unit is shown in Fig.4. The photographs of the
developed high power TWTA and Transmitter are shown in Fig. 5 and Fig. 6
respectively.


5.0    Electronically Beam Steering EA Transmitter


5.1    Multi Beam Transmitter (MBT): To overcome the limitations of the
conventional transmitter, a new approach like the Multi Beam Transmitter (MBT) with
electronic steering came into existence. The philosophy behind this configuration is
adding the RF power in space. The architecture of the MBT consists of Beam Switching
Matrix, Rotman Lens, gain and phased matched mini TWTs and a linear antenna array as
shown in Fig 7.

 5.1.1 Beam Switching Matrix: It consists of SPMT switch with associated RF
assembly and control circuitry. It is digitally controlled to connect the input RF signal to
the appropriate beam port of the Rotman lens as per the desired angle.
.



           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     XI - 7
5.1.2 Rotman Lens: The key element of the multiple beam arrays is the beam forming
Rotman Lens. The Rotman lens consists of a parallel plate cavity with plate separation
less than λ/2 so that only TEM mode can propagate. It also consists of cables of specified
length connecting the array elements to the parallel plate region. The geometry of the lens
and the cable lengths are designed so that all ray paths traced from a beam port to its
associated wave front on the array port side are equal. The significant difference between
the Rotman Lens and the optical lens is that, the former has three foci’s as compared to
the optical lens, which has only one focus.

        The Rotman lens provides both true time delay phase shift and amplitude taper in
one lens component. The true time delay is one of the distinct advantages of the lens over
the phase-shifted array since that makes it independent of frequency. It is in principal a
wide band system since its design is based on the electrical length compensation. This
means that the direction of emerging beam from the linear array at the lens outer face is
independent of frequency, but the beam width and cross over levels of the multiple beams
change with frequency.


        Excitation of each input port produces a suitable phase variation across the array
elements to produce a beam in a particular direction. It can be constructed in a variety of
transmission media such as parallel plate strip line or micro strip and does not require
multi layer circuits.

5.1.3 Gain & Phase Matched Mini TWTs: High ERP is achieved by means of spatial
Power combining of the mini TWTs outputs connected to the linear array. To achieve
optimum power combining, the TWTs are to be gain and phase matched so that the
TWTs do not alter the radiated wave fronts. The mini TWTs operate in saturation mode
to minimize any amplitude variations.


5.1.4 Principle of Operation: The actual steering of the beam is controlled in response
to the RF power appearing in one of the beam ports of the Rotman Lens. Based on the
direction of the threat, the ECM processor gives the command to the Switching Matrix to
connect the low power-jamming signal to one of the beam ports to produce the beam in

            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                      XI - 8
the required angle. The number of beams that can be formed is equal to the number of the
beam ports. Exciting the appropriate beam port of the Rotman Lens can energize any one
of the multiple beams. Mini TWT amplifiers are connected between the Rotman Lens and
the Array elements to minimize the RF plumbing loss and to provide high ERP. These
TWT amplifiers are energized with the Jam-On command issued by the ECM Processor.
Individual TWT amplifiers at each array element provide the required amplification to
feed the antenna array. The beam is formed in space where the individual powers of the
TWT amplifiers are spatially combined to concentrate the jamming energy in the desired
direction of the threat. By controlling the RF power at the beam ports of the Lens,
simultaneous multiple jamming beams can be formed in different direction to handle
multiple threats.


       The ERP of MBT is dependent on the TWT output power, number of array
elements and the antenna array gain. The ERP is given by,

ERP(dBm) = TWT O/P power (dBm) + 10 Log (number of elements) + Array Gain(dB)
(2) Example : For a 16 element linear antenna array with gain of 18 dB minimum and 16
TWTs with each TWT output power of 49dBm over the frequency range are used in the
MBT, the array factor is 10*log(16) =12dB, using Eq. (2), the ERP can be estimated as


       The ERP of MBT = 49dBm + 12dB + 18 dB = 79dBm min. (80 KW min)


The multiple beam coverage by MBT configuration is shown in Fig. 8.


6.0    Current Trends in EA Transmitters:


6.1    Micro wave Power Modules (MPMs) in Phased Array architecture


       With the ever-increasing demand of the Multi Beam Transmitter in relatively
small platforms like fighter aircrafts, UAVs and active decoys, the architecture with the
TWTs as power amplifiers is not a viable solution. Modular systems using MPMs can
meet the system power and platform requirements. The MPM characteristics and features

            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                      XI - 9
also make it an ideal replacement for TWTs of existing EW systems in order to upgrade
the performance and the reliability. The development of MPM is a technology
breakthrough in broadband microwave power amplifiers. The highly miniaturized MPM
subsystem combines the optimum features of a MMIC driver amplifier (low noise,
compactness, functionality), a Vacuum Power Booster (high power, wide bandwidth,
high efficiency) and an advanced Electronic Power Conditioner (high efficiency, RF
modulation, compactness, control and logic functions).


       The MPMs are regarded as the most significant advancement in the microwave
amplifiers. It offers high efficiency, high power, high gain, wide bandwidth, high
reliability and better noise figure coupled with reductions in size weight and the cost.

6.2    Important Specifications of a typical MPM


       The broadband MPMs used in EA application cover frequency ranges of 1-2, 2-8,
6-18 and 18-40 GHz. The technical specification of a typical MPM in 6-18 GHz band is
given below.


       Frequency Range         :              6-18 GHz
       Power Output            :              70 Watts Min.
       Gain                    :              50 dB
       Noise Figure            :              12 dB
       DC input                :              270 Volts
       Power Consumption:                     350 Watts
       Size                    :              9” x 5.4” x 0.8”
       Weight                  :              1.6 Kg


6.3 Active Decoys: Active decoys are basically repeater transmitters that are used to
protect the platform from the modern advanced missile seekers. This is a kind of off-
board jammer deployed from the platform.




           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                    XI - 10
       Present day missiles are with on board active radar opening up at short ranges, in
short time. The on board ECM techniques with higher reaction time cannot counter these
missiles. Due to their inherent limitations the off –board passive jammers (chaffs and IR
flares), they cannot counter the missiles effectively. The solution for these is the active
decoy. The active decoy is basically a repeater, that intercepts the radar/ missile signal,
adds modulations to mimic the plat form, amplifies the jamming signal to the required
level and transmits back towards the radar / missile. When an aircraft deploys an off-
board jammer such as a towed decoy, the angle error of the missile can be distorted by
the presence of second RF source, the first being the reflections from the aircraft. Further
more, if the Aircraft and the Towed decoy are irresolvable in range, velocity, or angle the
response of the missile seeker becomes a complex function of two RF sources
introducing an error in angle which translates into increased miss distance and thus
increased probability of survival of the platform.


       Hence Towed Active Decoy is most effective ECM against modern present day
missile seekers by virtue of its geometry with respect to the missile and platform. In the
case of a ship platform, the decoy can be towed to the ship or can be launched from the
launchers installed on the deck


       The Integrated Defense Electronic countermeasure (IDECM), systems is a
combination of on board / of board ECM system. A typical IDECM system is shown in
the Fig. 9. The on board portion is based on the Self Protection Jammers design. The
Technique Generator is designed to apply a variety of RF techniques against pulsed/ CW
and Pulse Doppler threats. This modulated signal is converted to light and transmitted to
the towed decoy through a fiber optic link. The fiber optic towed decoy converts the laser
signal back to RF, amplifies and radiates this jammer signal towards the missile / radar.


       As there is a constrain in the size and weight of the decoy, the conventional TWTs
with high Voltage Power Supplies can not form a part of the decoy system. Hence the
MPMs are the best alternatives.



           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                    XI - 11
       The present Electronic Attack Systems are aimed at high ERP, efficiency, light
weight, and modularity for various platform installations. Microwave power module is
identified as emerging technology in this area.


7.0 Conclusions:
       For an integrated EW, the EA system plays a major role in the tactical scenario.
Modern ECM systems employ multi-beam jammers for countering the multiple threats
simultaneously in all the directions. The emphasis on modern jammer transmitters is on
its reliability, cost and modularity that can be configured to suit any platform installation.
Future transmitter systems for EW will play operational capability to perform multi-
functional tasks and support future integrated RF sensors. This implies electronically
steered wideband arrays and modular architectures of the EA transmitters using MPMs.
Future EA transmitters can meet system power requirements ranging from few hundreds
watts to hundreds of kilowatts, using relatively low power (i.e., 1 watt to 200 watts) RF
amplifiers.


8.0 References

       1. “Introduction to Electronic Warfare”, D. Curtis Schleher, Artech house
              Publishing, Boston, USA, February 1986.
       2. “Introduction to Electronic Warfare in Information Age”, D. Curtis Schleher,
              Artech house Publishing, Boston, USA, 1999.
       3. “Microwave Power Module – Features and Applications”
              Communication & Power Industries - Travelling Wave Tube Products
              Division, New York, USA.
       4. Special Technology Area Review on Vacuum Electronics Technology for RF
              Applications Report of Department of Defense Advisory Group on Electron
              Devices Compiled from the STAR, USA, held on December 11-12, 2000.




              CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                       XI - 12
                                                                              RF Power
                                                COOLING
                                                                               Monitor
                                                SYSTEM


          LOW POWER
          RF SIGNAL SOLID STATE       RF IN           TWT           RF OUT           DDC
                      DRIVER
                     AMPLIFIER

                                                                                                PARABOLIC
                                                                                                REFLECTOR
SYNTH.            ECM            REPEATER                                                        ANTENNA
CHANNEL        TECHNIQUE         CHANNEL
               GENERATOR
                                                                                           DETECTOR


                              MAINS AC              HVPS &
   JAMMING                    INPUT SUPPLY        MODULATOR            REFLECTED
 PARAMETERS          ECM                                              POWER SENSE
   FROM ESM       PROCESSOR
  PROCESSOR

                                                      MOD
                                                    PULSE IN


                  FIG 1 : BLOCK DIAGRAM OF EA TRANSMITTER




                      CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                               XI - 13
     Fig. 2 : ELEMENTS OF HELIX TWT




CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                         XI - 14
     EMI/EMC            LV              Rectifier              Heater          HTR             HTR Trans./                Heater
      FILTER           Trans             Filter                 Reg.           INV             Rectifier Fil.
230 VAC (Ph-N) ,
50 C/s 3ϕ, 5 Wire                       Un Reg.
                                        Auxiliary                                                          Rect.
                                        Voltages               Focus           Focus        Focus          Filter
                                                               Reg.             Inv.        Trans.         & Reg.       Focus
                                                                                                                        Electrode/
                                                                                                                Beam ON
                                                                                                                        Grid
                                                                                                                Voltage
  MOD Pulse IN
                                                                               MOD                         MOD
                                                                               Proc.
                                                                                                                Beam OFF
                        Rectifier                                                                               Voltage
                        & Filter
                                                                                                         Rectifier,       Cathode
                                                                                                         Filter &
                                                                                                           Reg.
                                                                                                                          Collector
                      PWM OSC &                                                 HV           HV             HV
                      Isolated drive                                            Inv.        Trans.         Rect &
                                                                                                                          Helix
                                                                                                           Filter
                                                                                   HV Feedback


 HV ON               Control &           Over Helix
 Beam ON             Processor            Excess Reflected power
 Battle Shot                              Over Temp
 Reset                                    Status Signals

                    Fig. 3: Functional Block Diagram – High Voltage Power Supply and Modulator

                                    CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                                             XI - 15
                      RF                                  RF
                      IN                                  OUT




                                                                      TWT




                                                                           Collector
Heater Heater Focus Electrode/ Helix
       Cathode Grid/Control Anode



           HIGH VOLTAGE POWER SUPPLY / MODULATOR



    Fig. 4 : INTERFACE DIAGRAM OF TWT WITH HVPSM

            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     XI - 16
                COOLING SYSTEM
                HP TWT
                                                SSA
        DDC



    HVPSM




Fig. 5: HIGH POWER TWT AMPLIFIER
  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                           XI - 17
Fig. 6 : HIGH POWER TRANSMITTER


CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                         XI - 18
                                                                         ANTENNA # 1
                                                           TWT 1

                                                                                      Linear Array
                            1       B1                                                       B15


              Beam                        ROTMAN LENS
Low Power                       8
   RF
             Switching              B8                                                        B8
              Matrix
Generation
                             15                                                              B1
                                    B15

                                                            TWT 16
                                                                              ANTENNA # 16



                         Fig.7 : Multi Beam Transmitter


               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                        XI - 19
FIG 8 : MULTIPLE BEAMS IN MBT CONFIGURATION

      CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                               XI - 20
RX ANT
         RWR                   ECM                    TECH. GEN               RF TO
                            PROCESSOR                                         LIGHT



                                 FIBRE OPTIC LINE



                 LIGHT TO RF                       TWT AMP
                    CONV
                                  TOWED                           TX. ANT
                                  DECOY




                   FIG 9 : BLOCK DIAGRAM OF IDECM




               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                        XI - 21
             CHAPTER – XII


MM WAVE RADAR SYSTEMS – SOME
        EW ASPECTS

  Shri. P. RAGHAVENDRA RAO, Sc’F’




    CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                               XII
                         CHAPTER - XII



             MM WAVE RADAR SYSTEMS



                            CONTENTS



1.0 Introduction:

2.0 Current MMW Military Radars – EW Aspects

3.0 Incorporation of ECCM

4.0 MMW Radars – ECM aspects

5.0 Electronic Support Measure (ESM)

6.0 Conclusion

7.0 References




      CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                  XII-1
           MM WAVE RADAR SYSTEMS – SOME EW ASPECTS

                            P.RAGHAVENDRA RAO, Sc’F’
                                 DLRL, Hyderabad


1.0    Introduction:

       The Prime characteristics of MMW Radars, when compared to conventional
Microwave Radars, can be listed as follows:


       •       Narrow Main Beams and Low Side Lobe levels which make jamming
               difficult.


       •       Larger Band Widths of operation for better Dopplers and High Range
               Resolutions.


       •       High Tracking/Guiding Accuracy and operation at Low Elevations and
               optimization of Multi path effects, because of Narrow Beam widths.


       •       High angular and Range Resolutions


       •       Effectiveness for an Exoatmospheric Satellite Target Acquisition due to
               minimal space attenuation.


       •       Due to small physical size, suitability for application as an active seeker
               for terminal homing missiles and for Anti Ballistic Missile Terminal
               Homing seekers.


       •       Feasibility of employing 2 Bands of Frequencies (Ex: Ka/W Bands) on a
               single platform to multiply the effects of Frequency Agility.




           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                       XII-2
               Generally, Terrestrial / On Board MMW Radars employ Frequencies
       around the ‘Atmospheric Windows’ (where the Atmospheric attenuation is
       minimum) like 35 GHz., 94 GHz., 140 GHz., etc.

2.0    Current MMW Military Radars – EW Aspects :


       In Electronic Warfare, Radar functions as an important sensor in a Multisensor
environment and it should not be vulnerable to ECM. However, in practice, the MMW
Radar sensors are mostly vulnerable to Barrage Noise Jamming. In this context, MMW
Radars with ECCM features incorporated will help the Radar Operator to continue the
mission of Tracking by appropriately choosing an optimum method.


       By virtue of the Narrower possible Beam Widths, the detection and tracking of
targets at low angles will not be degraded due to the reduction of ground clutter and
Multipath effects - in the case of MMW Radars.


       The signal processing philosophy in MMW Radars, maximizes the target
information by high resolution and it is reported that the target identification by imaging
Techniques has become a realistic goal without employing synthetic aperture techniques.


As discussed above, since the tracking of low flying targets (100 feet) encounters
problems1 like multipath propagation phenomenon which is complex, specular diffuse
reflections are involved. The occurrence of image is difficult to predict and its position
varies continuously since it is a function of target height, antenna height and surface
roughness which are never constant. More over, in the control systems, the reaction time
is very essential parameter.


       To over come these problems, consideration of the lower frequencies for quick
acquisition and higher frequencies for realising a narrow beam width, is an optimum
solution2 .   In this case, it is advantageous to have the same antenna for both the
frequencies so that parallax and alignment problems are minimized.


         CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     XII-3
       Scattering and reflective properties of land and sea clutter at MM Waves depends
largely on the surface roughness. It is reported by L. Klaver, that in X/Ka Band tracking
Radar, Ka system experiences relatively a rougher surface which indicates that the
Specular reflection coefficient for Ka band is relatively small.


       Fly Catcher Radar5 of M/s HSA, Netherland currently available, employs dual
frequency bands (X and Ka). The Chinese Hunting Dog Radar in Ka Band is reported of
being used for moving target search and gun pointing adjustment. The other Radar
system, is STARTLE (Surveillance, Tracking Radar for Tank Location and Engagement)
at 94 GHz., using Solid State Impatt Source. Use of the type of Gunn Oscillator as a
Local Oscillator and Low Noise Mixer Diodes’ use resulted in increase of Radar
detection range in a benign environment.


       Ka Band Fly Catcher (of M/s. Signal Apparaten, Nether Lands) at around 35
GHz., with a pulse width of 0.140 micro seconds, peak power of about 15 Kilowatts,
Antenna gain of about 42 dB has the tracking range capability of about 13 Km for a 1
Square metre Target where the Conical Scan Tracker is employed for tracking the target
with a tracking accuracy of less than one Milli Radian.


       Another important Military Radar developed by M/s. CONTRAVES, also utilizes
the dual frequency concept.       The significant ECCM features in this Radar are –
Frequency agility, Pulse compression CFAR( Constant False Alarm Rate) techniques and
better signal processing techniques.


       Fire Control Radar employing Ku and W bands (94 GHz.) was developed by US
Armaments R & D for use as a Tank mounted Gun Fire Control System. This dual
frequency Tracker employs a common monopulse Antenna at both the frequencies. Due
to inherent advantage of narrow beam width and high resolution properties, this fire
control tracker can provide significant advantages in respect of multipath propagation and
ground main beam clutter.

         CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     XII-4
       MMW Radars with wide band widths, provide high information rate capability for
determining target signature structures with narrow pulses or wide band FM. Various
wave form designs with coding can be implemented due to wider band widths. Moving
objects produce larger Doppler Frequency shifts resulting in increased classification
capability.

3.0    Incorporation of ECCM:


       The ECCM features that can be incorporated in MMW Radar systems either in
Transmitter or Receiver/ Signal Processor can be realised as follows:


       (a)     For a Target Tracking Radar broadly by:


       -       Employing Monopulse Pulse Doppler System
       -       Providing facility for Passive Tracking of Target
       -       Providing fast switching among various spot frequencies in a band
       -       Simultaneous Tracking in 2 different Frequency Bands (eg:Ka/W Bands)
       -       Frequency Agility
       -       Using Effective Band Pass Filters to Counter Broad Band Jamming
       -       Using Specific Waveform coding or FM chip or phase coding. If the
               ECM signals do not have the same coding as that of the defending Radar
               signals, the process gain in the Radar is less to the ECM signal compared
               to the Radar echo signal which results in improvement of S/J Ratio. If the
               coding is pseudo Random, then it will be much more difficult for the
               jammer system to interfere / jam the Radar system
       -       Employing Frequency stepped Pulse compression Technique which results
               in better Range Resolution and minimising Data sampling rate problems.
       -       Transmitting powers of Higher values, employing wide pulses and Dwell
               numbers of larger values.




           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                       XII-5
       -       Employing Antennas with low side lobe, side lobe blanking and coherent
               side lobe cancellation characteristics.


       (b)     For a Search Radar exclusively:


               By,
       -       Incorporating MTI using Digital cancellor and Digital Video Correlator
       -       Fast Switching between various frequencies in same band
       -       Floating False Alarm Level
       -       Staggered/Pseudo Random PRF


4.0    MMW RADARS – ECM ASPECSTS:

       Though the Narrow Beam Width of MMW Radars make them less vulnerable to
Hostile jamming, still there is a chance that jamming can be accomplished in either
Active or passive configuration. Passive Jamming is realized by locating chaff, Decoys
and employing techniques accomplishing tactics only- towards the Radars. Active ECM
is jamming with transmitted Electro-Magnetic Spectrum.


       Active ECM can be divided into two types VIZ Denial ECM and Deceptive ECM.


       In Denial jamming, usually, ‘brute force’ type high powers are radiated towards
the Radars so that the jamming signal covers the Radar echo-signal to the exclusion of
targets. The better way for the Radar to withstand this on-slaught is to have a fool proof
CFAR-threshold setting circuit.


       Spot Noise Jamming, where the jammers’ energy is entirely within the Radars
band pass is effective against a non agile Radar.


       Wide band Noise jamming (known also as Barrage Jamming) employed to
counter the agility of the Radar, is less effective than Spot Noise Jamming, because of the
rejection of the jamming energy by the Radar’s filters.

           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                       XII-6
        For employing the Denial Jamming, TWT Amplifiers are available with CW
Powers and Pulsed Powers of 63 dBm & 75 dBm respectively at 35 GHz and 50 dBm
and 65 dBm respectively at 94 GHz.

In Denial Jamming:

(A)     Self Protection Jamming (Self Screen Jamming (SSJ) is an interfering EM energy
deliberately emitted by vehicle, the Radar is trying to detect. The jamming is done
through the main lobe of the Radar Antenna.


        For different Target cross section values (1m2, 10m2), Cross Over Ranges are
estimated using the formula given below. Then for two different values of signal to
jamming Noise Ratio of 10 dB and 20 dB, the Detection Ranges are estimated as per the
jamming equation. The cross over and Detection Ranges of the Radar are independent
of the Frequency of Operation of the Radar.


Pj =    (PT GTσ )BJ x Processing Gain x System Loss x J/S
        (4π)RT2GjBR

Where                                                       for a typical case

S/J     =      Signal to Self Protection Jamming Power Ratio
PJ      =      Jammer Power                                         = 53 dBm
PT      =      Radar Transmitter Power                              = 70 dBm
GT      =      Radar Antenna gain                                   = 42 dB
GJ      =      Gain of Jammer antenna towards Victim Radar          = 10 dB
σ       =      Cross section (meter2)                               = 1, 10
BR      =      Radar’s Receive filters Bandwidth (MHz)              = 0.5
BJ      =      Effective Bandwidth of Jammer’s emission (MHz)       = 50
All system losses                                                   = 0dB for normal
                                                                      Conditions
Processing Gain                                                     = 16 dB
       For the Radars at both 35 GHz and 94 GHz system.

        Cross-over Range (Rc)3 can be defined as the range at which, for the Radar
System, the Jamming Noise to Signal Ratio (J/S) is unity. The Detection Range (Rd) can
be termed as the range at which S/J is sufficient for target detection in the Interference.

         CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     XII-7
For typical case, mentioned above, the Jammer power is estimated for required Crossover
ranges for various target cross sections as shown in Table 1. Detection range for typical
S/J ratio values of 10 dB and 20 dB is also estimated.

               Rd     =       Rc / √(S/J)

(B)     Stand off jamming is an interfering signal deliberately emitted by a vehicle other
        than the one, the victim (threat) Radar is trying to detect. Hence the jamming is
        done through, possibly, side lobe of victim Radar Receive Antenna.


        In standard notation, here also cross over ranges and Detection Ranges are
        estimated using the formula

               Pj =   (PT GT2 RJ2 σ )BJ x Processing Gain x System Losses x J/S
                      (4π)(GJGRJ)RJ4BR

Where                                                       for a typical case

G RJ    = Radar Receive Antenna’s Gain towards Jammer = 20 dB

RJ      = Range from Jammer to Radar (Meters)               = 10,000
PJ      = Jammer Power = 1000 W                             = 1 KW
GJ      =30 dB

And the values of other parameters remaining the same as in SSJ case. Here also the
cross-over Ranges can be estimated for the required Cross sections and Detection ranges
have been estimated for a typical S/J ratio of 10 dB and 20 dB as shown in Table -II.


        The Jammer Power is normal and adverse weather conditions is estimated in Ka
and W Bands. The Jammer signal suffers a one way path loss where as the Radar signal
suffers two way path loss. The data is summarised as shown in Table I & II.


        Approximate values of one way attenuation1 for 35 GHz signal for normal
condition and moderate rain condition are 0.15 dB/Km and 1 dB /Km respectively.
Similar values for 94 GHz signal for normal condition and moderate rain condition are
0.4 dB/Km and 4.5 dB/Km.

          CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                      XII-8
           Thus in MM Waves the atmospheric attenuation can actually reduce the amount
of Jammer ERP (Effective Radiated Power) required to jam an MMW radar in Self
Screening Jamming. In Stand off Jamming, the atmospheric attenuation magnifies the
Jammer ERP required to jam an MMW radar.

                                         TABLE – I

         NORMAL CONDITION
       σ     Cross over Range           Detection Range(Km)
          2
       (m )  (Km)                       S/J = 10 dB     =20 dB
             (Ka)    (W)                Ka      W       Ka     W
       1      5.0    5.0                1.5     1.5     0.5    0.5
       10    15.85   15.85              4.6     4.6     1.585 1.585

                                         TABLE – II


        NORMAL CONDITION
      σ    Cross over Range             Detection Range(km)for
      (m2) (Km)
                                        S/J = 10 dB      20 dB

                      Ka          W           Ka / W         W / Ka
      1         5.2         5.2         2.8              1.70
      10        9.4         9.4         5.1              3.10


5.0        Electronic Support Measure (ESM):           The receiver of MMW Track Radar
operating at 35/94 GHz., can be thought of to get modified to accommodate a built in
ESM Receiver. The Track Radar receiver chain can be broken in between the output of
Track Receive antenna after duplexer and the RF input of mixer of Radar receiver. A
frequency diplexer4 can be introduced in the gap.

