Docstoc

40120130405008

Document Sample
40120130405008 Powered By Docstoc
					International Journal of Electronics and Communication Engineering & Technology (IJECET), ISSN 0976 –
         INTERNATIONAL JOURNAL OF ELECTRONICS AND
6464(Print), ISSN 0976 – 6472(Online) Volume 4, Issue 5, September – October (2013), © IAEME
 COMMUNICATION ENGINEERING & TECHNOLOGY (IJECET)

ISSN 0976 – 6464(Print)
ISSN 0976 – 6472(Online)
                                                                             IJECET
Volume 4, Issue 5, September – October, 2013, pp. 80-89
© IAEME: www.iaeme.com/ijecet.asp                                           ©IAEME
Journal Impact Factor (2013): 5.8896 (Calculated by GISI)
www.jifactor.com




 TO ANALYZE THE PERFORMANCE OF VARIOUS DIGITAL FILTERS IN
       OCDMA MULTI-USER ENVIRONMENT WITH 3D CODES

                     Pankaj Sharma1, Sandeep Kaushal2, Anurag Sharma3
           1, 2
             Department of E.C.E, Amritsar College of Engineering & Technology,
                                   Amritsar Punjab, India
       3
        Department of E.C.E, CT Institute of Engineering, Management & Technology,
                                  Jalandhar, Punjab, India



ABSTRACT

        To achieve the high speed connectivity for access networks, the combination of the large
bandwidth of the fiber medium with the flexibility of the Code Division Multiple Access
(CDMA) technique is used and referred as Optical Code Division Multiple Access(OCDMA). This
paper presents the simulation results for 24 user OCDMA environment system with the help of
various system parameters like Bit Error Rate(BER), Q- factor and eye pattern with Raised Cosine
filter, Gaussian filter, Fabry Perot filter, Trapezoidal filter and Lorentzian filter. The proffered
network concede high number of users to communicate with high data rate over the conventional
OCDMA system. The analysis of simulation assures that the system frame with Fabry Perot filter
method is of minimum distortion, while maintaining BER 6.81×e-20 for the correctly decoded signal.

Keywords-OCDMA, BER, Wavelength Division Multiplex, Signal to Noise Ratio

 I.   INTRODUCTION

        OCDMA technology is a better option and achievable method to the general
traditional time and wavelength based multiple access methods for the high speed fiber optics.
OCDMA system became popular due to the availability of excess bandwidth in the fiber optic
medium [2]. In OCDMA system multiple users can transmit there data on a single optical channel by
various methods. Each user data is spread by a unique coding sequence, which are uni-polar
{0,1} sequence called optical orthogonal codes (OOC).These optical orthogonal codes are of
various types likewise 2-D,3-D [3]. At the receiver side combination of all the users data are received
over a single optical channel and each user can recognize its respective data by correlating the

                                                  80
International Journal of Electronics and Communication Engineering & Technology (IJECET), ISSN 0976 –
6464(Print), ISSN 0976 – 6472(Online) Volume 4, Issue 5, September – October (2013), © IAEME

received signal by specific transmitted spreading sequence, and then detecting the transmitted data
by optical receiver. As in conventional CDMA system is assigned a unique spreading code that
enables the user to distinguish his signal from that of the other users. In OCDMA the typical
modulation scheme used is On-Off Keying (OOK) and as a result, the spreading codes are binary
symbols in {0, 1} [1, 2]. Traditionally, as in the case of wireless communication, the spreading has
been carried out in time and we will refer to this class of OOC as one-dimensional OOCs (1-D
OOCs) [2, 3]. One drawback of one-dimensional (1-D) OOCs is the requirement of a large chip rate
[3]. For understanding OCDMA is a simple way, let say a brief case containing some
important document which is locked by a specific code. The briefcase can be opened only if the
code is known. Otherwise data cannot be retrieve this is called code division multiple access
(CDMA) and implementing this technique on optical fiber is called OCDMA. It has been
observed that the simulation setup by M.I Anis [5] uses two filters which are Bessel filter and
chebyshev filter to achieve the desirable results whereas our proposed work includes five filters and
endeavors has been made to get better results with minimum distortion. OCDMA has the advantage
of using optical processing to perform certain network applications such as addressing and routing
without reinstate to complicated multiplexers or demultiplexers or combiner and splitter [7]. The
asynchronous triple play transmission can simplify network development, sustain and overall control
[1,7]. Therefore, OCDMA is an attractive candidate for Local Area Network application.
Particularly, OCDMA can provide a secure network connection providing dynamic encoding.
For achieving multirate in OCDMA system many design schemes are used, most commonly used
techniques are by varying the length of the optical orthogonal codes (OOCs) sequence and power
controller. Wavelength Division Multiple Access (WDMA) is a channel access method based on
wavelength division multiplexing (WDM).In WDMA method data are transmitted via a stream of
signals of different wavelength. WDMA uses multiple lasers and transmit multiple wavelength of
light (λc) simultaneously over a single optical fiber. Each signal travels within its specific color band
which is modulated with the data on the transmitter side In traditional WDM system Erbium Doped
Fiber Amplifier(EDFA) are used which are very expensive, for reducing the cost of the system
powerful broadband light sources may be used at the transmitter side. This multiwavelength light
could be distributed by fiber to all the nodes to encode with the data. To further reduce the cost and
improve performance of the system multiwavelength fiber laser should be used that can generate
large no of wavelength. Micro wave optical code division multiple access (MW-OCDMA) efficiently
utilizes the available band- width of optical fiber, in a manner similar to WDM. Codes for MW-
OCDMA have to be two-dimensional (2D) to accommodate wavelength and time or three-
dimensional (3D) to accommodate space, wavelength and time [4-6]. 2D codes based on primes,
Reed-Solomon Codes have been reported such as Temporal/ Spatial Addition Modulo LT , where
LT is the temporal length, Multiple Pulse per Row (MPR), Wavelength - Time Spreading Codes
[1,