           The radar echoes from the targets are received and further processed in the Radar
Receiver. Any other emissions from the threat side whose frequency is other than 35±
0.5 GHz / 94 ±0.5 GHz can be processed in the built in ESM Receiver. The loss of
approximately 2 dB suffered by the Radar Target echo in this modified Radar Receiver
would result in 0.9 times the original range performance. But the advantage here is that

             CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         XII-9
the same Track Radar Receive Antenna is used for receiving the ESM Receiver signals
also.

6.0     Conclusion :

        EW at MM Wave lengths is very difficult if not impossible with the state of the
art technology and components’ availability; highly moderate EW system in different
bands MM Wave frequency range can be evolved.

7.0     References:

        1.       D.C. Schleher, “Introduction to Electronic Warfare”, 1986, Artech House
                 Inc.
        2.       T. Kirubarajan et al, “IMMPDAF for radar management and tracking
                 bench mark with ECM”, IEEE Trans. On Aero Space and Electronics
                 System Vol. 34 No.4, Oct 1998.
        3.       Byron Edde, “Radar Principles, Techniques, Applications”, PRENTICE
                 HALL INC. New Jersey 1993.
        4.       P. Raghavendra Rao, “On Development of 4 Port W/G Circulator In MM
                 Waves”, Proceedings of the first Asia – Pacific Microwave Conference,
                 Feb 24 – 28, 1986 New Delhi, India.




             CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         XII-10
        CHAPTER – XIII


ECM EFFECTIVENESS
   Shri. D.D.SARMA, SC’E’




CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                             .
                           XIII
                       CHAPTER – XIII


                ECM EFFECTIVENESS

                             CONTENTS




1.    INTRODUCTION

2.    MEASURES OF ECM EFFECTIVENESS

3.    DESIGN CONSIDERATIONS OF AN ECM SYSTEM

4.    SUITABLE JAMMING TECHNIQUES AGAINST DIFFERENT
      RADARS

5.    COMBINATION OF ECM TECHNIQUES

6.    JAMMING TECHNIQUES VS ANTI JAMMING TECHNIQUES

7.    CASE STUDY – DYNAMIC AIRCRAFT ENGAGEMENT

8.    ECM EFFECTIVENESS – PRESENT TRENDS

9.    CONCLUSIONS

10.   REFERENCES




         CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                      .
                                    XIII
                                  ECM EFFECTIVENESS

                            D.D.Sarma, Sc’E’ , DLRL, Hyderabad.

1.0   Introduction :

              Countermeasures include all those means of exploiting an adversary’s
      activity to determine his intensions or reduce his effectiveness. Countermeasures
      may be applied against weapons operating over the entire spectrum from acoustic
      to gamma wave frequencies. Since the Radio Frequency Spectrum is most
      predominantly used for weapons / Sensors, communications etc., most of the
      counter measures and counter countermeasures also operate in this region.


              The basic purpose of ECM is to interfere with the operation of sensors of
      air / surface defence systems and through them to interfere with the operation of
      the system itself. By this process, ECM attempts to make the defence more
      uncertain. The greater the defence uncertainty, the more effective is the ECM.
      This can be achieved by reducing the information content of the signals, the
      defence receives with its sensors and force the defence system to make mistakes
      or errors.


              One should always keep in mind that ECM does not have to prevent
      tracking completely to be effective. It can delay the establishment of a solid track
      on a target, causing a moment’s confusion or forces the decision maker to wait
      few more seconds to be sure of the proper response.


              Given that we want to interfere with an enemy air / surface defence radar,
      how do we go about it ? This involves development of an ECM system,
      considering the following aspects.


                        -        Threat and scenario definition
                        -        Design & Development of system architecture
                        -        Selection of ECM technique
                   CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                                .
                                             XIII-1
                     -       Parameters of ECM technique
                     -       Selection of deployments and tactics.
                     -       Analysis to evaluate system effectiveness
                     -       Establishment of one or more measures of effectiveness
                             keyed to mission objectives.


             Effectiveness evaluation can be a theoretical laboratory evaluation or
      field evaluation. Depending on the complexity of the system and the technology,
      the effectiveness evaluation is done at various stages of the system realisation
      from design to production. Many software packages are available and simulators
      are also used in the study.



2.0   Measures of ECM Effectiveness :


             The interaction between the countermeasure and the target is complex,
      involving a large number of parameters of the target platform (ship, aircraft or
      ground vehicle), countermeasure technique, target and the engagement
      parameters. Furthermore the interaction is dynamic, with the target and the
      platform positions changing with time and the countermeasure operating modes
      also potentially changing as the engagement unfolds. The dynamic complexity of
      the interaction represents a substantial and evaluation challenge.


             The primary means normally used to measure the effect of a specific
      countermeasure on the specific target has been through the execution of
      hardware-in-the loop laboratory or field trials using countermeasure equipment
      and simulator systems . The conduct of such trials, involving substantial
      equipment and human resources is expensive. Cost inevitably places a limitation
      on the number of test runs executable. In the face of these factors , another
      approach is to systematic engineering methodologies supported by appropriate
      tools applied to realize the reduction and analysis of the data. This gives the

               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                            .
                                         XIII-2
      necessity of identifying the measures of effectiveness to proceed in our
      evaluation.


             The measures of Effectiveness (MOE) to describe the ECM performance
      are Miss distance , Probability of kill and Length of time for the generation of a
      large tracking error. These measures of effectiveness can be studied for each of
      the countermeasure techniques and their combinations against different targets
      and platform.


             The ECM effectiveness for various types of Countermeasure like noise,
      range, doppler and angle techniques are discussed in detail. This chapter also
      discusses the passive and active ECM techniques and combinational techniques.


3.0   Design considerations of an ECM system :


             For an effective ECM, the architecture of the system to be arrived at
      depends on defence tactics, spectrum of region, angular coverage, Radars to be
      encountered and no. of simultaneous threats to be handled.


             Based on Defence tactics, an ECM system can be intended to play its role
      as
                      -     Self Screening Jammer
                      -     Stand-off Jammer
                      -     Escort Jammer
                      -     Stand forward Jammer




               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                            .
                                         XIII-3
       Self Screening Jammer(SSJ) carries both weapon system and jamming
equipment for its own protection. This results in trade-off between weight and
space reserved for ECM equipment and that reserved for sensors, weapons and
fuel. The trade-off is the most critical in aircraft and the least critical in ships. In
general, Self screening jammers are equipped with medium power transmitter
systems.


       Stand-off Jammer(SOJ) remains outside the range of enemy weapons,
providing screening for attacking units that actually penetrates through enemy
defence. The advantage of SOJ is that the jammer is safe from enemy Home-on
jamming             ( HOJ ) weapons. The disadvantage in that, the burn-through
occurs earlier on attack units because the jammer must remain at very long range,
while attack units being close to the enemy to a very short range. SOJ systems
are, in general, equipped with high power transmitter systems.


       Stand forward Jammer(SFJ) is placed between enemy sensors and attack
units. While maintaining proper geometry between victim sensors, attack units
and the jammer is difficult, this method allows the most effective use of jammer
power by reducing spreading and attenuation losses. This situation is the most
dangerous for the jamming unit because he is the prime target for all weapon
systems and well within the capabilities of Home-on-Jam (HOJ) and Anti
radiation Missile(ARM) weapons.


       Depending on the mode of operation, ECM transmitter is to be designed
for the required ERP, considering the output power of a transmitter, antenna gain
in the direction of victim receiver and the transmitter bandwidth. In addition, the
amount of ECM transmitter emission delivered into the victim receiver is a
function of receiver bandwidth, its antenna gain and RCS of the target. The
effectiveness of ECM transmitter depends upon its power density within the
bandwidth of victim receiver after considering antenna factors and propagation

           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                        .
                                     XIII-4
      path losses i.e the effectiveness depends on the ratio of jamming signal to the
      target echo signal received at the radar.

             The Jammer to Signal ratio for a Stand - Off Jammer is given by standard
      notation as

                       J / S = Pj Gj . 4Πr2 . Br . GSL
                               PtGt     σ     Bj GML

             It is evident that the value of J/S decreases as the range to the target
      decreases. This phenomena occurs for all radars at some range at which the
      strength of radar echo becomes equal to or greater than ECM signal.


             Usually, the burn-through range is calculated for standard set of conditions
      and is used to compare the effectiveness of ECM against various radars. But in
      practice, the burn-through range is never a constant.            First, the reflective
      properties (RCS) of the aircraft or ship can vary over a range of 1000: 1,
      depending on the particular aspect presented to the radars. Second, the ability of
      radar to distinguish its echo from the ECM depends on its design, its condition
      maintenance and on the signal processing circuits in use at that time. i.e, the burn-
      through to the operator or automatic detection may occur when the echo is either
      stronger or weaker than the ECM signal depending upon the radar configuration
      and condition.


             The effectiveness of an ECM technique, like Noise Jamming is greater as
      the J/S increases. This means that by decreasing S (which results from the RCS
      reflection of an aircraft ), the J/S increases within same J. This is achievable by
      using Stealth fighters and bombers which have large advantage in penetrating
      hostile air space that is protected by surface to air missile systems.


3.1   General Specifications of EW system :

             After arriving at the required transmitter power, the ECM system
      architecture to be designed keeping in view of the following parameters.
                CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                             .
                                          XIII-5
                     -       Sensitivity
                     -       Dynamic Range
                     -       Frequency Range of operation
                     -       Azimuth and elevation angle coverage
                     -       Choice of Jamming Techniques
                     -       Multi-threat handling ( Time share / Power share )
                     -       Reaction time

             To have a better effect of Counter measure action on the threat receiver, it
      should be ensured that ECM transmitter radiates


                     -       at the correct frequency
                     -       in the correct direction
                     -       at the correct time
                     -       with the correct ECM Technique



3.2   Electronic Support (ES) System :


               The EW receiver is the primary Electronic Support Measures (ESM)
      equipment and functions as a sensor for and as a means of identifying Friendly,
      neutral and enemy electronic emissions. It provides warning of potential attack,
      knowledge of enemy capabilities and an indication of enemy’s use of active
      counter measures to manipulate the electromagnetic spectrum. The design of EW
      receiver provides a special challenge to the engineer, in that no single antenna
      system or specific receiver circuit can cover the entire range of the electro
      magnetic spectrum. A set of components can be designed to provide maximum
      efficiency over a range of up to a few thousand Mega Hertz.. However, current
      requirements demand performance from a few KHz to 50 GHz with a wide range
      of signal strengths and other parameters such as Frequency, PW ,PRF, Scan rate,
      DoA , sideband characteristics and modulations ( Inter pulse, Intra pulse etc.).
               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                            .
                                         XIII-6
                 As the frequency spectrum range is too large for one receiver, either
        several different ECM receivers with different tuning ranges must be used or one
        receiver must use different tuning units to cover different parts of frequency
        coverage.

3.3     Choice of Radar Jamming Techniques :


               At present an Electronic Jamming system ( such as air-borne Self
        Protection System ) often has diversed jamming methods. However, various
        system radars and diverse anti-jamming methods are also employed in enemy air-
        defence systems. Thus, in battle procedure, the effects of tactics, including
        survival probability of own planes, will be different according to measures
        employed by both sides. It is necessary to make a strategic decision about use of
        different types of jammers and jammer techniques.

               Jamming techniques are basically classified into

                       1. Noise Jamming
                       2. Deception Jamming


3.3.1   Noise Jamming :


               One way of preventing a radar receiver from functioning correctly is to
        saturate it with noise. Noise is a continuous random signal and is dissimilar to
        radar signal. The radar signal or echo is a periodic sequence of pulses and the
        main objective of Noise Jamming is to cancel the echo. That is, the average
        amplitude of the radar echo is to be cancelled. This can be alternatively expressed
        by saying that the average power of the jammer must have the same effect as the
        peak power of the radar echo or the noise-to-signal ratio at the input is raised to a
        level beyond which the receiver can extract the intelligence.



                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                              .
                                           XIII-7
             Finally, when the radar antenna is pointed toward the jammer, the radar
      sees signals at all ranges. The effect on a PPI scope is to create a solid line at the
      azimuth of the jammer. This line called a strobe, dictates to the operator that a
      jammer is present and gives its azimuth, but he does not know the range of
      jammer, once the jamming is effective.

3.3.1.1 Major Noise Jamming Techniques :


             The following are the most commonly used Noise Jamming techniques.

                     1. Spot Noise
                     2. Barrage Noise
                     3. Sweep through Noise
                     4. Doppler Band Noise

             In Spot Noise technique, jammer concentrates its power into narrow band
      centred on the frequency of victim radar. Typically, jammer noise bandwidth
      equals or exceeds the radar bandwidth. This is the most effective technique
      against fixed frequency radars.


             In Barrage noise jamming, jammer spreads its power over a wide
      bandwidth. As a result, it can cover all the radars with frequencies in that band or
      frequency agile radars. This technique does not require accurate tuning            of
      jammer signal. Barrage noise is generated by frequency modulating the carrier
      signal by ramp signal added with noise. Each time the jammer signal sweeps
      across the radar receivers Instantaneous Band Width (IBW) an impulse will be
      produced. Repetition rate of ramp signal determines number of impulses. To high
      repetition rate ( when compared to instantaneous bandwidth of radar receiver )
      will produce more impulses, whose energy content will be very low. Too low rate
      will produce less number of impulses, but their energy content will high.
      Normally, frequency of ramp signal shall be of the order of the instantaneous
      bandwidth of radar receiver.
               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                            .
                                         XIII-8
                In Sweep through Noise technique, Spot noise is swept back from across
        the wide bandwidth. This technique will be effective, if bandwidth and sweep
        rates are chosen correctly.


                In Doppler Band Noise Jammer, the noise bandwidth is of the order of
        few KHz just covering all Doppler cells. Generally, the BW is selected to match
        with the doppler radar bandwidth. Doppler Band Noise disrupts the Doppler
        processing circuits of radar.

3.3.2   Deception Jamming :

3.3.2.1 Range Deception :

                In contrast to Noise Jamming, deception tries to mimic the radar echo, so
        that, the radar will respond as if it is receiving an echo from another aircraft or
        ship. For a radar to direct a fire control system correctly, it must accurately
        measure the target range, bearing and velocity. If either range or bearing is
        misrepresented without operators knowledge, the target’s location will be
        incorrectly established. Deception is generally accomplished by repeaters /
        transponders and is sometimes also called repeater jamming.


                The transponder repeater plays back, a stored replica of the radar signal
        after it is triggered by radar. It can be programmed to remain silent when
        illuminated by main lobe beam and transmits only when illuminated by side
        lobes, creating spurious targets on the radar display at a direction other than that
        of the true targets.


                A range gate stealer is a repeater jammer whose function is to break the
        lock of the tracking radar on the target. The most efficient jamming against Split
        gate range tracker is the Range-Gate Pull-Off ( RGPO ) and Range Gate Pull–In
        (RGPI) . If the apparent acceleration of the false target is within the response

                  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                               .
                                            XIII-9
capability of the radar, the average jammer power nearly has to exceed the
average skin return power. In order to be successful, high degree of
synchronisation between the jammer and radar is required for this type of
jamming. The repeater provides this synchronisation very effectively, but since
the transmission and reception occurs simultaneously and since a target with a
large effective cross section required a high gain repeater, isolation between the
Rx and Tx antennas can be a serious problem. To overcome this isolation of
problem, a small slice (50-100 nsec.) of receiver signal is sampled and stored in a
delay line memory and is re-circulated to generate the required PW of jammer
signal. This technique can be achieved using Frequency Memory Loop ( FML)
and has very wide band of operation covering tens of GHz.. But it has problems
such as


          -        Noise built up by means of re-circulation which limits the storage
                   time to 5 µsec.
          -        Generation of Non-coherent signals
          -        Phase discontinuities at the slice boundaries resulting in spectrum
                   spread and dilution of effective jammer power.
          -        Can not handle Intra-pulse modulations.
          -        Can not generate RGPI & false targets.


          In order to overcome the above mentioned problems, DRFM has emerged.
It is not as wide in bandwidth as FML, but it has specific advantages such as


          -        Storing and re-producing pulse for coherent jamming.
          -        Capability to handle Intra-pulse modulation.
          -        Unlimited storage to generate RGPO, RGPI and FTG
          -        Very high quality spectral quality re-production to provide highly
                   effective jamming.



              CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                           .
                                       XIII-10
              Hence, DRFM is more attractive to defeat the modern radars in
       comparison to FML. However, DRFM can not handle the radar signals with
       frequency agility on pulse by pulse basis. Hence, DFML mode of operation can
       be used to overcome this problem.

3.3.2.2 Velocity Deception :


              In CW or Pulse Doppler radars, the target discrimination is accomplished
       on the basis of doppler shift of the echo. Because of the extremely narrow BW of
       speed gate filter, barrage jamming of either CW or Pulse Doppler radar would be
       very inefficient. Spot Noise Jamming also would be very difficult because of the
       required accuracy in frequency set-on. Repeater Jammers naturally satisfy the
       accurate frequency set-on requirements and lend themselves very well to the
       deception techniques.


              The requirement of VGPO/VGPI jammer is that the repeated signal
       exceeds the normal skin return from the target and that the pull-off/pull-in rate be
       kept within the capability of the victim radar tracking loop.


              In the process, the deception repeater jammer is obviously more complex
       than the noise jammer. The deception jamming is the most effective one against
       modern radars which employ coherent integration techniques such as the Pulse
       Doppler and Pulse compression types. This is because those radar have large
       processing gain( an order of 20-60 dB ) and hence attenuates the noise jammer
       signal by that amount, necessarily accepting any target like returns ( Repeater
       jammer’s false target) unattenuated.


              A compromise between deception and noise jamming is so called “smart-
       noise” jammer where by noise bursts about radar’s centre frequency which are
       timed to coincide with and cover the true target return, are transmitted by the
       jammer.
                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                              .
                                          XIII-11
3.3.2.3 Angle Deception :



              Angle Jamming techniques induce false angle information into tracking
       circuits of radar. If a false target jammer signal enters the radar receiver through
       its side lobe then the target video is indicated on the radar display at an angular
       error equal to the angular displacement between the main lobe and the side lobe
       involved. This technique can be applied to any radar within effective side lobe
       suppression or cancellation. The radar in turn may use antenna with extremely
       low side lobe level. In this case, constant gain ECM system would be more
       effective than constant power ECM system.



              To attack the radar angle tracking loop, the jammer program management
       must ensure that the ECM signal with an appropriate modulation is passed
       through the radar range or velocity gate, which the angle tracking loop will
       accept, and process. The amplitude modulation should be near to the nutation
       frequency used by the radars Tracking system because the acceptable bandwidth
       of angle tracking loop is only about 1 to 2 Hz centred at the reference frequency.
       A high percentage of AM modulation relative to that imposed on the signal by
       the nutating radar antenna should be applied. An effective way for an ECM
       system is to amplify the detected amplitude modulation at the ECM receiver,
       invert it by 180° and use it to amplitude modulate (Square wave) the ECM
       transmitted signal.

              ECM against Lobe on Receive Only (LORO) and Conical Scan On
       Receive Only (COSRO) uses antennas which nutates on during the reception of
       the radar signals. And hence Inverse gain jamming will not be effective in this
       case. However to counter Radars using LORO or COSRO type of scan, Swept
       Square wave modulations are employed. In this ECM technique, the frequency of
       AM square wave is swept from the expected low scan frequency to the high scan
       frequency. As a result, when the ECM signal crosses the actual scan frequency,
                CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                             .
                                         XIII-12
perturbation in the antenna tracking servo loop occurs and the angle track breaks
down.

        The most effective ECM techniques to counter the COSRO type radar are
the use of Jog Detection with Swept Square Wave Modulation. This technique is
based on the fact that, even though the sweep speed is such that the ECM
modulation does not dwell long enough to effectively disrupt the radar angle
tracking function, still it perturbs the tracking loop. When the perturbation caused
by the sweeping frequency is detected, the ECM signal modulation sweeps over a
very limited portion of the sweep range. Continuous modulation of the ECM
signal about the detected frequency eventually drives tracking loop off the target.
The J/S required to have effective angle ECM is 10 to 20 dB, when the ECM
signal and the true target return lie in the same range gate/velocity tracking gate.
However, if the RGPO/VGPO programs are successfully employed to move
through range or velocity tracking gate away from the true target position, J/S of
3dB will be adequate. Having captured the range gates/velocity gate, the ECM
signal enjoys infinite J/S for an effective angle jamming.

        Using the above ECM techniques against monopulse (amplitude and phase
comparison) radar is ineffective, since the single returned pulse is less effective to
inverse gain and swept square wave AM jamming techniques. Jamming
sequential lobing radars is completely ineffective. However, Cross Eye ECM or
Cross-Polarization ECM technique will be effective against monopulse radars.
The cross-polarization ECM technique will be effective in breaking the angle
tracking loop if the J/S is 30 to 40 dB and if the intercepted radar signal and ECM
transmitted signals are completely orthogonal (within 1°) in polarization to that
received In the combination mode with RGPO/VGPO technique, the requirement
on J/S will be less than 30 dB. In phase comparison monopulse, the error signal is
inverted, because of 180° phase shift in the signal at one of the paired elements in
the radar antenna. This results in distortion in the angle tracking loop error signal
pattern.

           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                        .
                                    XIII-13
              Another important ECM technique to counter monopuse radar is Cross-
      Eye Technique. Cross-eye jamming is an angle deception ECM technique that
      employs two spatially separated jamming sources. Each source acts as a repeater-
      type jammer transmitting the same signal at the same time and if the two signals
      arrive at the radar antenna approximately 180° out of phase, wave-front distortion
      occurs. This creates null at the radar antenna when it is on bore sight on the ECM
      vehicle. The on board radar, presuming that the signal source lies along the
      normal to the wave-front, tries to re-align its antenna at right angle to the distorted
      wave-front. This antenna re-alignment results in incorrect radar angle tracking
      which in turn results in incorrect track information

4.0   Suitable Jamming Techniques against different Radars :

              A list of common ECM techniques which will be effective against
      different Radars is given below.

       1. Search Radar                   Spot Noise, Range false targets,
                                         Cross polarisation, alternate polarisation
                                         ( for Radar with SLB )
       2. Radar with Freq. Diversity     Barrage noise, Swept Jammers ( a few radars
          or different freq. radars      operating simultaneously at different swept
                                         velocities will generate twinkle jamming at high
                                         probability), Noise blinking.
       3. General Radar                  CW noise jamming
       4. Track Radar                    AGC deception, Data rate countdown,
                                         Co-operative jamming, Cross polarisation.
       5. Pulse Track Radar              Passive Angle track deception, RGPO,RGPO
                                         with AM, Range false targets, Cover pulse.
       6. Doppler Radar                  Doppler noise, Hopping false Doppler signals,
                                         Doppler False targets.
       7. Doppler Track Radar            VGPO|I, VGPO/I and chaff, Doppler false


                CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                             .
                                         XIII-14
                                     targets with chaff corridor.
8. Mono Pulse track Radar            Cross-eye, Cross- polarisation, Delta jamming,
                                     Image Jamming, RGPO with data rate count
                                     down, skirt freq. Jamming.
9. Pulse Doppler Radar               Range and Doppler false targets, Head to tail
                                     pulses
10. Conical scan track radar         Conical scan AM, Scan rate modulation, Swept
                                     Scan Rate modulation, Bounce jamming, jog
                                     detection, Pseudo Random AM.
11. PC Chirp Radar                   In Pulse off set freq., In pulse AM, In pulse
                                     swept freq. ,Leading edge distortion, cover
                                     pulse
12.Track radar acquisition           Noise blanking
    mode
13.Coherent search or track          False targets
    radar
14. Doppler Missile                  Narrow band AM / FM noise

           In general, the jamming effect can not be achieved by using a single ECM
technique. Suitable combination of techniques is to be employed depending on the
Radar processing and tracking methods. The Radar signal characteristics are
measured with the        help of    Electronic Support measuring equipment.     The
parameters that determine the signature of the enemy Radars are stored in threat
library.     Signal processing system compares the measured parameters with the
threat library data and identifies the threat Radar. Suitable combination technique
can be automatically selected by using a definite strategic method which can
produce a strong effect on the Radar system. For this purpose, suitable jamming
techniques are to be programmed so that the system can automatically generate
the effective jamming technique, the moment the threat radar signal is detected
and identified.


            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         .
                                     XIII-15
5.0   Combination of ECM Techniques :
             A list of few combinations of ECM techniques which will be effective
      against some radars is given below.

       S.no Combination ECM techniques                     Radar type
       1.    Velocity Gate Walk Off with Doppler 1. CW Doppler track radar
             false targets                                 2. PC Chirp radar
       2.    Range and Doppler false targets               1. Search radars, track radars,
                                                              missile guidance systems,
                                                              missile mono pulse track
                                                           radar
       3.    Synchronised RGWO & VGWO                      Pulse Doppler radars which
                                                           compares target Doppler and
                                                           range rate of a target.
       4.    Noise with Blinking                           Search radar with conical scan,
                                                           Acquisition mode of track
                                                           radar, some mono pulse track
                                                           radars, CW homing missile
       5.    RGWO or VGWO with Data rate Mono pulse track radar,
             reduction                                     Conical scan tracking radar
       6.    VGWO with chaff corridor Doppler Doppler track Radar
             false targets with chaff corridor
       7.    Noise with alternate polarisation             Search radar with Side lobe
                                                           blanking
       8.    VGWO with terrain bounce                      CW Doppler homing missile
       9.    VGWO        with   wobbulated         cross Semi-active          CW     Doppler
             polarisation                                  homing missile
       10.   RGPO with conical scan square wave Conical scan tracking radar
             noise
       11.   Cross polarisation and cross eye              Mono pulse track radar ( cross


               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                            .
                                        XIII-16
                                                              polarisation at far ranges and
                                                              cross eye at near ranges )



6.0   Jamming Techniques Vs Anti Jamming Techniques :



             In many cases, the types of the enemy radar, especially various anti
      jamming methods employed by them, are not easy to get quickly distinguished
      according to the received radar signal. And the side of radar always uses the most
      advantageous radar types and optimum anti jamming methods to achieve
      optimum results of battle under the basis of known jamming methods. This needs
      the game theory method to determine the jamming techniques to be used. The
      optimum tactics can be employed under circumstances of game-theory model
      using pure tactics, but this kind of model only has mixed tactics and often has no
      pure tactics. Thus, radar jamming systems separately need to use different
      jamming methods at a definite probability so that the battle efficacy            ( such as
      probability of survival of carrier planes ) achieve a definite numerical value.

         Jamming              Anti Jamming               Enhanced Jamming technique
         technique              technique
      Noise Jamming :
      1.Side lobe       1. Side lobe blanking       Alternate polarisation within a PW
        jamming         2. Ultra Side lobes         Co-operative      jamming     from     many
                                                    sources
                        3. Adaptive arrays          Electronically steerable antenna


      2.Mainlobe        1. Frequency agility        High Speed set on receiver,
       jamming                                      Barrage noise
                        2. Frequency diversity      Barrage noise, Sweep through noise
      Deception Jamming :
      1. RGPO           1.Leading edge tracking     RGPO with DRFM, long rise time cover
                                                    pulse

                CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                             .
                                         XIII-17
                        2. PRI Jitter &            Repetitive noise, Repeater systems
                          Freq. tagger             ( coherent or non-coherent )


                        3.Comparisionof Doppler    Synchronously using RGPO &
                          freq. & range rate       VGPO with DRFM


                        4. Pulse Doppler radars,   RGPO / VGPO with coherent ECM
                                                   system ( DRFM ).
                        5. Chirp radar             Leading edge distortion, Cover pulse, In
                                                   pulse offset freq., In pulse swept freq.,
                                                   Doppler false targets
                        6. Coherent radars         VGPO/ I with coherent ECM system like
                                                   DRFM
      2. Conical scan   1. LORO or COSRO           Swept square wave with or without log
        modulation                                 detection, Data rate reduction
        or Inverse      2.Mono pulse track radar   Cross polarisation, cross eye, wobbulated
        gain angle                                 cross polarisation, noise with blinking,
        jamming                                    Terrain bounce, Data rate reduction with
                                                   some gate stealing technique, Skirt
                                                   jamming




7.0   Case study – Dynamic Aircraft engagement :

             In DLRL, we have designed and developed an ECM system as a part of
      development of an integrated Air-borne Self Protection Jammer system for Indian
      Air Force. The effectiveness of the system is studied against an Airborne Pulse
      Doppler Radar. For this purpose, an experimental test set up was made with the
      actual radar in an anechoic chamber as shown in Fig.1 This is a cost effective
      method for testing and evaluating the ECM effectiveness by simulating real –
      time engagements.     This can also help us to establish suitable ECM strategic
      techniques against Air-borne Pulse Doppler Radar.

               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                            .
                                        XIII-18
Test Procedure :- Refer to experimental set up shown in Fig. 1.

       A portion of Radar output is tapped and fed to a Computer Controlled
Range and Doppler simulator. The output signal of this unit resembles echo
signal of a moving target. The signal through a variable attenuator was fed to a
horn antenna, which is kept at a distance of 10 meters. The power level of the
signal is adjusted such that the intensity of the echo signal received at the Radar
receiver input is about –90 dBm.

The radar can operate in two modes. 1) Head – on mode , 2) Tail – on mode.


       In Head–on mode, both radar and target move in opposite directions
approaching each other. As the Radar and the target approach each other, the
Doppler frequency shift in the Radar transmitted frequency is always positive. In
tail-on mode, both move in the same direction (i.e. the jammer follows the Radar
from behind).


       Initially, Head-on mode was selected and the parameters like velocity of
the aircraft and Range were fed into the system. When the Radar was switched
ON, it enters ‘scan mode’ of operation to search for the target. The Radar
receiver receives the signal transmitted by the horn antenna and detects it. An
indication appears on the display and the operator can lock on to the target with
the help of a joystick. Then the Radar enters to ‘LOCK-ON mode’. The Radar
tracks the target until the range becomes zero when the break lock occurs
automatically. This shows the health of the Radar system.


       The output of the Range and Doppler simulator is amplified and fed to
Radar Warning Receiver of the EW system. This system measures the Frequency,
Pulse width, Pulse Repetition Frequency and amplitude of the signal. Then the
system matches measured parameter values with threat library data and identifies


         CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                      .
                                  XIII-19
the signal. Then it displays a symbol on the ‘Display Unit’ which indicates type
of Radar signal being received by the system.

          A portion of the Range and Doppler simulator output is fed to the ECM
system, through a SPST switch, for which effectiveness evaluation is being made.
The parameters for generating various ECM techniques were stored in a Pre-flight
message data and is programmed into the system before start of the experiment.
Different ECM techniques can be invoked with the help of control panel in Radar
Warning Receiver.


          Once the Radar enters Lock-on mode, the SPST switch was turned /
switched ON. The ECM receiver detects the signal and starts transmission after
applying the ECM techniques on to the received signal. The output of the ECM
system is fed to the same horn antenna through a variable attenuator. The jammer
power density at the Radar receiver was controlled using this attenuator. This
facilitates us to study the effect of ECM technique on the Radar for different J/S
values.


          By using different ECM techniques, the Jammer effectiveness was
evaluated by observing its effect on the Radar display. The effect was observed
in three ways.


          1.        Break lock, in which the Radar breaks the lock on to the target and
                    is forced to enter       ‘search – mode’.       In this condition, the
                    operator/pilot can not launch any missile and has to re-acquire the
                    target.
          2.        Hanging, in which the radar display              hangs indicating the
                    saturation of radar computer. The radar can no more track the
                    target and the operator / pilot is forced to break the lock manually
                    and enter search mode.


               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                            .
                                        XIII-20
              3.        Range errors, in which the display indicates the false range of the
                        target and produces large range error. This increases the ‘Miss
                        Distance’ when a missile is launched.


              The experiment was conducted with various Range and velocity gate
      stealing ECM techniques like RGPO, VGPO, VGPI along with and without using
      angular detection technique like DRR & SSRM. Doppler band noise technique is
      also applied against the radar.


              The experiment was also conducted at different ranges and velocities of
      the aircraft for different J/S values.


              Several experiments were conducted to find out the best suitable ECM
      techniques and the ECM effectiveness is also evaluated.


              A similar kind of study was carried out for Tail-mode of operation and the
      best suitable technique is evaluated.


              It was observed that for suitable ECM techniques, the effectiveness is
      increased from 10-20% to 90 – 95%. Some ECM techniques could produce only
      partial effect and the probability survival depends on the target detection
      capability of the pilot under jamming environment.


8.0   ECM Effectiveness – Present Trends :


              The ECM effectiveness is not only evaluated through simulators. Various
      present jammer systems are built with automatic re-programmability to achieve
      better effectiveness. The present trend of the software and implemented systems
      are discussed here.



                   CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                                .
                                            XIII-21
Jammer Effectiveness Model :


       This Jammer Effectiveness Model (JEM) is developed by ITS for US
Army in windows 95/98/NT environment to evaluate EW scenarios for radar and
communication systems analysis.


       Jammer Effectiveness Model (JEM) is computer based ECM effectiveness
analysis software module used to know the performance degradation on radar
systems. This model is a very flexible analysis tool and can be used to perform
many different types of analysis, because it is highly structured and modular in
design. ITS has developed        two versions of JEM for systems performance
evaluation in an electronic warfare environment: one for radar analysis and other
for communication analysis. The frequency range of the JEM is currently 30 MHz
to 20 GHz in radar version and 2 MHz – 300 GHz in communication version.


       Each version of JEM includes a user- created catalogue of equipments,
ground stations, aircraft and satellite platforms. The software for creating and
maintaining this catalogue is a climatological database which includes subroutines
for use in calculating clear air attenuation, rain attenuation, diffraction losses and
tropo-scatter losses.


       There are many present jammer systems incorporating Jamming
Effectiveness Analysis in their onboard computers. ELT/553(V)-2 Airborne pulse
and CW jammer, AN-ALQ-187 Internal countermeasure system, AN/ALQ – 131
Self protection jammer pod etc. are few examples.




          CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                       .
                                   XIII-22
9.0    Conclusions :


              In this chapter, ECM effectiveness against various types of radars is
       described. A case study on ECM effectiveness for dynamic aircraft engagement is
       described, which is a cost effective method for testing and evaluating the ECM
       effectiveness by simulating real – time signals.


              Due to the integrated modern types of radar network in the hostile
       environment, it is essential to have integrated ECM systems with computer
       control jamming techniques with associated Real Time Embedded Software and
       RF signal processing features. Digital Radio Frequency Memory (DRFM)
       integrated with processor based Techniques Generator can play an active role in
       the design of an effective ECM System.




10.0   References :


       1. S.M. Sherman, “Monopulse Principles and Techniques”, Artech House, 1984.
       2. R.N. Lothes et al, “Radar Vulnerability to Jamming”, Artech House, 1990.
       3. Edward J. Chrzanowski, “ Active Radar Electronic Countermeasures, Artech
          House,1995.
       4. Li Neng-Jing and Zhang Yi-Ting -A survey of Radar ECM and ECCM,
           LEEE transactions on Aerospace and Electronic systems, vol.31, No.3
           July,1995
       5. Van Bruntt, Applied ECM, Volume III.
       6. J.A.Boyd, D.B.Harris,etal “Electronic countermeasures”.
       7. D.Curtis Schleher, “ Introduction to Electronic Warfare” ,Artech House,1984.
       8. David Adamy , “EW101 A First Course in Electronic Warfare” ,Artech
          House2000.

                CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                             .
                                         XIII-23
                                                                                    RWR

                                                                                 CONTROL
                                                                                   UNIT
         RANGE &
                       VAR.
         DOPPLER                       PD                        DC
                      ATTEN                        AMP.
        SIMULATOR


                                                                                  SPST        ECM      STEP
                                                                                  S/W        SYSTEM   ATTEN.




                                                                                                ISO
RADAR                                                                 PD             PC
                                                                                                ISO
DISPLAY
                RADAR ANT.                     ECM ANT.

                                                                 SPECTRUM
                                                                 ANALYSER




    Fig. 1 TEST SET - UP FOR EVALUATION OF ECM SYSTEM


                              CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                                           .
                                                       XIII-24
           CHAPTER – XIV


TEST AND EVALUATION OF EW
          SYSTEMS
  Shri. S. VIJAYA KUMAR, Sc ‘F’




    CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                               XIV
                         CHAPTER - XIV


      TEST AND EVALUATION OF EW SYSTEMS

                            CONTENTS


1. INTRODUCTION
2. OBJECTIVES
3. DESCRIPTION OF EW T & E PROCESS
4. EW T&E PROCESS
5. INTEGRATED PROCESSES
6. TOOLS
7. EW T&E RESOURCE CATEGORIES
8. T&E RESOURCE CATEGORY EXAMPLES
9. CASE STUDY
10. ECM SYSTEM TESTING
11. CONCLUSIONS




       CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                  XIV
         TEST AND EVALUATION OF EW SYSTEMS
                         S. VIJAYA KUMAR Sc ‘F’

                               DLRL, Hyderabad

1.    Introduction:

       Test and Evaluation is a disciplined method of predicting and measuring,
with known confidence that a system will function as intended in the expected
operational environment. The Phrase “with known confidence“ is not only
included to assure acquisition authorities that project milestones are being achieved
but also it is crucial to a War- fighter, who must act on reliable and realistic T & E
results when about to risk lives and high costly military platforms in a hostile
engagement. To evaluate means to establish worth or value by analyzing the data.
The Test and Evaluation go hand in hand.

       The T & E asks the question “ What the system is expected to do and
does it do.?” The need for a disciplined approach to T & E is therefore paramount –
so that outcome can be expressed objectively. Denial of problems at an early stage
simply means that, it is either costly to fix them or unaffordable to fix them at the
actual operational scenario.

       Test and Evaluation of EW systems require a disciplined application of the
scientific method, modern measurement systems and rigorous analysis of the
resulting data products. As a result, an improved T & E process has evolved that is
characterized by a Test- Evaluate- Compare strategy rather than the previously
practiced Fly- Fix- Fly approach (e.g. for an air borne platform).

        With the rapid evolution of military electronics and computer science, the
range, complexity, and sophistication of EW systems have grown significantly. It
would, therefore, be impractical to cover all aspects of EW systems testing in a few


           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                    XIV - 1
      pages. It is however, desirable to cover the core T & E concepts, procedures, and
      techniques that apply generically to most of the EW systems.

            This paper deals with the T & E process applicable to a generic EW scenario.
      The objectives, the need, resources, tools, process, testing and evaluation         are
      covered. To give more emphasis to the T&E process, a typical Air borne Electronic
      Attack system is considered as a case study.


2.     Objectives: The primary objectives of the EW T&E Process are



             •    To reduce the risk of hidden flaws in the product that will be very costly
                   to fix later
             •    To demonstrate system performance that proves new and modified
                   systems are being properly developed/improved and will meet the needs
                   of the user
             •    To contribute timely, accurate and affordable information to support life
                   cycle acquisition and support decisions.
             •    To provide a quality product to the user



 3.    Description of the EW T&E Process


                 Predict, Test, and Compare: The EW T&E Process is built upon the
                 three pillars of Predict, Test, and Compare and its foundation is to do
                 ground testing before flight-testing. Tools are used with the process to
                 plan, execute and record T&E efforts. This is illustrated in figure 1.




                   CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                            XIV - 2
Figure 1. EW T&E Process Structure


        The Predict pillar is supported by Modeling and Simulation (M&S) tools. These
are used to help develop test concepts, and predict test results prior to testing. A model of
the EW System Under Test (SUT) interacts with other models at various levels of
performance to estimate and predict system performance. The EW system model will
usually be a Digital System Model (DSM) that can be run on a digital computer with
other digital models such as platform models, threat models, environmental models and
scenario models.


        The Test pillar is supported by test facilities and reporting tools. These are used
to produce and record data and information derived from the tests. Test facilities will be
selected to be efficient and cost effective. Test reports must be concise, timely and
contain information needed by decision makers. Risk areas identified in pre-test planning
                   CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                            XIV - 3
must be addressed. Test reports will be written by the Responsible Test Organizations
(RTO) and Operational Test Agencies (OTA).


        The Compare pillar is supported by the T&E record tools. These are T&E
requirements, documents, test data, plans, evaluations, test results and summaries. The
T&E record is implemented in the form of a Test Process Archive (TPA) that is used to
track test progress as the EW system matures.

        In Summary, the EW T&E Process uses M&S (DSM), Test Facilities and
Reports, a T&E record in the form of a TPA and a six step disciplined test process (see
figure 2).


4.      EW T&E Process: Figure 2 illustrates the six steps of the process.




Figure 2. The EW T&E Process



4.1          Determine Test Objectives--. The first step of the process is to determine the
test objectives. This is an action step. This step defines

                  •    The technical and operational issues that must be proved
                  •    Risk areas
                  •    T & E information needed by the decision makers
                  •    Underlying assumptions

This step is the start of early test planning to include a test strategy, modeling and
simulation analysis / support requirements and data management.

                      CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                               XIV - 4
       To accomplish this, several source documents are required as mentioned below.

       1. Mission Need Statement (PDR document)
       2. Operational Requirements Document (ORD)
       3. System Threat Assessment Report (STAR)
       4. Cost & Operational Effectiveness Analysis (COEA)
       5. Concept of Operations (CONOPS)
       6. Design and Performance Specifications


       These documents will provide details on the user requirements and the threats that
may be encountered by the system, once it is deployed. From these, the developer, user
and test organizations will derive detailed specifications and test requirements for the
system.

The EW T&E Process will be severely impacted by poorly defined user
requirements.


4.2    Pre-test Analysis. Once the test objectives have been identified, there is a need to
determine the overall test concept and exactly what will be measured for each objective
& how it will be measured. This is an action step. Pre-test analysis, is used to predict the
results of the test in terms of events that will occur and to predict values for system
performance. Pre-test analysis is also used to determine test conditions and sequences.


Issues that should be addressed are:


                   •   Designing the test scenario
                   •   Setting up the test environment
                   •   Controlling the test resources
                   •   Sequencing the test trials
                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                          XIV - 5
                   •   Determining the types and the quantities of data
                   •   Predicting the outcome values for each of the objectives.


       A data management and analysis plan should be developed to define test data
requirements, determine how the test team will analyze data, and identify specific M&S
requirements. The analysis plan part should include the statistical requirements for
establishing performance measurements. By performing this analysis, a better
understanding of risk areas can be achieved, resulting in a test program that will find
problems not previously discovered and provide the required data for decision makers.
Modeling and simulation tools can aid in this effort.
       One outcome of the pre-test analysis could be the discovery that current test
resources are not available to accomplish the desired testing. In that case, alternatives or
develop needs to be defined.


4.3    Test--This is an action step. Test includes many activities ranging from


                   •   Test planning
                   •   Test conduct
                   •   Data management
                   •   Test reporting.


Test planning is necessary in order for test conduct to take place in an orderly and
efficient manner. However, it also should be done to define risk areas where early testing
is needed to identify problems so improvements can be incorporated while economically
feasible. In performance of test planning duties, DT&E and OT&E personnel will work
together as part of a test planning team. This will ensure they are minimizing duplication
by using common data requirements, T&E resources, analysis tools and instrumentation,
as much as possible / practical.



                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                          XIV - 6
       Detailed test plans are used by Testing Agency to plan and execute the actual test
conduct. These evolve from pretest analysis work, continuing modeling and simulation
work, and test planning documents such as DT&E and OT&E Test Plans, Program
Introduction Documents and Statements of Capability Detailed Test Plans define specific
test objectives, test conditions, test article configurations, numbers of tests, and test
events. They focus on testing in an orderly, efficient, and safe manner.


        The primary organizations involved during this step are the System Program
Office (SPO), RTO and OTA.


       Data management includes all the data handling and processing tasks required.
This starts with collecting the raw test data, converting it to engineering units, analyzing
& validating the data, and finally getting it to the people and organizations that need part
or all of it for information and storage.


       System deficiencies identified during this test step, and the following evaluation
step, will be documented and processed. The Deficiency Reporting (DR) system provides
a systematic way to document and validate problems. Then it must be used to investigate,
track, and resolve problems.


4.4    Evaluate--This is an action step. In figure 2 the evaluate step is shown following
test; however, test and evaluate go hand in hand. Test data must be evaluated to
determine if the predicted results were achieved. Where differences are found, evaluation
must determine if the differences are due to errors in pre-test analysis, flaws in test
design, or failures in system performance. Actions which address analysis errors, test
design flaws and system failures are to be properly addressed. It is important to review
the data as soon as it is available, to determine data quality and validity.


       The evaluation should result in conclusions and recommendations. Conclusions
should correspond directly to the test objectives. Recommendations should focus on both
                  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                           XIV - 7
performance and operator issues. Recommendations should be made 1) regarding the
ability of the system to accomplish its most critical task and 2) regarding implications of
the T&E on the next planned step for the program.


         Evaluation is not complete until all test data and objectives have been analyzed,
and any differences between predicted and measured values have been resolved. Input of
this step is data from the simulations and testing. Output is information for the customer
and decision maker. In this step digital system models and computer simulations should
be updated. A record of system performance prepared, dated, and placed in a test
program information archive called the Test Process Archive.


   4.5    Acceptable Risk--This is a decision step. This step is a yes/no decision by a
   decision maker that the test outcome was either satisfactory or not satisfactory. This
   step address the questions like


            •   Was the test outcome satisfactory?
            •   Have technical and operational risks been reduced to acceptable
                levels?
            •   Have the user’s needs are met?


         If the test results are not satisfactory, an unacceptable risk may exist in the
product until proven otherwise. Failures in DT&E may be serious but corrective.
However, failures in OT&E may result in program cancellation with no possibility for
corrective action.


         The test manager determines if the test demonstrated the objectives, insures test
adequacy, and then makes a recommendation. The acceptable risk decision is made by a
decision maker--the program manager (PM) or higher authority.



                  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                           XIV - 8
        If an acceptable level of risk was not achieved, it must be determined if it was
due to the system design, the test method, or if the pre-test predictions were in erroneous.
When potential solutions are found, the system should be retested until the decision
maker is convinced that an acceptable level of risk has been achieved.


4.5    Improve--This is an action step. It defines what must be changed or refined? Who
must take corrective action? These are actions to improve the EW system design, correct
a flawed test method, find and fix errors in models and simulations or improve the test
process.


       If problems were discovered during testing, they could be in the system design,
the test method, or flawed predictions. Analyzing and fixing these problems is the key to
reducing product risk. The benefit of the EW T&E Process is that feedback from the
evaluation step is available early, when design changes can be economically
incorporated. As development/testing of the product progresses, predictions / measures of
performance and effectiveness are verified / improved.


       Deficiencies found in models and simulations used in the pre-test predictions
must be brought to the attention of the responsible organization. Failure to respond to this
feedback information in DT&E can adversely affect OT&E and a program’s ultimate
success.


4.6    Quality Product--Low Risk: Properly employed testing at appropriate times will
reduce risk by identifying areas early that need improvement. The biggest benefit of a
well-run test is the customer’s confidence in the results and his appropriate reaction to the
feedback provided. Predict-test-compare … then fix, is the way to transform a product
with risk into the robust, low-risk product the user wants and needs … a quality product.

5.     Integrated Processes. The EW T&E Process supports and must be integrated
with the more complex Acquisition Process. The EW T&E Process interfaces with the
                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                          XIV - 9
system acquisition process as shown in figure 3. The user defines system requirements
and deploys the system after development. The SM controls program specifications,
design and production. The Responsible Test Organization (RTO) and Operational Test
Agency (OTA) are responsible for detailed test planning, conduct, evaluation and
reporting. Information must be developed and shared between user, tester, and acquisition
communities. Responsibility for the product rests with the SM who depends upon the
RTO and OTA to provide needed test information. EW testing requires an integrated
effort (teamwork) to get a quality product with low risk that meets user needs.




Figure 3. Integrated Effort.




6.     Tools. Modeling and Simulation, Test Facilities, and a T&E Record are required
tools to make the EW T&E Process to work. Modeling and simulation is used for analysis
and predicting outcomes. Test facilities will first be used for ground testing, and then
flight-testing. A Test Process Archive (TPA) will be used to maintain a complete history
of the T&E and permit traceability back to user requirements.




                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         XIV - 10
6.1     Modeling and Simulation (M&S): The EW T&E Process uses computer-aided
simulations and analysis prior to testing. This process is used to help design tests and
predict test results and after testing to extrapolate test results to other conditions. M&S
should also be used to provide constant feedback for system development/improvement.
M&S use should start as early as possible during concept exploration and continue
through open air testing. M&S is not limited to DT&E; M&S applications can be used to
assist in planning OT&E assessments and to replace certain portions of resource-rich
operational testing.

6.2    Test Facilities: EW systems require the use of both ground test facilities and
open-air range facilities for DT&E and OT&E. These facilities are tools, which provide
the data and test information to verify performance, risk mitigation and operational
suitability. Selecting appropriate facilities for specific EW T&E efforts requires
knowledge of the capabilities of such facilities.

6.3    Test Report: Output from these test facilities will be data and information
summarized in various test reports. The implementing command should use the test
results to guide design and development decisions. Decision makers need the results for
program reviews and briefings for milestone decisions. The user also needs test results to
determine if the system’s effectiveness and suitability will meet their requirements. A
number of other test reports are generated during the life of an EW program. For each
T&E effort the customer and RTO or OTA agree on the number of reports and the report
formats in pre-test planning meetings. One or more of the following reports is typically
required:

6.4    T&E Records: It is recognized that during the life cycle of an EW system, the
program will evolve and change. Technical considerations, schedule requirements,
budget realities, facility constraints, and decision makers will impact the T&E. Thus an
important part of implementing the EW T&E Process is maintaining a record of all T&E
associated with each EW system, the budgets, decisions, and reasons for the way the
T&E was planned and executed.
                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         XIV - 11
        To record this history each EW T&E effort will establish and maintain a Test
Process Archive (TPA). The TPA is a file of information and documents of all the T&E
efforts of an EW system for the life of that system. It consists of the T&E Structure, Test
Data Collected; Test Plans, Evaluations, and Results

Additional Benefits Derived from Use of the EW T&E Process:

         •     Early and thorough evaluation of system concepts.
         •     Early feedback to the design process.
               The creation and evolution of test requirements through rigorous analysis
               and evaluation.
         •     The identification of performance parameters that is critical to operational
               effectiveness.
         •     Establishment of validated linkages between operational requirements and
               test criteria.
         •     Timely and credible test results to support milestone decision-making.
         •     A closer tie between intelligence analysis, systems engineering, test
               facilities and test agencies.
         •     Early identification and acquisition of test assets.

7.0     EW T&E Resource Categories. Basically there are six general categories of
T&E resources. These are Modeling & Simulation (M&S), Measurement Facilities (MF),
System Integration Laboratories (SIL), Hardware-In-The-Loop (HITL) Facilities,
Installed System Test Facilities (ISTFs), and Open Air Ranges (OAR). Proper selection
and use of these resources (facilities and capabilities) is an important part of the EW T&E
Process. A thorough understanding of these categories and their interrelationships is
necessary.




                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         XIV - 12
Figure 4. Resource Categories Support Test Execution.




7.1    Modeling and Simulation (M&S): Digital models and computer simulations are
used to represent systems, host platforms, other friendly players, the combat
environment, and threat systems. They can be used to help design and define EW systems
and testing with threat simulations and missile fly out models. Due to the relatively low
cost of exercising these models, this type of activity can be run many times to check
what ifs and explore the widest possible range of system parameters without concern for
flight safety. These models may run interactively in real or simulated time and space
domains, along with other factors of a combat environment, to support the entire T&E
process. Computer simulations are realized up to the following levels of technical
complexity:

        The objectives of modeling a parameter in the test process are to:
                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         XIV - 13
            •   Define safety footprints or limits.
            •   Extrapolate test data into un testable regimes.
            •   Increase sample size once confidence in the model is established.
            •   Define test facility requirements (e.g., number and types of threats,
                airspace required, control of background noise and emitters and
                instrumentation).
            •   Define and optimize test scenarios.
            •   Select test points (i.e., successful results would not indicate the need for
                additional heart-of-the-envelope testing).
            •   Predict test results for each test objective.


         These modeling objectives must be tailored to the test program and specific DSM
requirements identified.      The EW T&E Process uses computer-aided simulations,
analysis prior to testing to help design tests and predict test results. In this way M&S is
part of all six-resource categories. M&S should also be used to provide constant feedback
for system development/improvement.


Limitations:    a) Prediction of absolute performance/effectiveness with high confidence
                b) Achieving the same degree of fidelity as an RF simulator for certain
                complex functions


7.2      Measurement Facilities (MF). Measurement facilities establish the character of
an EW related system/subsystem or technology. They provide capabilities to explore and
evaluate advanced technologies such as those involved with various sensors and multi-
spectral signature reduction. Figure 5 illustrates aircraft radar cross-section measurement
range.




                  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                          XIV - 14
Figure 5. RCS Measurement



Measurement facilities generally fall into the sub-categories of antenna measurement,
Radar Cross Section (RCS) measurement, infrared/laser signature measurement,
Electromagnetic Interference and Electromagnetic Compatibility (EMI/EMC) test
capabilities. Measurement facilities provide EW and platform antenna pattern
descriptions, and platform signature data critical for system design and refinement,
computer simulation and HITL testing.


Limitations: a) simulation of electronic Warfare
             b) Evaluation of EW performance / effectiveness


7.3    System Integration Laboratories (SIL): Illustrated by figure 6, are facilities
designed to test the performance and compatibility of components, subsystems and
systems when they are integrated with other systems or functions. They are used to
evaluate individual hardware and software interactions and at times, involve the entire
weapon system avionics suite. A variety of computer simulations and test equipment is
used to generate scenarios and environments to test for functional performance, reliability
and safety. SILs are generally used to test weapon system specifications.




                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         XIV - 15
Figure 6. System Integration Lab.

       SILs often employ a variety of real-time/near-real-time digital models and
computer simulations to generate scenarios and multi-spectral backgrounds. These
models are interfaced with brass board, prototype, or actual production hardware and
software of the systems under test. SILs are used from the beginning of an EW system’s
development through avionics integration and fielding. Moreover, SILs continue to be
used to support the testing of hardware and software modifications or updates occurring
throughout an EW system’s operational life.


Limitations:   a) Evaluation of dynamic performance
               b) Evaluation of closed loop EW performance against threat
               c) Evaluation of EW system effectiveness


7.4      Hardware-in-the-Loop (HITL): HITL, illustrated by figure 7, forms             an
important test category because it represents the first opportunity to test uninstalled
system components (breadboard, brass board, preproduction prototypes, etc.) in a realistic
RF, Laser, or IR environment. HITL operating environments can provide: terrain effects;
high signal/threat density; realistic interactive scenarios; multi-spectral capability;
background noise; modern threat representation via closed-loop hybrid threat simulator
for effectiveness testing; man-in-the-loop interaction; and Integrated Air Defense System
(IADS) networking. Capabilities provided by the HITL test environment are:

                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         XIV - 16
secure/covert (shield/screen room); high data pass rate; test replay/repeatability; and high
capacity data collection and recording.

        Thus HITL facilities are indoor test facilities that provide a secure environment
to test EW techniques and hardware against simulators of threat systems. Primary EW
HITL facilities contain simulations of hostile weapon system hardware or the actual
hostile weapon system hardware. They are used to determine threat system susceptibility
and to evaluate the performance of EW systems and techniques.




Figure 7 Hardware-In-the-Loop Testing.

        HITL testing should be done as early in the development process as possible -
even if that means using a brass board configuration. EW HITL testing is done to provide
repeatable measurements and verification of protection techniques and EW system
effectiveness.



       Limitations: a) Testing comp ability with the host platform
                     b) Simulation of all flight environment aspects with high confidence


7.5    Installed System Test Facilities (ISTF). ISTFs, illustrated by figure 8, provide a
secure capability to evaluate EW systems that are installed on, or integrated with, host
platforms. These test facilities consist of anechoic chambers in which free-space radiation
measurements are made during the simultaneous operation of EW systems and host
platform avionics. The EW system under test is stimulated by threat signal generators and

                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         XIV - 17
its responses evaluated to provide critical, integrated system performance information.
Their primary purpose is to evaluate integrated avionics systems (e.g., radar, infrared,
communications, navigation, identification, EW systems or subsystems, integrated
controls and displays) in installed configurations to test specific functions of complete,
full-scale weapon systems. Such testing is done to determine if any EMI/EMC problems
exist; to determine system reaction to electromagnetic environments of hostile and/or
friendly systems whose signals cannot be radiated in free space.


       Problems not observable in other ground test facilities but crucial to system
checkout prior to open air testing. Failure to evaluate installed EW system performance
adequately on the ground results in significantly increased flight test cost and lengthened
schedules.




Figure 8. Installed System Testing.


Limitations:   a) Dynamic test performance in free –space environment
               b) Evaluation of closed loop performance against a threat in a free –space
               environment
               c) Evaluation of EW effectiveness




                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         XIV - 18
7.6     Open Air Range (OAR). Open-air range test facilities, illustrated by figure 9, are
used to evaluate EW systems in background, clutter, noise and dynamic environments.
Typically these resources are divided into sub-categories of test ranges and airborne test
beds.




Figure 9. Open Air Range Testing.

        Open Air Range EW flight test ranges are equipped and populated with high
fidelity manned or unmanned threat simulators. The high cost of outdoor threat
simulators limits current range testing to one-on-one, one-on few, or few-on-few
scenarios. Open Air Range testing includes the subcategories of ground test, test track,
and flight test. The primary purpose of open air testing is to evaluate the system under
real-world representative environment and operating conditions. Open air range testing is
used to validate system operational performance/ effectiveness at a high level of
confidence. EW components, subsystems, systems, and entire avionic suites can be
installed in either a ground or airborne test beds or in the intended operational platform
and tested in open-air ranges. Real-world phenomena encountered during open air range
testing include terrain effects, multi-path propagation, and electromagnetic interference
from commercial systems (television and radio broadcasts, micro-wave transmissions,
etc.). Flight test ranges also offer the capability to conduct tests using captive carried and
live-fired missiles.

Limitations:           a) Achieving battle field threat densities

                  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                          XIV - 19
                         b) Scenario flexibility and statistical repeatability
                         c) Relatively high cost per test

8      T&E Resource Category Examples.



        Relative Cost. In general, the cost per test becomes more expensive as the testing
moves to the right as shown notionally in figure 10. The use of models, simulations, and
ground testing can reduce overall test cost, since open-air flight tests are the most costly.




Figure 10. Relative Cost--T&E Resource Utilization.


        Relative Use. Due to the complexity of EW systems and threat interactions,
modeling and simulation can be used in a wide range of progressively more rigorous
ground and flight test activities. Figure11, also notional, shows that modeling and
simulation and measurement facilities are used throughout the test spectrum. It also
shows how the number of trials/tests should decrease as the testing proceeds to the right
through the categories




                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         XIV - 20
Figure 11. Relative Use--T&E Resource Categories.



9.     Case Study:

       T & E process applicable for a typical air borne Electronic Attack System

       Electronic Attack (EA) is the use of electromagnetic or direct energy to attack
personnel, facilities or equipment. There are three basic sub-divisions of EA Jamming,
deception, and direct energy weapons. Jamming is generally defined as deliberate
radiation, re-radiation, or reflections of energy for the purpose of preventing or reducing
an enemy’s effective use of electromagnetic spectrum. With recent advances in
technology and frequent use of spectra out side the electromagnetic range, this definition
can be extended to cover similar actions against infrared, (IR), ultraviolet (UV), and
electro- optical (EO) systems.

A simplified block diagram of the jammer is given below in Fig 12.



                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         XIV - 21
                        Processor                     Technique
                                                      generator

         Rx Ant                                                                   Tx Ant
                         Receiver                       Modulator
                                                        transmitter


Fig 12: Simplified Jammer Block Diagram

9.1      RF Electronic Countermeasures system Description:

      An RF Defensive Electronic countermeasures (DECM) system has four basic
      subsystems.

      1) Receiver 2) Processor 3) Technique Generator and 4) Transmitter/ Modulator.

A brief functional description of the basic subsystem is given below.

9.1.1    Sensors: The antenna receives the RF energy. Typically two antennas are
employed, one pointing towards the front and the other towards oft.


9.1.2    Receiver/ Processor: The Receiver/ Processor uniquely identifies a threat to
make the proper response. The Receiver / Processor must also “ track” the incoming
signals in time or Doppler position to know the targets “ location “ in range or velocity
relative to the aircraft. The Receiver / Processor along with the Technique Generator
must also generate “ techniques “ that will disturb the tracking of the threat. Major
categories of jamming techniques include range, Doppler, and angle techniques.


9.1.3    Displays & Controls: The pilot interface usually consists of a control panel for
selection of system operating modes and indicator lights, identifying the threat
environment etc.


9.1.4    Transmitters: The transmitter module amplifies the technique selected and is
connected such that it radiates the energy either through a separate antenna or uses the
receive antenna in a time-shared mode.

                   CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                           XIV - 22
10     Electronic Countermeasures System Testing:


        The basic object of the test and evaluation is to test and evaluate whether the
designed system is able to meet all the requirements of the user with a defined jamming
effective. How effect a jammer is against a specific radar can be measured by several
parameters like the J / S ratio or the miss distance (in case of a missile threat). Technique
effectiveness must be evaluated with respect to the specific radars to be defeated. These
evaluations vary in sophistication based on the data available about the threat radar.


       Certain critical capabilities of a jammer must be understood; their parameters are
measured and evaluated. These capabilities are: antenna performance, transmitter/exciter
performance, threat recognition ability, counter measure generation/ effectiveness,
multiple signal environment operation, interface, ground testing, and flight testing.


       Parametric ranges of the technique generator can be defined after the total
operational range of the threat is determined. To perform effectiveness evaluation, a
thorough understanding of the threat radar is required. Ideally, this would encompass first
having the complete documentation of the victim radar to analyze the signal processing
and tracking algorithms. Next, analysis on computer model of the radar is accomplished.
Then, laboratory testing is conducted to gain a broad understanding of the technique
ranges that are effective against the radar. Finally, flight tests against the actual radar are
conducted.


10.1   Jammer Parameters:


Some of the important parameters of a Jammer are listed below. These parameters should
be tested and evaluated before the flight test commences.

1      Frequency               :       F1 – F2

2      Antennas (Rx and Tx) :          a)       Type
                  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                          XIV - 23
                                       b)       Beam width
                                       c)       Gain
                                       d)       Polarization

3      Receiver                :       a)       Sensitivity
                                       b)       Dynamic Range
                                       c)       Pulse width & PRI handling capability

                                                                        Spot
4      Techniques              :       a)       Noise                   Barrage
                                                                        Sweep Through

                                                                        Range
                                       b)       Deception               Angle
                                                                        Velocity

5      Transmitter             :               ERP

6      Platform                :                Aircraft / Helicopter

7      Environmental           :       a)       Temperature
       Condition                       b)       Altitude
                                       c)       Vibration Levels

8      Multi Threat Handling Capabilities

10.2   Additional Information required for T & E

       •      Scenario
       •      Operational requirement (Task Directives)
       •      Test Agencies
       •      Likely problems that are going to be encountered in real scenario
              (EMI / EMC, Multi Path etc)
       •      Mat lab program to calculate Path loss, Jammers effectiveness (J/S) ratio
              etc, Burn through range etc.
       •      Dispensing Mechanism: like Chaffs, IR Flares, and Decoys etc.
       •      Other on board avionics.
       •      Assumptions based on the previous experience on EW equipments.
       •      Test agencies (DGQA, RCMA, CRI)
                  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                          XIV - 24
10.3   Testing: The testing and Evaluation of the designed EW equipment can be
categorized as 1) Ground Testing and 2) Flight Testing.


10.3.1 Ground Testing: There are many different ground tests during the development
of a DECVM system


   a) Antenna & Radome Performance: Antenna performance is highly dependent on
the antenna position on the host platform. Actual antenna performance needs to be
characterized at a measurement facility,. The Beam pattern, Beam width, and the Gain of
the antenna are studied in the anechoic chamber, which simulates the free space
conditions. Similarly the deviations of the antenna performance are also to be tested and
evaluated when the antenna is covered with a Radome. It is also required to study the
transparency of the radomes for the EM enrage will be studied at the test facility.


   b) Receiver / Transmitter performance: Parametric testing of the receiver and
transmitter is usually conducted at the contractor’s facility but spot checks should be
done in the HITL to assure that the system is still performing as per the specifications. In
the HITL, threat detection, identification, technique generation, and multiple signal
environment capabilities are determined. The test begins by simulating the ECM system
with a single threat emitter to determine the one- on – one system identification,
operation. Techniques generated can be monitored on a spectrum analyzer. Techniques
effectiveness can be tested with closed- loop simulators. First in one-on- one and then in
a multiple signal environment.


       After the HITL tests are completed, the system moves on to an ISTF where the
ECM system is first tested on the actual platform on which it will be used.
Electromagnetic Interference tests are performed at this time. An ISTF is the best place to
perform avionics interface testing for the entire platform data base structure. Interface
testing involves exercising every link that will be operational during flight. Again, the
                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         XIV - 25
rule is to start slow. EMI and EMC tests are also performed at the ISTF. Installed antenna
patterns can be investigated in the ISTF environment. Some EW oriented ISTFs can
produce closed loop, high signal density, dynamic test scenarios


10.3.2 Flight Testing:


       Flight-testing is the final step in the testing process. The ECM system is installed
in an operational configuration and flown against the actual radars. Planning the flight
paths and maneuvers to take the full assets is a critical part of Open-Air Range (OAR)
flight tests. Instrumentation and the availability of data products may have a bearing on
which the flight profiles and test conditions are utilized. Initial flight may consist of only
one threat system being switched on at a time to get a base line capability and correlate
in-flight operation with the ground test observations. Effectiveness data can be collected
and should be compared against the laboratory results. More and more threats are then
switched on as confidence is gained that the system is performing, as it should be.

11.    Conclusion:


       Test and Evaluation plays an important role in the development of a product. It is
a systematic, methodical and logical sequence of various operations. Typically T & E is
accomplished as the design of a system matures from the component level to a fielded
system via sub-system, system and integration with other installed systems. A thorough
evaluation of the EW systems is not complete with out exploiting the maximum features
of the system. Though the paper describes the T & E process of a generic EA system, it
can be applied to the development of any branch of EW system (ES, EP).




                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         XIV - 26
            CHAPTER – XV


FREQUENCY MEMORY LOOP


Shri. Y. GOPALA KRISHNA, SC ‘E’




    CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                               XV
                       CHAPTER - XV




          FREQUENCY MEMORY LOOP




                          CONTENTS




1.    INTRODUCTION

2.    PRINCIPLE OF OPERATION

3.    THEORY OF OPERATION

4.    DESIGN METHODOLOGY

5.    DESIGN APPROACH

6.    CONCLUSION

7.    REFERENCES




     CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                XV
                   FREQUENCY MEMORY LOOP

             Shri Y. Gopala Krishna, Sc ‘E’, DLRL, Hyderabad.


1.     Introduction:


       Present day a war supremacy depends on supremacy on Electronic
Counter Measures (ECM). But the power, sophistication and dense deployment
of modern threat radar impose severe requirements of high ERP on noise jammer.
Under this condition deception jamming which requires significantly less energy
to counter a threat radar than an equivalent noise jammer, is the best choice. The
deception system in the form of repeater jammer radiates the replica of victim
radar’s signal, delayed in time to achieve range deception. The key to the
transmission of a signal that will cause the threat radar to lose the tracking, lies
within the ability of the Frequency Memory Loop (FML) device to store that
signal for a specified length of time while maintaining good spectral purity.


2.0    Theory of operation :


       Basically FML is an analog RF storage device, which operates by re-
circulating the RF signal in a closed loop. The main function of FML is therefore


               a. To store the input signal.
               b. To reproduce the signal with good spectral purity.


       It stores the signal by recirculating the RF signal with a definite delay in
each circulation which needs a delay line. It also needs an amplifier to
compensate the loss due to delay line. The amplifier and the delay line forms the
loop. The open loop gain of the FML is set well above zero decibels in the
absence of the instruction pulse inside the loop. Once the signal is detected, since
the open loop gain is more than zero decibels, the signal grows in each circulation
till the loop amplifier saturates limiting the growth. The saturation phenomena
             CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                       XV - 1
drops the loop gain such that the net loop gain becomes zero decibels to attain a
steady state condition provided the loop is stable. The configuration of FML is
shown in fig. 2. The FML essentially consists of an amplifier, a delay line, an
input RF switch to connect/disconnect the delay line in the loop and a directional
coupler.    Initially, the RF switch is connected to the input RF port and hence the
RF signal propagates through the loop amplifier and directional coupler and is
then divided into two paths, one path goes directly to the out put RF port, while
the other is coupled to the delay line of re-circulating memory loop. After a
predetermined delay, i.e. at the time when the RF signal through the delay line
arrives at the switch, the switch opens the input RF port and connects to the
feedback path. At this time the circulation loop is filled and the signal re-
circulates through the closed loop supplying RF continuously at the output port.
Relevant timing waveforms are shown in the fig. 3. This RF signal will be used to
transmit in accordance with the jamming technique. After the signal is transmitted
the RF switch disconnects the feedback path and connects to the input RF port.
This action terminates the storage and prepare the loop to receive and memorize
the next RF signal.


3.0     Performance of FML:


        Although Microwave Frequency Memory by circulation loop is a simple
concept, its dynamic operation introduces many interdependent relations that
govern the total performance. A brief review of the RF signal buildup and storage
will identify the principal factors limiting.


            1. Memory Loop RF Sensitivity.
            2. Memory duration
            3. Stability of output amplitude
            4. Frequency accuracy




              CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                        XV - 2
3.1     Memory Loop RF Sensitivity :


       The memory loop has a nominal open loop gain (OLG) of 10 to 15 dB
(say 15db). Introduction of an RF instruction signal begins the circulation process.
At the completion of the first circulation, the input switch closes the loop and
prevents the introduction of a new signal. A gain of 15 dB is achieved by the RF
instruction signal upon completion of the first circulation and an additional 15 dB
for every subsequent circulation until the power level is sufficient to drive the
loop amplifier into its nonlinear region. During this period, the loop noise power
is also building up at a similar rate, However, small signal gain suppression
occurs in the loop amplifier when it is driven into the non-linear region, thereby
inhibiting further noise power buildup. As the RF instruction signal’s power level
is reduced, more circulations are required before the signal reaches the power
level at which the amplifier is driven into its non-linear region. The noise power
will then reach a higher level before the small signal gain suppression mechanism
starts to take place. A point is reached where the noise power will also be at a
level high enough to have a suppression effect upon the signal, as the signal and
the noise compete the power available from the amplifier. The signal then loses its
ability to suppress the noise power-buildup and, in this case, noise will continue
to build up until it suppresses the signal. A 15 dB signal to noise ratio is a
practical guideline for the minimum S/N that can occur within a memory loop and
still insure that the loop amplifier is driven into the non-linear region by the
instruction signal and not by noise.


 3.2   Memory duration :


       The memory duration or Storage time would have been limitless if no
other signal including noise is present in the FML along with the instruction
pulse. A 15 dB S/N ration does not insure long-term memory duration. Random
noise capture of the memory loop will eventually occur. The number of RF signal
circulations (memory duration) before noise capture occurs is dependent on:
             CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                       XV - 3
          a. Noise power with in the loop,
          b. Small signal gain suppression provided by the loop amplifier,
          c. Open-loop gain,
          d. Open-loop gain fine structure variation.


          Noise power in a memory loop has frequency dependence due to non-
uniform amplifier noise figure, and open loop gain variation. If noise power with
in the loop is high comparative to the signal at any time, the memory frequency is
lost and the loop will re-circulate only those frequencies at which the noise is
high.


          The principal parameter that controls memory duration is small signal gain
suppression in the loop amplifier. When an amplifier is operating under saturated
condition, it exhibits small signal gain suppression so that signals other than large
experience decreased gains in the presence of large signal. This property of the
amplifier enhances the memory duration by suppressing the noise as long as it is
less than the signal. The input output characteristic of an amplifier indicating
large signal gain compression, noise depression and noise suppression is shown in
fig. 4.


          Fine grain gain structure of the loop amplifier also plays an important roll
in the memory duration. If there is no general loop-gain level or noise-level
advantage at any frequency in the band, the noise will tend to build up at gain fine
structure peaks, and in this case it will select preferred-mode frequencies in the
immediate area. Gain fine structure can be controlled by alignment with multi-
tuned equalizers. Open loop gain peak to peak fine structure variation should be
equalized to less than 1 dB to prevent noise capture.




               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         XV - 4
       According to the definition of memory time, it is defined as the time by
which the signal power falls by 3 dB from the initial saturated signal power. That
is the total noise power is half the total saturated power of the amplifier.


           The noise power at the end of n circulation is given by



                                   (GH)2n +2 - 1
               Non =        ( NiGF)²                   + (NeG ) ²(GH)2n
                                    (GH) ² - 1

               Where, Ni is the KTB Noise at the input of amplifier.
                       Ne is the external noise along with the signal.
                       G is forward gain of amplifier under saturation
                       H is the loop loss
                       F is the noise Figure of the Amplifier
                       G H is the loop gain
       The maximum number of circulations corresponding to the memory time
is ,when
                       Initial saturated signal power / Total noise power = 2


       If Ne the external noise is very much higher than the normal KTB noise,
       the Non can be approximated as
                       Non = NeG (GH)n
       And hence the maximum number of circulations ‘n’ can be estimated from
                       NeG (GH)n = Saturated power of amplifier / 2
       Finally the memory time T is given by
                       T = nτ
               Where τ is the loop delay time.


       It can be seen that ‘n’ depends on closed loop gain, initial noise, noise
figure and the forward gain. The most predominant factor is the closed loop gain
GH for noise. If GH = 1 the memory time is independent of ‘n’ and hence very
              CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                        XV - 5
long memory times can be achieved. If GH > 1 the noise goes on building as the
number of circulations are increased and ultimately the memory will collapse. For
the successful design of FML, the closed loop gain at all the frequencies other
than the instruction frequency also should be maintained at unity. This is possible
only with good small signal suppression characteristic of the amplifier in the loop.


3.3    Stability of output amplitude:


       Aside from the loop amplifier function in overcoming the loop insertion
loss, the amplifier properties such as small signal gain and noise suppression in
the overdrive mode are critical to long time storage. The amplifier is designed
with a small signal gain greater than the loop insertion loss. This allows an input
signal to be amplified on successive circulations through the loop until power
saturation occurs in the loop amplifier. This sequence of signal buildup is shown
in fig. 5. Final operation takes place at the output power point where the amplifier
gain equals the loop loss. As long as slope of the overdrive curve does not exceed
45°, the signal will buildup and lock on to the final operating point after a finite
number of circulations.


3.4    Frequency Accuracy :


       While storage time is an important measure of the effectiveness of an
FML, there is another figure of merit for the capability of the system to work
effectively in an ECM system. One can imagine an FML whose spectrum does
not change over the desired duration of storage time, but the spectrum is not
sharply peaked at the RF frequency of input pulse. The storage time of such a
device would be very long but the FML is ineffective against the threat radar
because there would be only a small fraction of the loop energy near the RF input
frequency. This is due to the characteristic of re-circulating systems, molding,
proves to be a more severe limit on achievable frequency accuracy in practical
systems. To determine the frequency accuracy of an FML, analysis of the output
             CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                       XV - 6
wave train, in time, must be made. As demonstrated in fig.6, the wave train enters
the loop, progressing through. At some point the wave has traveled all the way
around the loop, and reaches the point of entry. At this instant, the loop is filled
with RF, and the switch is thrown to close the loop. No more RF can then enter
the loop, and the wave inside the loop continues to travel, the entire pattern inside
the loop rotates clockwise. As a result, there is generally a discontinuity in the
waveform at the point of entry to the loop. Unless the length of the loop is exactly
equal to an integral number of wavelengths of the radiation inside the loop, there
is a phase jump at the junction between the leading and trailing edges of the pulse.


       If the frequency of instruction signal is such that the total electrical length
of the loop is an integral number of wavelengths


                   Nλ = LL


       Where N is any integer, λ is wavelength, LL is the electrical length of the
loop. Such frequencies are called preferred frequencies or preferred mode, then
each circulation is in phase, the frequency accuracy is maintained, and the FML
output spectrum peaks at the input RF frequency.


       However, if the frequency of the instruction signal is such that the loop
length is equal to an odd number of half wavelengths (non preferred mode)


                   (2N+1) λ/2 = LL


       Each circulation will be 180° out of phase. This situation is identical to the
instruction signal being 180° phase modulated at a modulation frequency equal to
1/2τ. Energy will then be concentrated in side bands approximately 1/2τ away
from the instruction signal frequency. The concentration of energy at frequencies
other than the instruction frequency is commonly called frequency inaccuracy.
The maximum frequency offset is ±1/2τ
             CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                       XV - 7
Improving Frequency Accuracy with Phase Shifter :


       Since the phase error is the source of frequency inaccuracy at non-
preferred modes, controlling phase error can improve the frequency accuracy of
the FML. An FML with phase shifter is shown in fig.7. The phase shifter with
180° phase shift causes the preferred frequencies of the loop to be shifted by 1/2τ
from their positions in the loop with out phase shifter. The preferred frequencies
are now determined by the equation,


                      (N+0.5) λ = L


       Where the 0.5 represents the contribution of the 180° phase shifter. The
frequency that suffers a 180° phase error in the FML has zero phase error in the
FML with 180° phase shifter, it is exactly on a preferred mode.


       Since the fixed phase shifter described above does not decrease the
spacing of the preferred modes in the FML spectrum, it does not increase
frequency accuracy over a large frequency band, it shifts the preferred frequency.
However, by properly controlling a phase shifter, it can be used in either of two
ways to improve the frequency accuracy of an FML.


       The simpler use of a phase shifter in an FML is the free-running phase
shifter. In this scheme, the phase shifter is repeatedly switched between 0° and
180°. The shifting is generally done in random fashion. With the use of 0° and
180°phase shifter the frequency accuracy improved to 1/4τ.


4.0   Design Methodology :


       For the satisfactory operation of FML, it is essential that the amplifier in
the forward path of the loop should be matched to the attenuation characteristics
             CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                       XV - 8
of the remaining part of the loop with tight tolerance limits of the order of one dB.
This ensures the closed loop gain of unity for the FML at all the frequencies when
it is operating in the saturation mode.


       If the shaping of the gain characteristics of the amplifier to match the loop
attenuation characteristics through out the band is not possible, the problem is to
be tackled by resorting to channelizing the total band into sub-bands. How ever in
each sub-band, the amplifier gain should be matched to the attenuation
characteristics of the loop. Sub-band to sub-band matching can be avoided. This
approach requires additional hardware and fast switching.


       The frequency accuracy of the memory loop is degraded by the phase
discontinuities produced by moding and the noise present in the loop. The
frequency accuracy will improve in a memory loop with increased in the loop
length and using low noise components. The accuracy can be further enhanced by
using a phase shifter in the loop and repeatedly switching between 0° and 180°.
The shifting is generally done in some random fashion to keep it unpredictable.
The input RF frequency in such an FML is never more than 1/4τ of the original
mode spacing away from a preferred mode. In a plain FML it can be 1/2τ a
spacing from the nearest preferred mode. This process improves the frequency
accuracy of an FML.


       The design of FML basically depends on system specifications and the
limitations imposed by the dynamic operation of the loop itself. The selection of
components and specifying their characteristics is the first exercise in the design
of a specific FML. Of all the components the loop amplifier, the delay line and
gain equalizer are the most important to consider.


       Loop Amplifier : In the early development of FML the TWT was the
only component for the loop amplifier. Presently only solid state amplifiers are
used in the FMLs after their availability in the high frequency band. As is brought
             CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                       XV - 9
out the important characteristics of the loop amplifier are its small signal gain
suppression, less gain and fine grain variation across the band and low noise
figure.


          Delay Line : The selection of delay line involves compromises on many
conflicting requirements. Longer the delay line, better is the frequency accuracy
as maximum error in worst conditions is the reciprocal of the delay time, which is
prepositional to the length of the delay line. Hence the length of the delay line
depends on the requirement of the minimum instruction pulse width.


          The choice is between wave guide, co-axial and acoustic delay lines.
Which again depends on the specific application. Wave guide delay lines are
more suitable when long memory times are required. They will impose less gain
requirements on the loop amplifier because of their less insertion loss. But they
are bulky and dispersive in nature.


          Acoustic Delay Line : These are small size and light weight. Their
attenuation characteristics are less dependent on temperature. But their insertion
loss is very high and they are basically of smaller band width.


          Co-axial Delay Lines : These delay lines compromise between the two as
long as relative insertion losses, size and weighs are concerned. But the insertion
loss of      co-axial line directly varies with temperature. Typical co-axial delay
line of 100 nano seconds will have 16 dB at –54° C, 23 dB at 25°C and 29 dB at
95°C. While designing the system these aspects are                 to be taken in to
consideration.


          Equalizers : If small signal gain suppression and loop noise level were
constant with respect to frequency it is desirable to have a constant open loop gain
through out the band. But in practice neither of the above are constant. In addition
the gain characteristics of the loop amplifier and the attenuation characteristics of
               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                        XV - 10
the delay line are independent and may not have any correlation. In such cases to
attain an open loop gain of about 15 dB through out the frequency band can be
achieved only by inserting the equalizers between the delay line and the loop
amplifier.


       Gain equalizers are passive microwave components that have an insertion
loss characteristics that varies as a function of frequency. They are commercially
available with a variety of loss shapes, linier positive slope, linier negative slope,
half sine,        etc. Gain equalizers can be made according to the customer
specifications with multiple controls. Twelve to sixteen point control gain
equalizers are more use full for this application.


5.0     Application:


       Most deception repeater jammer provide programmed delay in re-
transmission of the threat signal to achieve range deception. Range tracking can
be broken by re-transmitting sufficiently strong false signal (at the frequency of
the threat signal) to activate the radar’s Automatic Gain Control (AGC) circuitry
and then progressively delaying the transmission of the false signal relative to
reception of the threat signal as shown in Fig. 1. To achieve range deception, it is
necessary to reproduce the threat signal frequency after the pulse is no longer
present and for a duration greater than the maximum programmed delay.


6.0   Advantages:


       Advantages of the FML device over the other ECM devices, such as the
frequency synthesizer techniques are:
             •    Nearly instantaneous response capability utilizing the first pulse of
                  the received signal.
             •    Ability to handle a new input frequency immediately after storage
                  of the first input signal at a different frequency.

                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                          XV - 11
            •    Wide bandwidth, fast performance and field reliability at low cost.
            •    High degree of spectral purity.


7.0 Conclusion:


       Channelizing the operating band into sub bands improves the S/N ratio of
the loop. Small signal gain suppression of the loop amplifier enhances the
memory time. The open loop gain of the total loop can be made uniform to the
maximum possible extent using Fine Grain Gain Equalizers. Individual units
require tuning of the gain equalizers to suit their open loop gain characteristics.
The inherent limitation of FML as short time storage device is overcome in the
present day development of DRFM which can store the signal digitally for infinite
duration.


8.0   References :

       1.       Harry F. Egbert, “Octave Bandwidth, Acoustic M/W Frequency
                Memory Loop”, Microwave Journal, Sept. 1973
       2.       Kellet J.D. , “ Some Recent Findings in Microwave Storage”, IRE
                Transactions on Microwave Theory and Techniques, July 1961.
       3.       Richard J. Wiegand, “Radar Electronic Countermeasure System
                Design”, Artech House, Boston, 1991
       4.       Richard N. De Gunther and Stephen P. Maccabe, “Frequency
                Memory Loop”, Vol-10, No.-2 , March/April 1983, WJ Tech-notes.




                CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         XV - 12
                                                                                Radar Pulse In



                                                                                  Stored RF Signal
       FML                       LNA
                                                                                 Jammer Tx. Pulse
                                                                       Rx.
                                                                   Antenna



HIGH         POWER
AMPLIFIER

                                                     Jammer Tx. Antenna



RGPO TECHNIQUE




                 Fig. 1 : Repeater Jammer
                                                                                        4 µS

                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                          XV - 13
                                                                                 RF OUTPUT
             RF SWITCH
                                                                                 PORT
                                   AMPLIFIER

INPUT PORT




                                   DELAY LINE


                             Fig. 2 : Frequency Memory Loop




                  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                           XV - 14
 INPUT RF PULSE




SWITCH COMMAND




 RECIRCULATED
   RF OUTPUT




                                                              VARIABLE DELAY

   OUTPUT PULSE




                   Fig. 3 Timing Waveforms




          CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                   XV - 15
Output
Power                                                         Compression
(dBm)
                                                                Saturation
                                                                                Overdrive

                         Signal # 1               Constant Gain
                                                     Slopes

                Signal # 2
                                                                  Compression


                                                                  Suppression




            Small Signal Region         Large Signal Region

                                  Input Power of Signal # 1 (dBm)

         Fig. 4 : Small Signal Suppression Phenomenon in SSA




              CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                       XV - 16
                     Loop Amplifier Gain                                            Stable Operating
                                                                                          Point
Power Output




                                                                     Delay Line
                                                                   Insertion Loss




               Pin
                                               Power Input

                            Fig. 5 : Signal Buildup in the FML

                     CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                              XV - 17
CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                         XV - 18
                                    AMPLIFIER                           RF OUTPUT

       RF SWITCH                                                        PORT


INPUT PORT

         180° Phase
           shifter



                                  DELAY LINE

                       Fig. 7 : FML with 180° Phase




         CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                  XV - 19
           SPST
RF Input                                                                                                  RF Output

               DC     A1           PAD          A2               ISO            DC          DC



                                                                      QUADRUPLEXER

                                                                                              Detectors     Controls &
                                                                                                            Tell backs
                                                                  ACTIVITY DETECTION
                                                                  & CONTROL CIRCUIT


                                                                                                              Delay
                                                                                                              Line

                    FGGE                 BPF            ISO
           S                                                         S
                    FGGE                 BPF            ISO
           P                                                         P                    BI PHASE
           4                                                         4          ISO
                                                                                            MOD.
                    FGGE                 BPF
           T                                            ISO          T
                    FGGE                 BPF            ISO


                               Fig. 8 : Schematic diagram of FML
                           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                                    XV - 20
            Chapter XVI




RFP Technique for unique Identification
         of emitter system
                          CHAPTER - XVI




     FINGER PRINTING TECHNIQUES FOR UNIQUE
           IDENTIFICATION OF EMITTERS




                             CONTENTS




1.     INTRODUCTION

2.     RADAR FINGER PRINTING SYSTEM

3.     ISSUES FOR A FINGER PRINTING SYSTEM

4.     EXPERIMENTAL SET–UP FOR FINGER PRINTING

5.     RESULTS OF THE EXPERIMENT

6.     CONCLUSIONS

7.     REFERENCES




        CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                   XVI
   FINGER PRINTING TECHNIQUES FOR UNIQUE
         IDENTIFICATION OF EMITTERS

      Shri R C Agarwal, Sc ‘G’, Shri KSVM Shyam Kumar, Sc ‘D’

                             DLRL, Hyderabad

1.0     Introduction :
        The role of an EW receiver is to intercept and classify various emitters in a
dense hostile environment. The conventional EW receivers[10],[11] measure a
number of parameters of the emitter, e.g. DOA, frequency, pulse amplitude, pulse
width, time of arrival and generate PDW’s (Pulse Descriptive Word) for each
pulse detected in the scenario. All the PDW’s formed are sorted/de-interleaved,
grouped together and tracks are formed. Each of the tracks formed is assigned a
Track Data (TD) which basically consists of the various ESM parameters
measured of the emitter. This TD is then compared with the library and the
identification of the emitter is declared. But the identification process through the
above conventional means is limited to the mode of radar. Such a limitation arises
due to the number of parameters measured as well as accuracy of parameters
measured. Other problems with the above conventional process include detection
of exotic intra-pulse modulation schemes on the radar waveform as well as
problems with Pulse-on-Pulse or simultaneous signal conditions. In order to
identify a unique serial number of the emitter even among the emitters of the
same type, demands Finger Printing of the radars.


        The text is organized as follows. Section-2 explains a Radar Finger
Printing System. It also explains various features on which a Finger Printing
System is based. Section-3 discusses various issues in the design of Finger
Printing system. Further various Signal Processing algorithms that can be utilized
in a Finger Printing system are discussed. In section-4, an experimental
configuration set-up for carrying out Finger Printing is described. The various
simulations carried out with this experimental set-up on a Commercial ATC radar


          CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                   XVI - 1
as well as on simulated data are described. Section-5 discusses the results of the
simulation studies carried out. Finally in section-6, conclusions are discussed.


2.0      Radar Finger Printing System [2][8] :


         In order to extend the capabilities of identification to the extent of
recognizing the serial number of an emitter even among the emitters of the same
type requires a Radar Finger Printing System. Finger Printing which refers to
unique identification of emitters among the same mode of radars is basically
based on Inter pulse analysis of radar waveforms and measures the frequency,
phase and amplitude variations within the pulse of that radar. These variations
may be intentional and/or unintentional. The intentional modulations on pulse
(IMOP) are well known, the simplest being linear chirp and barker code phase
modulation. The unintentional modulations on pulse (UMOP) are due to inherent
characteristics of all high-powered radar transmitters. The amount and type of
modulation varies with the transmitter type. These modulations are present in the
output of high power transmitting tubes and are due to pushing, pulling and other
effects such as temperature, aging and poor maintenance. The modulations-IMOP
and/or UMOP generate minute variations in the signal characteristics of every
emitting system thereby creating a unique Signature for the emitter.


         To capture the minute variations, which may be due to intentional and / or
unintentional modulation, an intra pulse analysis is needed. Such an intra pulse
analysis, extracts as many parameters (features) of radar pulses as possible with
fine grain accuracy. With these extracted features, the unique emitter
identification of the emitter can be carried out.


         In essence, a Radar Finger Printing System is basically a digital receiver
which will


-     Provide unique identification of emitter among a class of emitters,

           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                    XVI - 2
-     Track a specific target/identification of platform,
-     Identify new emitters,
-     Generate Electronic Order of Battle (EOB).


3.0     Issues for A Finger Printing System :


         The design of a Radar Finger Printing System involves digitizing the
signal and processing the digitized data for extraction of features of the emitter.
The various issues that determine the approach of a Radar Finger Printing System
are –


      (1) IF frequency & BW
      (2) Sampling rate
      (3) SNR requirement
      (4) Instantaneous parameter measurement
      (5) Signal Processing Algorithms
      (6) System architecture for finger printing


3.1      IF Frequency & Bandwidth :


         The selection of instantaneous bandwidth of the system depends on the
type of signal, SNR and full-power bandwidth of the A/D. The type of signal
determines the bandwidth to be provided in the system. For example pulsed
signals require more bandwidth in comparison to CW and frequency agile signals
require more bandwidth than pulsed signals. So it is advantageous to open the
receiver for a bandwidth depending on the type of signals. This approach results
in two significant advantages-first, it generates a better SNR, and second, it filters
the unwanted signals. With these considerations, the appropriate bandwidth
required for the scenario is to be selected.




           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                    XVI - 3
       The spurious components generated with the selection of a particular
combination of IF & bandwidth, availability of standard RF components and
finally A/D requirements determine the selection of IF.


3.2    Sampling Rate :


       The important criteria which determines the sampling frequency is the
problem of aliasing. The sampling rate requirement can be reduced, if instead of
conventional low pass sampling, the technique of under-sampling or band-pass
sampling is employed. But even with such a technique, the A/D should support
the full-power-bandwidth demanded at the input of the A/D. Further more, when
the criteria of providing equal guard band for the lowest and the highest frequency
components is used, the options available are further reduced. The other important
consideration is the number of effective bits available for A/D at the chosen
sampling frequency. The number of bits of the A/D influences the accuracy with
which various parameters can be measured. Also, considerations are to be taken
on the requirements of roll-off rate of the anti-aliasing filter required for the IF,
sampling frequency and bandwidth chosen.


3.3    Instantaneous Parameter Measurement :


       The essential emphasis of RFPS is to perform intra pulse analysis. Such an
intra pulse analysis consisting of instantaneous parameter measurement reveals
the signature of the emitter. The frequency profile of a pulse is a vital parameter
to detect intentional frequency variations of a chirp radar as well as any
unintentional frequency modulations particularly during rise time and fall time.
The phase profile is an important parameter that identifies the important class of
Barker-coded Radars. Similarly the instantaneous amplitude profile will reveal
any unintentional/intentional amplitude modulations. The ensemble averaged
amplitude and frequency profiles of the pulses of the emitter is a very good source
for identifying the emitter. Hence to cater to the above needs, it is required to

         CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                  XVI - 4
measure instantaneous amplitude, phase and frequency on a sample to sample
basis.


3.4      SNR Requirement :


         Proper de-interleaving and identification of emitter through Electronic
Finger Printing demands a high signal-to-noise ratio[10]. Bringing pulse shape
out of many pulses corrupted by noise generally requires averaging many pulses.
Pulse shape averaging involves taking many individual wave shapes, aligning
them in time and frequency and then averaging them point by point across the
waveform ensemble.        The alignment is critical. Thus the averaging must be
conducted on pulses arising from a single source i.e. pulse stream that must have
already been de-interleaved. The higher the SNR, the fewer pulses are needed for
meaningful averaging. The requirement of SNR to electronic finger printing can
be critical when applied to scanning beams where the opportunity to gather many
pulses is not there. For this reason, only those pulses that have sufficient SNR are
to be processed.


3.5      Signal Processing Algorithms :


         One of the important areas in a Finger Printing System is the feature
extraction of the emitter. The various Signal Processing techniques that can be
utilized in the feature extraction are –


      (1) Frequency measurement through zero-crossings
      (2) Instantaneous phase, frequency, amplitude measurement using I-Q
         decomposition.
      (3) FFT method
      (4) STFT method
      (5) Wavelets


           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                    XVI - 5
3.5.1   Frequency Measurement Through zero crossings [12] :


        One of the simplest methods of finding the instantaneous frequency is
through zero crossings of the signal, wherein, the time difference between the
points of the signal crossing the zero can be measured and through the time
difference, frequency can be measured. This process can be viewed as a
comparator in which a sine wave is converted to a square wave and the time
difference between the changes of sign of the square wave is measured.


        If x(n), x(n+1),x(n+2) are three consecutive samples of the signal, the
frequency of the signal can be found from
               2πfts = cos-1 { [x(n+2) + x(n)]/2x(n+1) },      - Eq(1)


        where,
        ‘f ‘ is the frequency of the signal,
        ‘ts ‘ is sampling period of the signal


        A further extension of the above approach is to use interpolation schemes
to exactly identify the zero crossings. Other scheme is to find timing over a larger
samples instead of the time-difference between two consecutive zero crossings.
This approach brings in some integration generating better accuracy for frequency
non-varying signals.


        Although the zero crossing time calculated from Eq(1) is based on exact
solution, the calculation itself is rather tedious. Further there may be some ill-
conditioned cases if x(n+1) is very small. The error in the frequency measurement
will be small only under very high SNR conditions. This process of measurement
demands a very high sampling rate to exactly detect the zero crossing points.
Also, the underlying assumption is that there is only one frequency component
present and the method fails when multiple signals exist. Although the method


          CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                   XVI - 6
looks simple, it is prohibitive practically because of the demands on the sampling
frequency.


3.5.2   I-Q Method :


        A more universal solution to the problem of feature extraction for Finger
Printing is the decomposition of the signal into I & Q components. From the I &
Q components, the profiles of amplitude, phase and frequency can be measured


                                                          LP F       A /D       I



                                            LO
                    IF
                  signal
                                            90 0
                                       phase shifter


                                                          LP F       A /D       Q

                      Fig-1 M ethod(a) of realizing I-Q decom position




                                                           LPF        I


                                            COS
                                            G EN.
      IF              A /D
    signal
                                             SIN
                                            G EN.


                                                           LPF        Q
                     Fig-2 M ethod(b) of realizing I-Q decom position




                                                                            I

      IF              A /D
    signal
                                                    H ILBER T
                                                 TR AN SF O R M ER          Q

                      Fig-3 M ethod(c) of realizing I-Q decom position



             CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                      XVI - 7
using
        A(t) = sqrt[ i2(t) + q2(t)]                     -       Eq(2)


        Φ(t) = tan-1 [q(t)/i(t)]                        -       Eq(3)


        f(t) = Φu(t2) - Φu(t1) / (t2 - t1 )             -       Eq(4)




        Where,
        i(t) is the in-phase component of the signal,
        q(t) is the quadrature component of the signal,
        A(t) is the instantaneous amplitude profile of the signal,
        Φ(t) is the instantaneous phase of the signal,
        Φu(t) is the unwrapped phase from Φ(t),
        f(t) is the instantaneous frequency of the signal


        The phase unwrapping can be a simple processing of the principal value of
the phase and detecting any abrupt phase changes. A more complicated, but
accurate procedure can be to use an adaptive numerical integration scheme[6] that
combines the information contained in both the phase derivative and the principal
value of phase.


        The conversion of signal into in-phase and quadrature components[7],[12]
can be accomplished in
(a) An analog conversion followed by digitization of the in-phase & quadrature
   components
(b) Digitizing the signal and then converting into I-Q components in digital
   domain
(c) Digitizing the signal and then using an Hilbert transformer to generate the
   quadrature component


          CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                   XVI - 8
         All the three schemes are shown from Fig-1 to Fig-3. Among the three
methods, method(b) is the best in term of simplicity of implementation. Although
for method(a), A/D sampling frequency is half as that of method(b), so that A/D’s
at lower sampling frequency can be used, it suffers from two disadvantages. One,
all the components in both I & Q path including A/D’s and filters should match
over the complete band of interest and both A/D clocks should be synchronized.
Second, this method requires two A/D’s.


         The important derivative of instantaneous parameter measurement using I-
Q decomposition is the scope for using intermittent as well as the final results of
the process for detection of intentional exotic modulation schemes. The overall
disadvantage of the I-Q method is that the method fails under simultaneous signal
conditions as well as the procedure demands slightly higher SNR.


3.5.3    FFT Method :


         Besides looking for the characteristics and features of an emitter in the
time domain, an important domain for the feature set is the frequency domain.
Given a discrete time-domain signal x(n), the frequency domain can be obtained
using the Fourier transform by
            ∞


X(ω) =     Σ x(n)e-jωn                   -       Eq(5)
          n = -∞




         The overall procedure can be visualized as if there are a group of band
pass filters. The availability of efficient Fast Fourier Transforms (FFT)
particularly make this method attractive for implementation. The advantage of the
FFT method is the possibility of better detection and estimation of features than
the time domain approaches because of the underlying processing gain. The
presence of simultaneous signals can also be detected by the use of this method.



           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                    XVI - 9
3.5.4 STFT Method [3][9] :


        The disadvantage with the FFT methods is the complete lack of timing
information. One way of extracting the timing information is to move a window
function on the time domain signal and compute the Fourier transform for the
windowed signal. Such an approach called as Short Time Fourier Transform
(STFT), localizes in time domain, the frequency content of the signal. The STFT
output is obtained from
             ∞

X(n,ω) =     Σ x(m)w(n-m)e-jωn                           -       Eq(6)
           n = -∞




        Where, w(m) is the window function. Among all the window functions,
the use of Gaussian window is of special interest which provides the optimum
time-frequency localization, in which case an STFT becomes a Gabor transform.


3.5.5   Wavelets [9] :


        As mentioned above, it is not only important to detect the frequency
content of the signal, but also its behavior in time. A wavelet transform is a linear
time frequency transformation that maps the one-dimensional signal of time into a
two-dimensional function of time and frequency. This method is an alternative to
STFT. But the advantage with wavelets is that a wavelet transform uses short
windows at high frequency and long windows at lower frequency, generating the
so called “constant-Q” analysis.


For a signal, x(t), the wavelet transform is given by


CWT(τ,a) =          1_ ƒ x(t) h*(t-τ/a)dt        -       Eq(7)
                 √|a|


Where, h(t) is the mother wavelet.
           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                   XVI - 10
             The wavelet analysis basically results in a set of wavelet coefficients that
      indicate how close the signal is to a particular basis function. The major
      advantage with wavelets which provide multi-resolution analysis for feature
      extraction, is the possibility of zooming to time discontinuities and providing
      localization in time and frequency using ortho-normal basis. Other advantages
      include a better picture for multiple signal conditions because of the
      discrimination capabilities in time and frequency simultaneously.


3.6        System Architecture for Finger Printing [1][11] :


             The role of a Finger Printing system is to perform fine grain inter-pulse
      analysis of the emitter signals. The demands of such an analysis can be better
      satisfied in a digital domain than in analog domain. Realization of such a structure
      requires digitizing the analog signal with the A/D and then processing the
      digitized data using Signal Processing techniques. Using such a process, Finger
      Printing is to be carried out in a typical frequency band of 0.5-40 GHz. Because
      A/D cannot digitize this large frequency band of interest, it is required to down
      convert to a comfortable IF signal from 0.5-40GHz before being fed to the A/D.


             Fig-4 shows one of the architectures of realizing a Finger Printing system.
      The architecture is a scanning type of receiver in which the receiver is tuned
      through the local oscillator to the frequency band of interest. Important
      considerations are the instantaneous bandwidth of the system which is to decided
      by the typical bandwidths of the radars as well as the bandwidth required for
      Finger Printing. For tuning to a particular frequency, the required band of interest
      is to be known apriori, which can be found out through other systems or by this
      system itself by scanning the various frequency bands and detecting the presence
      of emitters.




                CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                        XVI - 11
                       DEMOD




                                     A/D              DSP
                                                   HARDWARE


               LO

     Fig-4 Scanning type of architecture for Finger Printing




                                    A/D      DSP

                                                            POST
                              A/D         DSP            PROCESSOR




                        A/D         DSP



      Fig-5 Channelized type of architecture for Finger Printing




                                    A/D     MEMORY


                                                                   DSP
                              A/D          MEMORY




                        A/D          MEMORY




      Fig-6 Hybrid type of architecture for Finger Printing


CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                        XVI - 12
       Fig-5 shows the channelized type of architecture in which the entire
frequency band is channelized into various bands which can be comfortably
handled by the A/D’s. Each of the chains in the architecture has an A/D and a
processing element. This type of architecture provides the capability of
simultaneously Finger Printing many emitters. But the overall architecture is
highly hardware intensive.


       A hybrid type of architecture and a trade-off between the above two
architectures is shown in Fig-6. The entire frequency band is channelized and
digitized using multiple A/D’s. All the digitized data is stored in memories and
only a single channel is processed at a time. This sort of architecture, at the cost of
some extra hardware provides more bandwidth, even though not instantaneous,
than possible through scanning type of architecture.


       In any architecture, one of the key elements is the A/D which ultimately
decides the architecture and the overall scheme for Finger Printing. For the
realization of the A/D, instead of using a single A/D at high sampling frequency,
an alternative approach is to use a bank of A/D’s at a low sampling frequency, all
running from a multi-phase clock. Although, this realization reduces the sampling
clock requirement and the associated working speed of the components that
follow the A/D, it still demands that the input full-power bandwidth of the A/D to
be at least the highest frequency component present. One more extension and
alternative is to use some further channelization with a bank of A/D’s in which
the input to each of the A/D chains is further down-converted to a lower
frequency which may solve the full-power-bandwidth problem of A/D.


       The other issues in the implementation of extracting features for Finger
Printing is the ability for the system to process data only during the pulse on time.
Since Finger Printing is essentially intra-pulse analysis, processing the off time of
the pulse train is not required. This sort of strategy of capturing only the pulse on
time data reduces the overall memory requirements as well as brings down the

          CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                  XVI - 13
overall reaction time of the system. For implementing this approach, one
alternative is to generate Real Time Gating signal for the emitter of interest. This
can be implemented by knowing the pulse width, PRF and then giving proper
margin for the gating signal width depending on the type of PRF. Another
attractive and reliable scheme is to use a circular buffer type of writing into the
memory effectively creating a sort of post-trigger. In this scheme, the digitized
data can be written continuously and repeatedly until the occurrence of a pulse.
Once the pulse has occurred, writing into the memory is stopped and the stored
data which contains the pulse is used for processing. With the use of two such
circular buffers in a double buffer mode, effective storing of only the pulse data
can be implemented.


4.0     Experimental Set –up for Finger Printing :


       The experimental set-up used for carrying out studies for feature extraction
of a commercial ATC radar is shown in Fig-7. The signal from ATC radar was
captured using a horn antenna. By using the apriori information of the signal
frequency, the RF was down converted to 160 MHz IF with 40 MHz bandwidth.
This IF was digitized using an 8 bit A/D at 500 MSPS. After the A/D, a memory
buffer of 4MB was provided. Signal was captured and stored in real time. The
process of digitization was continued till a memory buffer of 4MB was filled.
Sufficient gain was provided in front of the A/D to match the signal level to that of
the A/D. The captured data is then passed on to MATLAB for analysis. Only the
pulse on time data was processed.


       The input signal was converted into I-Q components using the I-Q
decomposition shown in Fig-2. From the I & Q components, instantaneous
amplitude, phase and frequency were measured using the methods described in
section 3.5.2. The instantaneous amplitude envelope measured is then filtered
using Median filter. It was observed that the median filter, which is a non-linear
digital filter, was very effective in removing the impulsive noise over the

          CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                  XVI - 14
instantaneous amplitude profile. On the filtered amplitude profile, the selected
threshold was applied and the exact location of the pulse was detected. To remove
the problems of noise chatter, during the rise time and fall time of the pulse,
hysterisis technique[10] was used. To avoid detection of noise, the logic of
minimum pulse width was used, where only those pulses that have sufficient
minimum


         RF front end                       8 bit,
                                          500 MSPS        4 MB
                                    160 MHz
                                                A/D       memory         Matlab
                               I\    IF


                 LO


    Fig-7 Experimental set-up for feature extraction of ATC radar


        pulse width were declared as pulses. This effectively removed all the noisy
spikes that were above the threshold level.


       Using the filtered amplitude profile, the reference amplitude during the
stable time of the pulse was estimated. From the estimated reference amplitude,
the 3dB pulse width as well as the rise time/fall time of the pulse were measured.
Because, the measurement of parameters is to be done on a single pulse, as such
there is no integration of pulses available for parameter measurement. Hence each
pulse should support sufficient SNR for measurement. The SNR of each pulse was
estimated and only those pulses having sufficient SNR were considered for
analysis and others were dropped out.


    The qualified pulses were then processed for phase measurement. The phase
was then unwrapped and the instantaneous frequency profile was measured. For
achieving better frequency accuracy, instead of considering the phase difference
between successive samples, frequency from phase differences between every 8
          CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                  XVI - 15
samples was measured. Using the method described in Section 3.5.3, the spectrum
of only for the pulse duration was measured. To perform time-frequency analysis,
STFT of Section 3.5.4 was used. A statistical estimation of all parameters was
done till second order to identify the statistical behavior of the results.


5.0     Results of The Experiment :


        The digitized signal that is fed for analysis is shown in Fig-8. The effect of
scanning on the digitized IF data can be seen. The plot does not show the complete
off-time of the pulse train. The data used for analysis is the pulse on time with
some buffer before the rising edge and after the falling edge. Fig-9 depicts the
instantaneous amplitude profile measured from the raw data. The filtered
amplitude profile using the median filter is shown in Fig-10. The result of using
the median filter in effectively removing the noise can be clearly seen in the
figure. The instantaneous frequency measurement done only during the pulse on
time is shown in Fig-12. For comparison purposes, Fig-11 shows the instantaneous
amplitude above instantaneous frequency. The first pulse is not having sufficient
SNR and is therefore dropped and not used for further processing. Hence for the
first pulse, there is no frequency measurement in Fig-12. The zoom plot of
instantaneous amplitude and frequency for one of the pulses is shown in Fig-13
and Fig-14 respectively. The results of spectral analysis using FFT are shown in
Fig-15. This plot shows a superimposed plot of spectrum of each of the pulses.


        Besides, a complete time domain and a complete frequency domain
analysis, a joint time-frequency analysis is done using STFT. The results from this
approach are shown in Fig-16. The x-axis in the plot is time scale. Y-axis shows
the frequency. The other dimension, the signal strength of the signal is colour
coded. The darker regions in the plot indicate the presence of signal. The result of
using STFT on signal from the signal generator is shown in Fig-17. The input
signal in this case was a 50% duty cycle signal that was frequency modulated with


          CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                  XVI - 16
a sinusoidal base band signal. The dark regions along the time axis, which are the
strongest frequency components, show the base band sinusoidal signal.


       All the above analysis was repeated on the ATC radar a number of times,
during different parts of the day and the end results of the study validated the
signal processing contemplated for the purpose. It was observed from the analysis
that there was a consistency in all the measurements including rise and fall times.
It was also noted that the standard deviation of the frequency was more during the
rise and fall time than the stable time of the pulse.


        All the above analysis was conducted on synthesized data as well as data
from digitized signals from different signal generators. The various results
obtained by the use of various Signal Processing algorithms, clearly reflected the
signal conditions demonstrating the success of the above algorithms in the feature
extraction.


6.0     Conclusions :


       In this text, we have discussed the concept of Finger Printing of emitters.
Various issues in the realization of a Finger Printing system are described. Various
Signal Processing algorithms that can be utilized for extraction of features of the
emitters are discussed. The architectural alternatives for evolving a Finger Printing
system are explained. An experimental set-up which was utilized for extracting the
features on a Commercial ATC radar was described. The Signal Processing
algorithms that were validated in the experimental analysis were discussed.
Finally, the various results obtained from the analysis were presented. The Finger
Printing system described in this text with minor modifications can also perform
the role of conventional ESM systems making it a more complete and universal
solution for the EW needs.




          CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                  XVI - 17
         Besides, using the Signal Processing techniques discussed in the text, other
promising areas are the use of cyclo-stationarity[5], Wigner-Ville, Choi-Williams
type of time-frequency transformations[3]. The other key area in Finger Printing is
the emitter identification algorithms. The use of pattern-recognition algorithms[4]
is one of the specific possibilities. The primary goal of pattern recognition is
supervised or unsupervised classification. Supervised classification (e.g.
discriminant analysis) is the identification of the input pattern as a member of a
pre-defined class whereas the unsupervised classification (e.g. clustering) is the
assignment of the input pattern to a hitherto unknown class. Some of the best
approaches for pattern recognition are template matching, statistical classification,
syntactic or structural matching and neural networks. Other techniques of
identification include the use of graphical matching techniques on the ensemble
averaged profiles of instantaneous parameters.


7.0       References :


      1. Dietmar Matthes, Digital receiver – A technology for future electronic
         warfare system.
      2. Don Herskovitz, The other SIGINT/ELINT, Microwave Journal, Sep 1996,
         pp 22-34
      3. F.Hlawatsch and G.F.Boudreaux Bartels, Linear and Quadratic time-
         frequency signal representations, IEEE SP Magazine, April 1992, pp 21-
         67
      4. A.K.Jain, R.P.W.Duin and J.Mao, Statistical             pattern recognition : A
         review, IEEE Transactions on Pattern Analysis and Machine Intelligence,
         January 2000, pp 4-37.
      5. James P. Stephens, Advances in signal processing technology for
         electronic warfare, IEEE AES Systems Magazine, Nov 1996, pp 31-38.
      6. Jose M. Tribolet, A new phase unwrapping algorithm, IEEE Trans. ASSP
         April 1977, pp 170-175


              CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                      XVI - 18
7. B.Liu et.al., A new quadrature sampling approach, IEEE Trans. ASSP Sep
    1989, pp 733- 747.
8. P.Michael Gale, Intrapulse and ELINT system design, J.Electronic
    Defense, Oct 1999, pp 86-106.
9. Oliver Rioul and Martin Vetterli, Wavelets and signal processing, IEEE SP
    Magazine,    October 1991, pp 14-38
10. Richard G. Wiley, Electronic Intelligence: The analysis of radar signals,
   Second edition, Artech house, 1993.
11. B.D.Trimmer & V.Barker, Signal detection: Future receiver technologies,
   Conf. IEE 1994,U.K., pp 3/1-3/7.
12. J.B.Y.Tsui, Digital techniques for wide band receivers, Artech house, 1995.




      CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                              XVI - 19
   0 .2



 0 .1 5



   0 .1



 0 .0 5



      0



-0 . 0 5



  -0 . 1



-0 . 1 5
           0         0 .5             1               1 .5             2         2 .5
                                                                                     4
                                                                              x 10




                 Fig.8 Digitzed IF from A/D vs Sample No.



               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                       XVI - 20
 -40


 -60


 -80


-100


-120
       0        5               10                 15                20            25   30
              Fig-9 Inst. Amplitude (dBm) vs Time(usec)

 -30


 -40


 -50


 -60


 -70
       0        5               10                 15                20            25   30

           Fig-10 Inst. Amplitude (dBm) vs Time(usec) after Median filtering




                    CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                            XVI - 21
 -40


 -60


 -80


-100


-120
    0   5               10                   15                   20          25   30
            Fig-11 Inst. Amplitude (dBm) vs Time(usec)


200


150


100


 50


  0
   0    5             10               15
             Fig-12 Inst. Frequency (MHz) vs Time(usec) 20                    25   30
               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                       XVI - 22
 0


-20


-40


-60


-80
      0   0.2        0.4              0.6             0.8               1       1.2   1.4
                Fig-13 Inst. Amplitude profile (in dBm) of one of the pulses

200


150


100


50


 0
      0   0.2        0.4              0.6             0.8               1       1.2   1.4


                Fig-14 Inst. Frequency( in MHz) profile of one of the pulses




                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         XVI - 23
 -4 0

 -6 0

 -8 0

-1 0 0

-1 2 0

-1 4 0
     1.2   1.4          1.6              1.8               2               2.2         2.4      2.6
                                                                                                    8
                                                                                             x 10
             Fig-15 Superimposed plot of Spectrum of all pulses




                        CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                                XVI - 24
                         8
                  x 10
            2.5




             2




            1.5
Frequency




             1




            0.5




             0
                   0         0.5                1                   1.5                    2      2.5
                                                            Time                                               -5
                                                                                                        x 10

                             Fig-16 Spectrogram output through STFT on ATC data
                                   CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                                           XVI - 25
                         8
                  x 10
            2.5




             2




            1.5
Frequency




             1




            0.5




             0
                   0         0.5                 1                    1.5                    2    2.5
                                                              Time                                             -5
                                                                                                        x 10

                             Fig-16 Spectrogram output through STFT on ATC data

                                   CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                                           XVI - 26
        CHAPTER – XVII


TRAVELLING WAVE TUBE (TWT)
      FUNDAMENTALS
 Shri. PRATULANANDA PAL, SC 'F'




      CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                 XVII
                     CHAPTER - XVII



           TWT - FUNDAMENTALS OF



                          CONTENTS




1.   INTRODUCTION

2.   CONSTRUCTION

3.   PRINCIPLE OF OPERATION

4.   CHARACTERISTICS

5.   CONCLUSION




      CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                 XVII
              Travelling Wave Tube (TWT) – Fundamentals
                              Shri Pratulananda Pal, Sc 'F'
                                   DLRL, Hyderabad


1.0   Introduction

                A traveling wave tube is a unique type of electron tube capable of high
      amplification of varying in frequency over multi octave without the need for any
      tuning or voltage adjustment. The principle of the traveling wave tube was first
      conceived by Rudolph Kompfner in 1942. Dr. John Pierce Of Bell Telephone
      Lab later recognized and solved the technical limitations faced by Kompfner to
      achieve the practical device.


                As the name suggests, the wave travels along the axis of the tube. The
      construction of the tube is such that the beam formation, interaction and collection
      are made in three separate sections of the tube so that they can be optimized for
      their performance independently. As there is no cavity, the tuning element, it is
      broad band in nature.


2.0   Construction


      TWT essentially consists of three distinct components. They are


      (i)       The electron gun which emits electrons
      (ii)      The slow wave structure i.e. the interaction space
      (iii)     The colletor where the electrons are collected at the end of interaction.


      The schematic of the TWT is shown in Fig.1.




                  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                          XVII - 1
2.1   The Electron Gun


              It consists of a cathode, heater and control grid. Some TWTs contain
      focus electrodes and grids. Generally cathode forms the convergent beam with
      cathode area to be 100 times the area of beam interaction space. Even with this
      high convergence the available perveance is quite small. The perveance is an
      important parameter of the gun and is denoted by P where I=PV 3/2 where I is the
      beam current and V is the accelerating voltage. The value of P of TWT various
      from 0.2 to 2up. The limitation in perveance results from allowable electrode
      spacing without voltage breakdown and allowable cathode current density of long
      life operation. Impregnated cathodes have 50,000 hour life in a well designed
      tube.

2.2   The Slow Wave Structure


              There are different types of slow wave structures used in TWTs. The best
      structure in terms of bandwidth is helix. But the power handling capability of
      helix is limited due to RF losses and interception of some electrons from the
      beam. For high power application the helix slow wave structure is replaced by
      ring bar, ring loop and by the coupled cavity for still higher power but of the cost
      of bandwidth.


      The functions of the slow wave structure are two viz.


              It provides space for interaction between electron beam and RF wave.


              It slows down the RF wave for interaction with electron beam.


              The electron beam velocity is much less compared to RF wave velocity
      which is equal to the velocity of light i.e. 3x10**8 metre/Sec. But for proper
      interaction these two velocities should be approximately equal. The slow wave

               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                       XVII - 2
       structure is used to reduce the velocity of RF wave. While the RF wave
       propagates at the velicity of light along the circumference of the helix, the axial
       velocity of the wave is reduced by a factor which is equal to pitch by
       circumference of the helix. It can also be considered that the slow wave structure
       modify the space charge wave propagation constant in such a way it becomes
       complex and represents a growing wave. In fact the helix products four waves
       viz:-


       (a)     Forward wave with velocity slightly higher than beam velocity. This wave
               gives energy to the electron beam and will die down shortly.


       (b)     The other forward waves with velocities slightly lower than beam velocity
               but one with growing in nature and other decaying in character.          The
               growing wave only will sustain till end and will be amplified by taking
               energy from the beam whereas decaying wave will die down.


       (c)     One backward wave will propogate from output side to input side and may
               be refelected by the input mismatch. It may creat oscillations in the tube. A
               small attenuator is placed at the input end of the tube to avoid the
               oscillations.

       A single characterstic which makes TWT a supreme among all microwave tube in
the field of ECM is the band width and the slow wave structure is responsible for it.


The phase velocity Vph = wavelength * frequency


                           = (2 π/β) x freq.


                               = ω/β

       Where β is the phase constant i.e. phase shift per unit length and ω is the radial
velocity.
                   CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                           XVII - 3
       This shows that the phase velocity is frequency dependent.                It is easy to
synchronize the RF phase velocity with electron beam velocity at a particular frequency
but it is quite difficult to synchronize over a wide frequency band for a fixed beam
voltage. The rate at which RF velocity changes with frequency determines the required
variation of the operating beam voltage to maintain the interaction between the beam and
the RF wave. The w/β diagram for the slowwave structure has major influence on the
tube’s band width.


       Fig.2 shows the ω/β diagram for the helix slowwave structure for the forward

wave. The slope of the curve (ω/β) at any point on the curve gives the phase velocity at
that particular frequency corresponding to the point. Figure shows that there is a very
little variation of phase velocity with the frequency over the operating band (F1 to F2) of
the tubes. Since the electron velocities must match with RF phase velocities for proper
interaction to occur, there is need for little variation of electron velocities and hence the
beam voltages.

       The best structure in terms of band width is the helix. Some tube with helix as
slow wave structure shows band width up to two octaves.

       The helix is supported inside the metalic vacuum envelope by some dielectric
rods which are held by simple contact obtained by mechanical pressure. This gives noise
to thermal resistance’s and poses a problem in transferring heat from helix to the ground
plane and reduces power handling capacity of the helix TWT. This problem is overcome
in brazed helix TWT where the helix is actually brazed to its support rods and the rods
inturn are brazed to the surrounding metal sleeves. The process also improves the
efficiency of the tube.

       The helix is normally fabricated from tranfor tape supported on dielectric rods.
Thermal energy from losses and beam interception must be conducted through these rods
to a heat sink in an efficient manner if a high degree of tube reliability is to be achieved.


                  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                          XVII - 4
        A new material for helix support rod, anisotropic boron nitride, has the high
thermal conductivity and also has very low dielectric loss, is now being employed.


        TWT fabricated with this new helix support material can safely & reliably
generate 500 Watts of CW power in C & X bands.



2.2.1   Focusing structure



        The slow wave structure is surrounded by focusing device which confines the
electronics in the beam by a strong axial uniform metallic field. The electronics in the
beam tends to repel each other causing the broadening the beam due to random variation
of their speed and the direction of motion. Some electronics are also intersected by the
slowwave structures. All these effects generate noise and the only remedy to limit the
performance degradation is to confine the beam and protect it from dispersing by some
good focusing field. There are different types of focusing devices used depending on the
requirement of the tube. Solinoid is used for high power tube where other types are not
sufficient for focusing. Permanent magnets of hollow cylindrical shape are used in TWT
from the very beginning but it has the disadvantage of size and weight.


        Periodic Permanent Magnet (PPM) : Since the beam is properly focused by a
magnetic field of either polarity it is greatly advantages to reverse alternate magnet
polarities as shown in the fig. 3 to reduce the size and weight of the tube. This is known
as Periodic Permanent Magnet (PPM) focusing.             As the weight of the magnet is
proportional to the cube of its length, the weight of the focusing structure can be reduced
by a factor of four where two magnets of opposite polarity for a given length are used.
The weight can be further reduced by increasing the number of magnets and hence the
number of field reversal. The disadvantage of PPM focusing is that the beam diameter
increases and hence some noise is generated along the length over which the reversal of
field takes place and the noise figure increases with the number of field reversal. This
means that reduction of size and weight of the tube due to use of PPM focusing structures

                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         XVII - 5
is at the cost of the noise figure. The use of PPM focusing also complicates the tube
design, not only because of noise figure but also because of iron pole pieces are usually
provided inside the tube and iron is a poorer conductor of heat than copper by a factor of
6.


2.3    Collector



       The collector is a hollow cylinder made of copper and is used to collect the beam
electronics at the end of transfer of energy to the RF wave. For low power TWT the
collector is generally connected to the helix. For high power TWT the collector may be
isolated from the helix and is operated at some potential between cathode and helix
voltage to improve the efficiency. The collector depression is limited to 50% by the
danger of electrons back streaming.

3.0    Principle of Operation



       The working of the TWT depends on the distributed interaction between the
travelling wave and the electron beam. The advantage of TWT over other microwave
tube in producing high power is due to its geometry. The regions of beam formation,
interaction and collection are independent to each other. It is, therefore, possible to
design each section to its optimum performance level. TWT employs a magnetically
focussed electron beam and a slow wave structure, such as helix. The applied RF signal
propagates around the turns of the helix and produces a travelling wave along the axis of
the helix. The electron beam velocity is adjusted to be slightly greater than the phase
velocity of the axial travelling electromagnetic wave. Under these conditions a strong
interaction takes place between the beam and the wave. In the process the beam delivers
the energy to the wave which grows stronger and stronger as it travels towards the
collector and the amplification takes place.




                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         XVII - 6
3.1    Mechanism



       The mechanism of operation for the energy transfer can be understood with the
help of the Fig. 4. The solid line shown in Fig. 4(a) represents the electric field due to
applied RF signal. It is assumed that the field travels towards right and the +ve field
accelerates the electrons. Consider an electron e, in the beam near the input end of tube
is under the influence of zero axial electric field and is un effected by the signal on the
helix. An electron e', just left of e, encounters a +ve electric field and is slightly
accelerated to catch the electron e. But the electron e'', just right of e, encounters a –ve
field and slows down to fall with the electron e. Thus electrons centered around e are
velocity modulated and the modulation continues to take place as the electrons continue
to travel with the wave towards the collector. The velocity modulation is transformed
into density modulation as the electrons around e are bunched together as shown in 2 of
Fig. 4 (c).   As the beam velocity is approximately the same as the wave velocity the
electron e in 2 of Fig.4 (b) still will be under zero wave field influence. However, the
bunch of electrons induces a second axial electric field, as shown by the dotted line in
Fig. 4(a) that lags (by a quarter wave length) the original field (solid line) produced by
the applied RF signal. The resultant wave shown in Fig. 4(b) lags a small fraction of
wavelength and is now closer to the electron gun.           So the electrons in the bunch
encounter a negative or retarding field and as a result deliver energy to the wave on the
helix. The wave becomes larger than at the input. The concentration of space charge in
the bunched electrons produces a repelling effect and the beam becomes de-bunched i.e.
the density modulation is converted back to the velocity modulation. As the electrons
travel further along the helix, the velocity modulated beam is again converted back to the
density modulation as shown at 3 in Fig. 4 (c) and the process is repeated. Fig. 4 (c) at 3
shows that the bunching is more completes and the induced wave grows in amplitude.
Each electron in the bunch now encounters a stronger retarding field and delivers more
energy to the wave which is much larger than the original signal. Analysis shows that the
amplitude of the resultant wave travelling down the helix increases exponentially.




                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         XVII - 7
3.2     Stability

        In order to avoid the oscillation from being generated in the TWT it is necessary
to prevent internal feedback arising from reflected wave. The situation is controlled by
introducing an attenuator moderately near the input end of the tube.

        The choice of high voltage has also improved other parameters that enhanced
system performance. The better focused electron beam results in much lower helix
interception, thereby, improving reliability. The noise figure or noise power is also
reduced due to better focusing.

4.0     Characteristics :


        The typical characteristic for Power Input vs. Power Output is shown in Fig.5(a).
The effect of Gain vs. frequency and Output power variation over the frequency band are
shown in the figure 5(b) and 5(c) respectively.


4.1      AM / PM Conversion :


        It is defined as the change in phase angle between input & output signal as the
input signal level varies. It is expressed in degree per db at specified output power.


        AM / PM conversion is due to the reduction in beam velocity as the input signal
level increases, causing greater energy exchanged but the beam and input required for
satisfaction.

4.2     Dynamic Range :

        It refers to the variation between the highest and lowest signal levels that can be
amplified linearly. Linearity is defined as the point where further increase in input
produces 1db of gain compression.

        Dynamic range (db)
        = 10 log (PIN / 1 mw) - (10 Log (KTB / 1mW) + N.F)

                    CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                            XVII - 8
4.3    Noise Figure


       It is ratio of S/N at input to S/N at output. Generally noise figure of high power
TWT is approx. 20 dB. Low noise TWT offers Noise Figure = less than 5 db.


4.4    Serrodyning:


       It is the technique where by a periodic saw tooth voltage applied to helix causes a
frequency shift of input to output signal.




5.0    CONCLUSION


       TWT dominated the vaccum tube industries for past few decades for its broad
bandwidth at high power. Unless there are some quantum breakthrough in high power
solid state device technology in the near future, the situation will remain unchanged for
some years to come.




                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         XVII - 9
                           Fig. 1



CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                        XVII - 10
a. Forward Wave




b. Backward Wave




                               FIG 2        ω-β DIAGRAM
                   CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                           XVII - 11
                  FIG 3



CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                        XVII - 12
                           FIG 4
CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                        XVII - 13
FIG 5 TWT CHARACTERISTICS
 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                         XVII - 14
     CHAPTER – XVIII


    DECOY SYSTEMS

Smt. Y.HEMALATHA, Sc'D'




 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                           XVIII
                        CHAPTER - XVIII


                       DECOY SYSTEMS


                             CONTENTS



1.0   INTRODUCTION


2.0   DECOY MISSION & STRATEGIES


3.0   ACTIVE DECOY ANALYSIS


4.0   CASE STUDY


5.0   CONCLUSIONS


      BIBLIOGRAPHY


      ACKNOWLEDGMENTS




         CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                   XVIII
CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                          XVIII
                                Decoy systems

                                Y.HEMALATHA,-Sc'D'
                                    DLRL, Hyderabad


1.0     Introduction :


        Expendables are the electronic warfare world’s “ bread - and - butter” .
Things have come a long way since the initial use of chaffs in World War II. On
the night of July 24th & 25th 1943, British aircraft dumped more than 40 tons of
chaffs- narrow strips of metal foils over Hamberg, Germany. More than 92
millions metallic dipole strips fluttered down and baffled the enemy’s radar. The
strips were pre cut to appropriate lengths to resonate with the known radar
frequencies. The enemy’s radar was saturated and the confusion followed to a
highly successful raid with minimal loss. Chaff is also an extremely versatile,
inexpensive countermeasure.          The use of chaff is not confined to protecting
aircraft – marine and ground-based assets can also benefit from the radar-clouding
effects of chaff.


            Radar systems can use different frequencies to see through chaff clouds
produced by precut dipoles. Doppler filtering can discriminate against the slowly
moving clouds of chaff. To overcome this problem, the EW community has
developed the Infrared (IR) missiles. Infrared Missiles home upon the infrared
(IR) emissions of aircraft engines have been the leading factor in aircraft losses
since their introduction in the early 1950s. IR Surface-to-air missiles were the
most destructive agents. The need to protect both military and civilian aircraft
from attack by effective, inexpensive, easily launched IR homing missiles is
critical.




                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                         XVIII - 1
       Conventional flares provided cost-effective protection against first-
generation, heat-seeking, surface-to-air or air-to-air missiles. The band pass of
early IR seekers was in the short wave length region of the IR spectrum, from
about 1-3 micrometers. This is the spectral range associated with hot engine
                                            0
exhaust temperatures in the 1,300-2,000         K range. Thus, protective flares were
designed to maximize the spectral output in this region. Another approach to
protection from missiles was to cool the aircraft’s hot exhaust gases before
venting them into the atmosphere. The response by missile designers to cooler
exhausts was to increase the sensitivity of the missile seeker and to shift sensor
responses to the longer wavelengths of cooler emitters.


       Missiles had advanced to the point where a simple flare could no longer
act as an effective target. Dual-spectrum sensors could discriminate between a
flare and an actual target. This, in turn, generated the need for smarter flares –
flares whose spectral signatures more closely emulated the emission of the target.
This resulted in the development of dual-component, bright/dim flares.


       But even if a flare’s spectral output can be customized to accurately
simulate that of the platform to be protected, a sophisticated seeker can
discriminate between a target and a flare by on-board, real-time computer analysis
of the flare’s kinematics’ or motion properties.


       In addition, chaffs & flares are for a short duration and stationary or slow
moving compared to the platform and hence can be easily filtered out by the
missile seekers onboard electronics.


       Present day missile seekers are with onboard active radar opening up at
short ranges, in short times. Modern missle designs have been developed which in
addition to    providing the required angle sensing capability have proven
invulnerable to airborne Defensive electronic counter measures (DECM)
Techniques. Hence, the on board ECM techniques cannot counter due to the short


              CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                      XVIII - 2
reaction times required by the threat. The Inadequacy of on - board DECM
Techniques has precipitated the need for the development of off-board Electronic
counter measures (ECM) for defence of aircraft. Offboard passive ECM alone is
proved to be ineffective and hence offboard active ECM is resorted to. In
response to this, the EW community searched for a new element to add to the
countermeasure suit      and developed Decoy. The job of Decoy is simple, if
jamming cannot prevent missile launch then the decoy will lure              the missle
away by advertising it self as the Target. When an aircraft deploys an off - board
Techniques such as a Towed decoy, the angle error of the missile can become
distorted by the presence of second RF source, the first source being the
reflections from the aircraft. Further more, if the Aircraft and Towed Decoy are
unresolvable in range, velocity, or angle the response of the missile seeker
becomes a complex function of the two RF sources introducing an error in angle
which translates into increased miss distance and increased probability of
survival of aircraft. Hence Towed decoy is most effective ECM against modern
present day missile seekers by virtue of its geometry w.r.t missile & platform.
Due to this towed decoy provides more J/S (RCS) and missile will look towards
decoy and detonate its warhead at a distance, safeguarding the platform.


2.0 Decoy Applications and Strategies :
       Decoys can be classified according to the way they are placed into service,
the way they interact with threats or the types of platforms they protect. Table 1
gives some of the general types of Decoys , their mission, and the platform they
protect.

       The Decoys can be divided in to three types namely expendable, towed
and “independent maneuver.” Expendable decoys are ejected from pods or
launched from aircraft and launched from tubes or rocket launchers from ships.
These decoys typically operate for short periods of time (seconds in the air,
minutes in the water).




             CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     XVIII - 3
        Table 1 Missions and platforms typically associated with types of
decoys.

        A Towed decoy is attached to the aircraft by a cable, with which it can be
controlled and/or retracted by the aircraft. Towed decoys are associated with long-
duration operation.

        Independent maneuver decoys are deployed or propelled, typically on
airborne platforms. Examples are UAV decoy payloads, ducted fan decoys used
in ship protection and decoys mounted on or below helicopters. When
independent maneuver decoys protect a platform, they have complete flexibility
of relative motion.

2.1     Decoy Missions

        Decoys have three basic missions: to saturate enemy defenses, to cause an
enemy to switch an attack from the intended target to the decoy and to cause an
enemy to expose his offensive assets by preparing to attack a decoy. Modern EW
decoys deceive the electronic sensors which detect and locate targets and guide
weapons to them.

2.1.1   Saturation Decoys

        Any type of weapon is limited in the number of targets it can engage at
one time. Since a finite amount of time is required for the weapon’s sensors and
processors to deal with each of the targets it attacks, the limitation is more
accurately described as a limit on the number of targets it can attack in a given
amount of time. The total time period during which a weapon can engage a target
starts when the target is first detected. It ends either when the target can no longer

             CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     XVIII - 4
be detected or when the weapon has succeeded in performing its mission. The
weapon will only be able to engage some maximum number of targets at once; if
more targets are present, some will escape attack, because the weapon must
operate above its saturation point.

       A large number of decoys can be used to saturate a weapon or a
combination of weapons       for example, an air-defense network. In general, the
radar processing associated with a weapon system can either ignore or quickly
discard the tracks of radar returns that are significantly different from the returns
of intended targets. Thus, to be effective, decoys must look enough like real
targets to the weapon system’s sensors that they cannot be easily rejected. The
more that is known about the sensors to be fooled, the more effective (and cost
effective) the decoys can be. Ideally, the attributes of the decoy are only those that
can be detected by the weapon system sensors; anything else adds size, weight
and cost. By the time an air-defense network processes all of the targets in Figure
2.1, the actual target may have accomplished its mission or may no longer be
vulnerable to attack.

       A special case of the saturation decoy mission occurs when the weapon
system acquires a decoy first and then stops looking for the target (Figure 2.2).
This is particularly important in protection against missiles with active guidance
— for example, antiship missiles which typically scan a narrow antenna beam to
acquire the target ship after the missile breaks the horizon.




Fig. 2.1 Saturation decoys force weapons sensors to deal with
large numbers of apparent targets, reducing their ability to attack the real
target.

             CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     XVIII - 5
Fig. 2.2 If a weapons sensor acquires a decoy before it detects a true target, it
may attack the decoy, wasting an expensive guided missile.

       Saturation decoys can be either passive or active, but they must provide
RCS approximately equal to that of the target. They must also provide other
characteristics detectable by the radar which are sufficiently close to those of the
target to “fool” the radar. Examples of these characteristics are motion, jet engine
modulation (if detected) and signal modulation

2.1.2 Detection Decoys

       A new and particularly valuable use of radar decoys is to cause a
defensive system, such as an air-defense network, to turn on its radars — making
it susceptible to detection and attack. This typically requires independent
maneuver decoys. If decoys look and act enough like real targets, the acquisition
radars or other acquisition sensors will hand them off to tracking radars. Once the
tracking radars turn on, they are vulnerable to attack by anti radiation missiles
fired from aircraft outside the lethal range of the enemy weapon (Figure 2.3).




             CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     XVIII - 6
       Fig. 2.3 If a decoy forces an air-defense radar to track it, an attacking
aircraft beyond the lethal range of the weapons system can attack it with an
anti radiation missile

2.1.3 Seduction Decoys

               In the seduction mission, the decoy attracts the attention of a radar
that has established track on a target, causing the radar to change its track to the
decoy. Then, the decoy moves away from the target, as shown in Figure 2.4.
Tracking radars consider only narrow segments of azimuth (and sometimes
elevation), range and return-signal frequency by use of angle, range and
frequency gates. If the decoy can move any or all of those gates far enough away
from the true target, the radar’s tracking lock on the target will be broken. Thus,
what we are calling seduction decoys could also be called break-lock decoys.

       Seduction decoys are so called because they “seduce” the tracking
mechanisms of threat radars away from their intended targets. Decoys function in
the seduction role after the threat radar has acquired the target. The purpose is to
capture the tracking mechanism of the threat radar and break its lock on the target.
Their function is much like that of a deceptive jammer (for example, a range-gate
pull-off jammer). However, the decoy is more powerful in that it holds the
attention of the threat radar, which continues to track it. The range-gate pull-off
jammer, on the other hand, pulls the radar’s range gate to a location which does
not contain a target allowing the radar to try to reacquire the target.




              CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                      XVIII - 7
  Fig. 2.4 In the seduction mission, the decoy activates within the radar’s
resolution cell with the target, but with high apparent RCS. It captures the
radar’s tracking gates and moves them away from the target.


         Another advantage of the decoy is, that its signals are transmitted from a
location away from the target. This defeats monopulse radars and home-on-jam
modes.

Operating Sequence

         As shown in Figure 2.5, the seduction decoy must turn on within the threat
radar’s resolution cell after the radar is tracking the protected target. To be
effective, the decoy must return radar signals with sufficient power to simulate a
radar cross section (RCS) significantly larger than that of the protected target. For
active decoys this requires adequate throughput gain and maximum power. For
passive decoys (e.g., chaff bursts for ship defense), the effective decoy RCS must
be greater than that of the target. The RCS of the target is a function of the
azimuth and elevation from which it is viewed, and maneuvering to reduce the
target RCS presented to the attacking radar may be an integral part of the
defensive strategy. It is also worth noting that the reduced RCS of modern
“stealthy” platforms allows a better level of protection at any given decoy RCS.

         As shown in Figure 2.6, the decoy captures the tracking mechanism of the
threat radar, so the resolution cell moves to center itself on the decoy as it

              CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                      XVIII - 8
separates from the target. In this figure, the decoy is falling behind the target, but
if the decoy is propelled it could as well move away from the target in any
direction. If the decoy seduction is successful (Figure 2.7), the decoy will pull the
radar’s resolution cell far enough away so that the protected target is completely
out




Fig. 2.5 Initially, the threat radar centers its resolution cell on the target. The
seduction decoy turns on within the threat resolution cell, presenting an RCS
significantly larger than that of the target

of the cell. At this point the effective “jamming to signal ratio” of the decoy is
infinite. It is important to note that for the decoy to be effective, it must be
indistinguishable from the target as perceived by the threat radar. If the threat
measures any signal return parameter that the decoy does not produce, it will
ignore the decoy and continue to track the target. Examples of parameters that
might be important are jet engine modulation and effects related to the size and
shape of the target.




              CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                      XVIII - 9
Fig. 2.6 The greater RCS of the decoy causes the threat radar’s resolution
cell to track it as it moves away from the target.




      Fig 2.7 When the threat radar's resolution cell is pulled far enough away so the target is
      no longer in the cell , the radar sees and tracks only the decoy



2.2    ACTIVE DECOYS:


       Today’s radar-guided missiles have the ability to counter jamming by
switching into a receive-only, home-on-jam mode that allows them to continue
their attack in the presence of jamming, in effect turning electronic
countermeasures into electronic guidance. In addition they open up at short ranges
and provide short reaction times with on-board computer analysis.


       As Passive decoys can be easily filtered out by the active missile seekers
onboard computer, active decoys are resorted to, to closely mimic the radars
return signal. The ultimate goal of an active decoy is to dupe an enemy radar or



             CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     XVIII - 10
missile into thinking that it is the legitimate target, thus protecting the platform
that deployed it.


       With active decoys, operators are able to specifically control what type of
signature is transmitted. The same can be done with decoys by generating
signatures that look ship- or aircraft- like and are specific to the threat that is
guiding it. The higher the power level, the better the signature and the more
attractive the decoy appears the threat. Active decoys are preferred in this case
when compared to passive decoys, because passive decoys are limited in terms of
the signature or radar cross section (RCS) they can produce.


       In practice the decoy can actually use a primed oscillator to output a signal
at some large fixed power at the frequency of the received signal. In this case the
effective gain and the resulting equivalent RCS can be extremly large and the
radar is distant from the target. However as the radar approaches the target the
effective gain (hence RCS) decreases.


2.3    REPEATER DECOYS :-


       Typical Repeater jammers simply receive the transmitted signal of the
radar, amplify it, and retransmits it back towards the radar and/or to disturb the
tracking capabilities of the radar and /or missile. "Repeater" decoys comprise two
main elements. A launcher/controller subsystem houses the decoy before it is
deployed and provides power to the decoy after it is launched. The decoy body
contains a receiver and transmitter and a self-contained system except for the
power supply. When it receives a signal from a threat radar, it amplifies it and
retransmits it, making it look like an aircraft that has reflected the original radar
signal. Ofcourse the radar receives two signals— one bounced off the aircraft and
a stronger identical signal coming from the decoy. Unable to distinguish between
the two signals, the radar or missile seeker assumes that the stronger of the two is
the target. In addition to the repeated signal, the system can also add a small


              CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                      XVIII - 11
modulation to mimic the aircraft’s engine signature to fool radars that look for
such discrimination. A typical repeater decoy is shown in fig 2.12.




                       Fig 2.12 A typical Repeater decoy
       One interesting feature repeater decoy is the ability to recover the decoy
after it has been Deployed. When used on fighter aircraft the decoy is equipped
with a parachute, thus making it potentially recoverable. When used on transports,
it can be retracted after use and placed back into its dispenser.


2.4    Towed Decoys : -


       The Towed decoys is an aerodynamic body tethered to a moving
platform.Typical Active decoy towing from the mother aircraft is shown in fig
2.13.When an aircraft deploys and off-board technique such as towed decoy, the
angle error of the missile can become distorted by the presence of the second RF
source. Further more, if the aircraft and the towed decoy are un resolvable in
range, velocity, or angle, the response of the missile seeker becomes a complex
function of the two RF sources introducing an error in angle which translates into
increased miss distance and increased probability of survival. Taking advantage
of the spatial separation between the expendable and the host aircraft, the stronger
signal lures a threat missile away.     The towed decoy may be passive or active.
Active decoys are normally repeaters. The aim in both cases is to provide a signal
from the decoy which is much larger than that from the towing vehicle so as to
force the radar or missile tracker to lock to the decoy. One immediate advantage
of a towed decoy is that velocity discrimination between target and decoy is not
possible. As long as the aircraft and decoy remains un resolved by the radar or


              CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                      XVIII - 12
missile seeker the radar will either track the signal “center of gravity “ of the
aircraft – decoy system or will lock on to the dominant source, the decoy. The
angular error caused by the decoy obviously depends on the geometry and will be
at maximum when the aircraft is flying on a tangential course past the radar.
The effectiveness of the deception depends on the relative magnitudes of the
jamming signal (J) from the towed decoy and the radar signal (S) reflected from
the aircraft. For airborne towed decoys practical tow-line lengths are of the order
of a few hundred meters and this type of counter measure is primarily of use
against semi-active and active missiles.




                                                                    Fibre optic link




                                                  Towed decoy
Fig 2.13 Towed Active Decoy integrated to the aircraft via fibre optic link

2.5    IDECM :-


       The Onboard countermeasures aimed at the RF threats, are positioned
either internal or external to the platform. Offboard Countermeasures refer to
those systems that are carried by the aircraft, helicopter, or ship but are deployed
when under attack (i.e., missile launch) by the threat. The use of offboard


             CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     XVIII - 13
countermeasures implies a missile attack warning(MAW) system that triggers the
offboard action at the appropriate time and in the appropriate direction. The
integrated onboard/offboard system potentially provides a higher level of
survivability than either system alone. The onboard system can degrade the
threat's target acquisitions, target track, and missile guidance functions ; while the
offboard system is effective only in the endgame to decoy the misile from its
intended target.


       The Integrated Defence Electronic Countermeasure (IDECM) system is a
combination of onboard/offboard ECM system. A typical IDECM system is
shown in figure 2.14. The onboard (techniques generator TG ) portion is based
upon the Self Protection Jammer design. The TG is designed to apply a variety of
RF techniques against pulsed/ CW and PD threats. This modulated RF signal is
converted to light in a laser modulator and transmitted to the towed decoy through
a fibre optic link. The FOTD (fibre optic Towed decoy) converts the laser signal
to RF and amplifies using a TWT amplifier and radiates this jammer signal
towards the missile/radar.


       Using an onboard jammer to generate RF techniques and FOTD to
transmit them gives the aircraft the benefits of both a smart onboard jammer and
an offboard repeater.


       Despite all the fanfare, active decoys are not a replacement for such
passive expendables as chaff. Chaff still serves a useful purpose against a wide
range of RF threats. It is not an "either/or" of passive and active decoys. One is
not better than the other. In fact, active decoys can be more effective when
deployed just after a few rounds of chaff have been dispensed. In this type of
scenario, chaff can temporarily confuse the missile’s RF seeker, and when the
missile re-acquires the target, it is in fact tracking the decoy. Active decoys do,
however, have definite advantages over passive countermeasures. Obviously the
best decoy one could have for a ship would-be another ship. With an active decoy,


             CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     XVIII - 14
      a better representation is transmitted for the enemy to see than with a passive
      device. In some ways, active decoys can even be better than another ship because
      the decoy can look even larger. Typically, a missile will detect a target presenting
      the biggest RCS; therefore it is ideal to make an active decoy that appears to have
      a tremendously large RCS.




                 RWR                    Processor                      Technique              RF to
                                                                       Generator              Light
Rx
Ant
                                        Aircraft




                                                  Fibre optic link


                                                      Towed decoy
                                                                                  Tx ANT
                                      Light to
                                      RF                         TWTA




                         Fig 2.14 BLOCK DIAGRAM OF IDECM
                         ARCHITECTURE

      3.0 Active decoy analysis:


             The effectiveness of the deception depends, in part, on the relative
      magnitudes of the jamming signal (J) from the towed decoy and the radar signal
      (S) reflected from the aircraft. Typical repeater jammers simply receive the
      transmitted signal of the radar, amplify it, and retransmit it back with a time delay


                   CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                           XVIII - 15
toward the radar and /or an oncoming missile. In the usual case, the received
signal is modulated in some way to disturb the tracking capabilities of the radar
and /or missile.   ECM techniques like Range Gate Pull Off (RGPO) and/or
Velocity Gate Pull Off (VGPO) are often used to disrupt the missile tracking
capabilities. However, in order for the ECM techniques to be effective, the
jamming signal (J) radiated at the radar has to exceed the skin return (S) reflected
from the aircraft toward the radar. The typical geometry of the missile, aircraft
and radar is considered as shown at fig 3.1. The geometry clearly shows the
length (TL), angle (γ ) at which the decoy is towed from the platform. The
analysis is carried out initially considering the decoy as a collocated jammer as
onboard ECM . For this the simple J/S equation is given by


       J/S (dB) = 10 log{[4πPJGJTLr) / (PTGTσL)]R2 }           --------------   (1)


       Where PT is the radar transmitter power in watts. GT is the radar antenna
gain as a numeric, R is the distance between the aircraft and radar in meters, L
and Lr is the path dependent losses , σ is the radar cross section (RCS) of aircraft
in square meters , PJ is the saturated power output of the jammer and GJT is the
jammer transmit antenna gain.[1]


       The J/S ratio for the case of Towed decoy scenario in fig 1, considering
linear repeating jammer is given by


       J/S (dB) = 10 log[(GC(γ)GEλ2 /(4πσL)] + 20 log((sinγ)/(sin(γ-θ)))2
                      + 24/θβ2 ) (sin-1(TLsinγ/RT))2.                   ---------(2)


where GC(γ) is the combined gain of the Transmit and receive antennas and GE is
the loop gain of the repeater jammer.[1]


       However at short range, the second term gives a slight increase in J/S due
to the proximity of the towed decoy to the missile, as compared with the aircraft.

             CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     XVIII - 16
The third term describes a very large increase in J/S, at short range, due to the
angular discrimination of the missile seeker to the towed decoy.


       For the given geometry a similar equation can be derived for the case of
saturated jamming. The result is [1]


       J/S (dB)
          = 10 log{[(4πPJGJT(γ)Lr)/(PTGTσL)]RT2} + 10 Log((sinγ)/(sin(γ-θ)))2
                         + 24/θβ2 ) (sin-1(TLsinγ/RT))2.                 ----------- (3)




      Fig 3.1 Spatial separation between aircraft & Towed decoy


4.0 CASE STUDY :


       The Towed decoy Jammer can create a high RCS, so that the tracking
radars will lock on to the decoy instead of the actual target. This present case
study is carried out to compare the J/S ratio created by the decoy as compared to
that of onboard ECM .It shows how the offboard ECM (Towed Decoy) is
preferable to that of its counterpart On board ECM. Consider an Aircraft with a
towed decoy (repeater) providing jamming signal for three cases as given in
section 3.0.
                  •   on-board self protection ECM case

               CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                       XVIII - 17
               •     Offboard active decoy in linear repeating mode.
               •     Offboard active decoy in saturation.


This is carried out for a typical ground tracking radar and for typical missile
seeker parameters.


4.1 GROUND TRACKING RADAR :


       The ground tracking radar illuminates the attacking aircraft and as well as
the towed decoy. The received signal of the decoy is amplified by the repeater
electronic gain (GE) and undergoes additional transmitting antenna gain (GTx) and
is transmitted towards the radar. The radar receiver receives two signals, the echo
from the aircraft and the jammer power from the repeater decoy. Initially, we
consider onboard ECM and compare its performance that of a Towed Decoy .
The J/S is constant at long ranges and dependent on electronic gain of the
amplifier in the repeater. After saturation range Rr, J/S falls with distance. This is
shown in fig 4.1(a). As the Self protection jammer approaches the radar and
crosses the saturation range Rr , the J/S ratio falls steeply with a slope of
20dB/decade and at a certain distance, it can be as low as 0 dB(Burn through ).
The decoy as off board jammer case in linear and saturation conditions is also
plotted . The Fig 4.1(a) also demonstrates the increase in J/S ratio for linear
repeating Jamming at short ranges due to the favorable geometry. It is seen that
the J/S ratio instead of falling at 20 dB / decade rate, actually increases sharply at
the short ranges, and can reach a very large value (30 to 40 dB). Hence the
situation of the echo signal burning out from the jammer signal will not arise at all
for the repeater decoy. Even though the repeater enters the saturation range (Rr ),
the third plot shows the J/S ratio is always positive and the decoy thus protects the
aircraft even at very short ranges.




              CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                      XVIII - 18
4.2 ACTIVE MISSILE SEEKER :


       The same exercise is carried out for the case of a typical missile seeker
and the three plots are shown at fig 4.1(b). For the same typical missile seeker ,
the dependence of J/S ratio on towing angle and towing distance is also worked
out. The J/S ratio at the missile/radar is highly depends on the towing angle γ . It
is evident from the Fig 4.1© that the geometry of the decoy with respect to the
missile bore sight, plays an important role for the J/S ratio. When the angle γ =
900 , the J/S ratio is around 6 dB, and when γ = 450 , the J/S ratio is increased to
16dB ( 10 dB increase), and hence corresponding increase in J/S ratio at /below
saturation ranges . The effect of Towing length on J/S is also studied for 150,
200,300 m of towing length It is evident from the figure that the increase in
towing length increases the J/S ratio. By increasing the Tow Length from 150m
to 200 m, the J/S ratio is increased @ Rt = 1Km by 4 dB, where as for 300 m it
is increased by 9 dB. It is shown at fig 4.1(d).
From the study, It can be seen that the J/S ratio increases very rapidly at
short ranges due to the geometry of TOWED DECOY.
It is further emphasized that in saturation condition, J/S tends to reduce, but
increases enormously for short ranges and never enters the burn through.


4.3 REPEATER TOWED DECOY:


       A simplified block diagram of a repeater decoy is worked out to cover the
front/ rear sector is shown in fig 4.2 . The receiving antennas (Horn/cavity
backed spirals) receive the emissions from radar/missile. These antennas have a
typical gain of 5 to 6 dB with a beam width of 750 (azimuth and elevation) . A
limiter protects the other components in the microwave chain from excess power
emanated from the radar/ missile and limits the power to +20 dBm. A solid state
amplifier (SSA) with adequate gain(around 35 - 40 dB) and P1dB of +25dBm
provides the necessary amplification to the intercepted signal and drives the high
power traveling wave tube amplifier. It is preferable to modulate the input signal

             CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     XVIII - 19
before giving it to the TWT amplifier. This modulation simulates the
characteristics of the target under attack by the radar or missile, so that the
radar/missile cannot




                                                                  (c)
            (a)




                                                                        (d)
                  (b)
                                                   Fig 4.1



distinguish the real target and the decoy. The missile then will be lured away by
the Repeater decoy. A Technique generator generates the required signal. The
TWT amplifier with a typical output power of 300 watts and gain of around 35 -
40 dB amplifies this signal further and thus generates the required jammer power
to counter the radar/missile with a high J/S ratio. The TWT amplifier must be
protected from excess VSWR and over helix current (either in the Tx antenna or

             CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     XVIII - 20
              in the high power RF cables). A portion of the reflected power is sampled in the
              dual directional coupler , detected and used to trip off the high voltages for
              protection. The transmitting antenna with the same characteristics as that of the
              Rx antenna will direct the power towards the radar / missile.
              Power Budgeting -active missile seeker : The power budgeting of the typical
              repeater decoy at fig 3 is calculated against active missile seeker with typical
              parameters
              PT = Transmitted power of the missile radar = 80 KW (79dBm)
              GT = Gain of the radar transmitting antenna = 28 dB ; Frequency = 13300 MHz
              D distance of the target from the radar = 5 Km
              ERP of the radar = PT +GT = 79 + 28 = 107 dBm
              Path loss = 32 + 20log(D Kms) +20 log (F MHz) = 32 + 20log(5) +20log(13300) =
              128 dB
              Power received at the receiving antenna = ERP - Pathloss = 107-128 = -21 dBm
              Input to the Tx. Antenna = 50 dBm ; Gain of the Tx antenna = 6dB
              The ERP of the repeater jammer = 50 +6 = 56 dBm
              If more power is required , a TWT of higher output power can be selected. As
              the path loss depends upon the frequency of the missile radar and range from the
              target, the jammer power varies. For 3 Km distance the ERP of jammer is 59
              dBm.




            -15dBm              +19 dBm              18dBm                 +53dBm
  -21
                                                                                            Ant G = 6dB
  dBm
                             SSA            Tech                 TWT
                Limiter                                                             DDC        Transmitted power = 56 dBm
F Antt Rx                                   Gen              Gain = 35dB
                (1 dB)                                                              (1dB)
                                            (1dB)            P =300Watts
                                                                                            F Ant TX
                       G = 35dB
                       P = +25dBm                                                             +50dBm

                                    Fig 4.2 Power budgeting of Repeater decoy




                            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                                    XVIII - 21
5.0 CONCLUSION:


       The J/S ratio for two cases, one for the ground radar and one missile
radars is calculated. The dependence of J/S ratio with the Towing angle and the
towing length of the decoy are worked out. Normally the J/S decreases with
decreasing range. However, the additional J/S provided by the geometry of the
Towed decoy is very desirable result which makes the Towed decoy a particularly
attractive countermeasure. Increasing towing length will increase the J/S, but
increasing towing length beyond certain limit will make the decoy out of the
Radar's range resolution call.


REFERENCES:


       1.    William J. Kerins "Analysis of Towed Decoy " IEEE Transactions
             on Aerospace and Electronic System           Vol. 29, No. 4 October,
             1993.pp1222 - 1227
       2.    "Anti-Ship missile threat drives Decoy Development "           Defence
             Electronics , March 1988, pp109-120
       3.    EW 101:Tutorials – by David Adamy Year: 2001, Artech House,
             Boston.
       4.    Electronic Warfare in the Information Age – by D.Curtis Schleher
             Year:1999, Artech House, Boston.


ACKNOWLEDGEMENTS :


       The author thank Director DLRL, Sri G. Kumaraswamy Rao Sc'H' for his
continuous encouragement in this work. The author also thank            Mrs V.Shoba
Rani, for the documentation help.




             CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     XVIII - 22
         CHAPTER – XIX


MULTISENSOR INTERFACE FOR
 INTEGRATED EW SYSTEMS

    J. SHANKAR RAO, SC-‘F’
               &
        M. SANTHA, SC-B




     CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                XIX
                       CHAPTER - XIX


MULTISENSOR INTERFACE FOR INTEGRATED EW
                           SYSTEMS


                          CONTENTS


1. INTRODUCTION
2. RS 232C SERIAL INTERFACE
3. ETHERNET LAN
4. MIL – STD – 1553
5. IEEE – 488 INTERFACE BUS (HP – IB / GP- IB)




     CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                XIX
      MULTISENSOR INTERFACE FOR INTEGRATED EW
                      SYSTEMS

                 J. Shankar Rao, Sc-‘F’ and M. Santha, Sc-B
                             DLRL, Hyderabad


1.0     INTRODUCTION:


        Modern Integrated EW systems acquire data from multiple sensors
such as Radar, EW receivers, IR/EO systems and lasers. This data is analyzed
and useful information is extracted for generating an appropriate EW response.


        Data is communicated over a variety of interfaces depending on the
volume of data to be communicated and platform constraints. Some of the
standard interfaces can be broadly classified into two types – serial and
parallel interfaces.


       Serial Interfaces                             Parallel Interfaces


1. RS 232C/RS 422/RS 423                        1. IEEE 488


2 RS 485                                        2. CENTRONIX


3. ETHERNET LAN                                 3. PARALLEL PORTS


4. MIL STD 1553                                 4. PC BUSES - ISA/EISA, PCI,
                                                   CPCI
5. FDDI


6. ARINC


7. USB/FIREWIRE


        In addition, the recent trend towards Network Centric systems – both
wired and wireless systems. The extensive set of interfaces available for
             CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                      XIX - 2
sending command & control information and measuring the data from various
sensors forces the EW system designer to cater for appropriate interfaces in his
EW system design.


        Let us now examine in greater detail the physical            and electrical
characteristics of various types of interfaces,          various modes of data
transmission, Hardware and software protocols, data rates achievable and
various techniques used to ensure data integrity.


        Though information can be sent in an analog or digital form,
transmission of data in digital form has many advantages:


        It is less susceptible to noise
        Error detection and correction codes can be used
        Allows for re-programmability of functionality


2.0     RS 232C SERIAL INTERFACE :


         RS 232 C       is a Serial interface standard defined by Electronics
      Industries Association (EIA) and defines the physical and electrical
      interface between a Data Terminal Equipment (DTE) and a Data
      Communication Equipment(DCE). Fig 1 gives the typical example of
      DTE and DCE are shown in fig 1.


      Typical examples of a DTE’s ara PCs or FAX machines.
      Typical examples of DCE’s are Data and FAX Modems.


      Electrical Specifications: The electrical specifications of RS 232 C
      defines the minimum and maximum voltages of a logic ‘1’ and ‘0’.
      Negative logic is used for data signals .i.e , logic ‘1’ is represented by a
      voltage between –3v to –25v and a logic ‘0’ is represented by a voltage
      between +3v to +25v. Hand shake signals follow positive logic.



            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     XIX - 3
     Physical characteristics: The full RS 232 C standard is defined by 25
     signals and allows full duplex operation of a main and auxiliary channel.
     fig 3 shows pin configuration of connector.
     Description of the pins (signals) is as follows:-


     Pin No.     Name                              Description
     2          TD          Data is sent from DTE to DCE
     3          RD          Data is sent from DCE to DTE
     4          RTS         DTE sets this active when it is ready to transmit data
     5          CTS         DCE sets this active to inform DTE when it is ready
                            to receive data
     6          DSR         Similar to CTS but activated by DTE when it is
                            ready to receive data
     20         DTR         Similar to RTS but activated by DCE when it is
                            ready to transmit data.
     7          GND         All signals are referenced to the signal ground
                            (GND)


     In addition to the above signals, other signals, other signals are defined
such as Ring Indicator(RI), Data carrier Detect(DCD) etc. which are used by
modems for communication between two DTE’s such as between two pc’s
through a PSTN ( Public Switched Telephone Network)


Data Format & Synchronization:                RS      232 C uses asynchronous
communications       which means, synchronization is maintained between
Transmitter and receiver on character by character basis. Each character (data
or byte) is preceded by a start bit and ends with 1 or 2 stop bits . An optional
parity bit is added to ensure error detection. An active low transition on the
line indicates, start of transmission. Asynchronous         communication frame
format is shown in fig 4.


Coding : while the most popular code used is 7-bit ASCII (American
Standard Code for Information Interchange). Other codes such as 8 bit

            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     XIX - 4
EBCDIC (Extended Binary Coded Decimal for Information Change) or 5 bit
BAUDOT code are also used in various applications.


Baud rate: This is the rate at which transitions of the symbol are occurring
(Symbol rate). In the present case, each signaling element (symbol) is encoded
into one bit and therefore it is identical to bit rate. Typical baud rates used
        varies from 300 baud to 9600 baud. The RS 232 c standard defines the
interfacing of a DTE to DCE over maximum distance of up to 50 ft and at a
maximum data rate of up to 20 KBps ( kilo bits per second)


EIA Standard RS 422 & RS 423 : To upon the specifications of RS 232 C, EIA
has defined RS 422 & RS 423 standards improve which support higher data
rates up to 10 MBps(1 Mega bit per second) and longer transmission distances
(4000 ft) or 1200meters. This is made possible by having line drivers with
more current buffering capability and balanced lines. RS 422 uses balanced
lines with differential line drivers. Fig 4a shows RS 422 interface using
balanced lines. Fig 4b shows RS 423 interface using single ended line.


HANDSHAKING UNDER RS 232C :                   The full specification of RS 232C
need not be implemented in every application which uses interface. This is
mainly because not every device requires the full functionality of RS 232 C,
for example a modem requires many more lines than a serial mouse. The rate
at which data is transmitted and the speed at which the transmitter and
receiver can transmit/receive the data dictates the amount of data handshaking
required.


        There can be either no handshaking, hardware handshaking or
software handshaking


3.0     ETHERNET LAN


      Ethernet LAN specification has been developed by Xerox corporation,
and is the most widely used LAN technology.

            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     XIX - 5
Definition: Ethernet is a multi-access, packet switched network that uses a
passive broadcast medium, with no central control. Its architecture provides
for error detection but not error correction.


       Data units transmitted over the network reach every station and each
station is responsible for recognizing the address contained in a data unit and
for accessing data units addressed to it from the transmission medium. Access
to the transmission medium is governed by the individual stations, using a
statistical arbitration scheme.


       Functional model: fig 9 shows the functional model of Ethernet. The
software layer operating above the data link layer is called client layer . The
client layer works with data units called packets. The client layer passes
packets to the data link layer for transmission through the network.


       Data Encapsulation/ Decapsulation: When a packet is received from
the client layer for transmission, it is encapsulated with the control information
necessary for transmission , called a frame. When the frame is received , the
control information is removed, before the data is passed to the client layer in
the form of the original packet.


       Link management: The link management function is responsible for
collision avoidance and collision handling. Collision avoidance involves
monitoring the carrier sense signal and defering transmission if the channel is
already in use. Collision handling involves transmitting a jam signal, to ensure
that all stations are aware of the collision, and then scheduling a re-
transmission attempt , using a defined algorithm, to determine the time to wait,
before retransmitting.


       Data encoding/Decoding:        Encoding involves adding bits needed for
synchronization and converting a binary signal to phase-encoded form, using
the Manchester encoding scheme. Decoding involves converting back from
Manchester to binary and removing the synchronization bits.

            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     XIX - 6
        Channel access:        The channel access function transmits bits to and
receives bits from the transmission medium or cable. It also senses the carrier
(The presence of data transmission on the medium) and detects collisions.


        Layer to layer services:       A lower layer provides services to its
immediate upper layer. The physical layer to the data link layer and the data
link layer to client layer .


        Ethernet Frame format: The data link layer is primarily concerned
with producing a correctly formatted frame for transmission across the
network and then processing a received frame to be sure it has arrived
correctly before passing it to the client layer. The processing performed is
determined by the frame format that is defined as part of Ethernet
specification. The Ethernet Frame format is shown in fig 10.


        Preamble: The preamble is used to provide synchronization and to
mark the start of the frame.


        Address fields: The frame inmcludes both the source and destination
address fields. Ethernet specifies the use of 48-bit addresses.


        Type field: The value specified in the type field is meaningful to the
higher network layers.


        Data field: The data is passed to the data link layer by the client layer.
It is a multiple of 8 bits. The minimum frame size is 72 bytes and maximum
frame size is 1526 bytes, including the preamble. If the data to be sent is
smaller or lesser than these sizes it is the responsibility of the higher layers to
pad it or break it into individual packets.


        Frame check sequence: Ethernet uses a frame check sequence field as
way of providing error checking(CRC) value that is calculated from the other
fields of the frame. When the frame is received, the value is compared to the

             CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                      XIX - 7
value in the frame. If the two do not match., an error has occurred, and this is
reported to the client layer.


          Ethernet physical model:             Fig 11 illustrates the physical
implementations model that is assumed in the Ethernet specification. The data
link functions and the data encoding/decoding function of the physical layer
are packaged in a controller board, or card, that is installed in a network
device, such as a personal computer. A device called a transceiver implements
the channel access function and is located close to, or directly on the coaxial
cable. A transceiver cable is used to connect the transceiver to the controller
card. In Ethernet, functions of the physical layer are referred to as the physical
channel, or just channel. This includes the logic in the controller board that
does encoding and decoding, preamble generation and removal, and carrier
sensing. The transceiver contains the logic required to send and receives bits
over the coaxial cable and to detect collisions.


          Data Encoding/decoding:       Ethernet uses the Manchester encoding
scheme shown in fig 12. With Manchester encoding there is a transition in
every bit. A ‘1’ bit signal goes from low to high, ‘0’ bit from high to low. The
transition in every bit allows clocking to be combined with data transmission.
It also    allows    carrier to detected by the presence of transition       in the
transmission medium. The data encoding function is also responsible for
adding a preamble to every frame transmitted. Data decoding removes the
preamble before the frame is passed to the data link layer. Carrier sensing is
also performed at this level.


          Ethernet physical specifications: The Ethernet specification defines
electrical, mechanical and physical characteristics for the components of the
physical channel. The most commonly use Ethernet implementation uses base
band transmission over co-axial cable at a data rate of 10 Mbps. The
maximum cable length is 500m, making this an implementation of 10 Base 5
transmission.



              CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                       XIX - 8
          Thin and thick Ethernet: The initial Ethernet implements used a
relatively expensive coaxial cable, now called thick Ethernet cable . Ordinary
CATV- type coaxial cable called thin Ethernet cable is used in many low-cost
Ethernet implementation. A form of Ethernet has been developed that uses
twisted-wire pair cable, to support a data rate of 10 Mbps.


4.0       MIL – STD – 1553


          MIL-STD-1553 “Aircraft Internal time –Division command/response
multiplex data bus “. It is a differential serial bus used in Military and space
equipment has been in use since 1973 and that is widely applied. MIL-STD-
1553 is referred to as “1553 with the appropriate revision letter (A or B) as a
suffix.


          In recent years the use of digital techniques in aircraft equipment has
greatly increase. Because analog point- to- point wire bundles are inefficient
and cumbersome means for interconnecting the sensors, computers, actuators,
indicators, and other equipment on board the modern Military vehicle, a serial
digital multiplex data bus was developed. MIL-STD-1553 finds all aspects of
the bus, therefore, many groups working with the Military tri-services have
chosen to adapt it.


          There are only three functional modes of terminals allowed on the data
bus. They are 1. bus controller 2. the bus monitor 3. the remote terminal.
Devices may be capable of doing more than one connection, typical bus
configuration is given in fig 13.


          The bus has single active bus controller(BC) and upto 31 remote
terminals (RTs). The BC manages all data transfers on the bus using the
command and status protocol. The bus controller initiates every transfer by
sending a command word data if 1553 defines Bus monitor as the terminal
assigned the task of receiving bus transfer and extracting selected information
to be used at a later time. The selected RT will respond with a status word
and data if required .
              CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                       XIX - 9
          1553 command word contains          a five bit RT address, transmit or
receive bit, five bit sub- address five-bit word count. This allows for 32 RTs
on the bus, but only 31 RTs may be connected, since the RT adddress (31) is
used to indicate a broadcast transfer ie all RTs should accept the following
data. Ch RT has 30 sub-address reserved for data transfers. The other two sub-
addresses (0 and 31) are reserved for mode codes used for bus control
function. Data transfers contain upto(32) 16-bit data words. Mode code
command words are used for bus control functions such as synchronization.
1553 bus supports ten message transfer types, allowing basic point- to- point
and broadcast BC-to-RT data transfers.


          Word formats: There are only three types of words in a 1553 message.
They are a command word(CW), a data word (DW), and a status word(SW).
Each word consists of three bit sync pattern, 16 bits of data and a parity bit,
providing the 20 bit word.


          Information transfer: Three basic types of information transfers are
defined by 1553.


     1. Bus controller to Remote terminal transfers.
     2.    Remote terminal to bus controller transfers.
     3. Remote controller to remote controller transfers.


          These transfers are related to the data flow and are referred to as
messages.


          The normal command/response operation involves the transmission of
a command from the BC to a selected RT address. The RT either accepts or
transmits data depending on the type of command issued by the BC. A status
word is transmitted by the RT in response to the BC command if the
transmission is received without error and is not illegal. Data encoding of
1553 is shown in fig 14



              CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                      XIX - 10
5.0       IEEE – 488 INTERFACE BUS (HP – IB / GP- IB)


       In the early 1970’s, Hewlett Packard came out with a standard Bus
(HP- IB) to help support their own laboratory measurement equipment product
lines, which was later adopted by IEEE in 1975 as IEEE std. 488. The IEEE
488 Interface Bus (HP-IB) was developed to provide a means for various
instruments and devices to communicate with each other under the direction of
one or more master controllers


       Description: The HP-IB specification permits up to 15 devices to be
connected together in any given setup, including the controller . A device may
be capable of any three types of functions. Controller, listener, or talker. A
device on the bus may have only one of the three types of functions active at a
given time. A controller directs which device will be talkers and listeners. The
bus will allow multiple controllers, but only one may active at a given time .
Each device on the bus should have a unique address in the range of 0 – 30.
The maximum length of the bus network is limited to 20 meters total
transmission path length. It is recommended that the bus be loaded with atleast
one instrument or device every 2 meter length of cable(4 meters is maximum).


       Electrical Interface:     The GP-IB is a bus to which many similar
modules can be directly connected, as is shown in fig 15. A total of 16 wires
are shown in the figure 12 . Eight data lines and eight control lines. The bus
cables actually have 24 wires, providing eight additional for shielding and
grounds. The GP-IB defines operation of a three wire handshaking that is
used for all data transfers on the bus. The bus operation is asynchronous in
nature. The data transfer rate can go up to ! megabyte per second. The bus is a
two way communications channel and data flows in both directions.


       GP-IB Bus structure: Fig 16 illustrates the structure of the GP-IB bus
and identifies the 16 connections of the interconnecting cable. Cabling
limitations make it less- than-ideal choice for large separation between
devices. The cabling specifications of the GP-IB interface system permit
interconnecting all devices together in a star or linear configuration. The GP-
            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     XIX - 11
IB connector is a 24-pin ribbon – type connector. Fig 17 shows the pin
assignment on 1 24-pin amphenol connector.


       Since the Introduction of IEEE –488, it has been the interface of choice
for a generation of medium speed, low power implementation which had a
need to operate in an automatic test system setup. With the availability of
device drivers for popular languages such as C, C++, VB, and VC++, it has
simplified the development of application software for microprocessor/pc
based applications for automation and industrial control.


       Summary: During the course of this lecture, I have touched upon the
most popular and widely used interfaces in some detail such as RS 232 C,
MIL-STD_1553B, Ethernet , LAN and GP-IB. Historically RS232C serial
interface standard was developed to support point to point data
communications, and the capability to communicate to larger distances and
high data rates was extended using RS 422 & RS 423 interfaces. RS 485 was
evolved as a multi-drop network with one controller and several slaves. Over
the some period GP-IB evolved as instrumentation interface standard. It’s
great advantage is its simplicity and the availability of suitable language
drivers from many sources such as HP & National Instruments to help in
automation and industrial control. The wide spread usage of components and
computer network has resulted in the development of Ethernet LAN as one of
the most popular technology to communicate data between computers and
peripherals. The trend is continuity towards fast and switched Ethernet and
FDDI networks to support higher data rates of 100 Megabites/second and
beyond. Simultaneously on the PC front, universal serial bus (USB) is fast
becoming the choice for communicating data between the pc and its
peripherals such as mouse, keyboard and printer etc. will gradually replace
RS 232 C & centronix ports.




           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                   XIX - 12
Conclusion:


          The need for catering to multi-sensor interfaces in EW systems have
been discussed in brief. Various interfaces in EW systems have been described
in some detail including the most popular serial interfaces such as RS 232 C,
MIL-STD- 1553b and Ethernet LAN. The most popular parallel interface GP-
IB has been discussed. Finally a comparison of the various technologies and
their evolution has been discussed.


REFERENCES :


   1.      PC interfacing using Centronix, RS232 and Game Ports by Pei An
   2.      Applied PC interfacing, graphics and interrupts by Willian
           Buchanan
   3.      Local Area Networks, Architectures and Implementations by James
           Martin.




           CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                   XIX - 13
                           RS 232 C
                           INTERFACE
 PC OR FAX MC                                          DATA/ FAX
   (DTE)                                              MODEM (DCE)



FIG 1. TYPICAL EXAMPLE OF A DTE AND DCE



                                 1
                                      14




                                      25

                                 13




     FIG 2 : 25 – PIN D TYPE CONNECTOR




       CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                XIX - 1
                      MODEM                      PSTN                       MODEM
   PC                                                                                              PC




     FIG 3. A TYPICAL EXAMPLE OF COMMUNICATION BETWEEN TWO DTE’S




Idle (high)                                                                parity Stop bits   Idle (high)


              start   D0                                              D7    P



         Fig 4 : RS 232 C ASYNCHRONOUS COMMUNICATION FRAME FORMAT




                       CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                                XIX - 2
                                                      NOISE




      TX+                                             RX+
TX
      TX-                                             RX-                  RXD


                                                                  NOISE

 FIG 5A: RS 422 INTERFACE USING BALANCED LINES




                                COMMON
TX                                                                           RX
              COMMON GROUND



            FIG 5B : RS 423 INTERFACE USING SINGLE ENDED WIRE
            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     XIX - 3
 DTE                               DTE                                    DTE                   DTE
       TD                          TD                                           TD              TD

       RD                          RD                                           RD              RD

       RTS                         RTS                                          RTS             RTS

       CTS                         CTS                                          CTS             CTS

     DTR                           DTR                                        DTR               DTR

       DSR                         DSR                                          DSR             DSR

     GND                           GND                                        GND               GND

FIG 6 : RS 232 Connections with no handshaking
                                                               FIG 7 RS 232 C CONNECTIONS WITH HARDWARE
                                                                     HANDSHAKING




                                 CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                                          XIX - 4
 X-mitter
                                           Receiver




FIG 8 SOFTWARE HANDSHAKING USING X-ON AND X-OFF CHARACTERS




            CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                     XIX - 5
                Client layer




                 Data link layer

      Data                            Data
    encapsulation                  decapsulation



                                   Receive link
    Transmit link                  management
    management




                 Physical layer

        Data                          Data
      encoding                       decoding


      transmit                     Receive
    channel access                 channel access


                       cable



  Fig 9. Ethernet Defence model

CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                         XIX - 6
Preamble Destination Source       Type      Data field                   Frame
         Address     address      field                                  check
                                                                         sequence


         FIG 10. Ethernet Frame Format




          CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                   XIX - 7
                                      Controller board


                                                          o

                                                                         connector


             Data encapsulation
Data link      / decapsulation
layer                                                                                    Transceiver
                                                                                         cable
             Link management

                                                                     Connector
 Physical   Data encoding/decoding
 layer


                                                                                  Transceiver
              Channel Access
                                                                                           Tap

                                                                                 Coaxial cable

                 Fig 11. Typical Ethernet Implementation




                  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                           XIX - 8
                          1                0               0                1           1
0 volts




2.5 volts


                                       Fig 12. Manchester Encoding




               Bus                      Remote                                           Remote
            Controller                 Terminal                  Monitor                Terminal
              BC                         RT                        M                      RT




                                                   Remote
                                                  Terminal
                                                    RT


                              Fig 13. 1553 Bus structure



                         CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                                  XIX - 9
                             1 bit time


       1 MHz (+)
       Clock (0)



       NRZ (+)
       Data (0)



Manchester II (+)    1           0           1            1          0            0     0
Bi-phase level (0)
               (+)




                                       Fig 14 1553 Data encoding




                         CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                                 XIX - 10
                              8 Wire data bus
                              Attention
                             Ready for data
                             Data accept
                             Data available
                           Service Request
                             Interface clear
                             Remote enable

                             End or Identify




Fig 15 : IEEE – 488 (HP-IB / GPIB ) Bus Configuration




       CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                XIX - 11
                GPIB Listener            GPIB Listener                       GPIB Listener       GPIB Listener
                and / or Talker          and / or Talker                     and / or Talker     and / or Talker



          DAV
         NRFD
        N DAC   Handshake Bus




          IFC
         ATN
         SRQ       Management Bus
         REN
         EOI




DIO 1


                                        Data Bus

DIO 8



                                    FIG 16 : GPIB Instrumentation Bus Structure


                                  CEP Course on Radar EW held at DLRL on 25 Oct to 29 Oct 2004
                                                          XIX - 12

								
To top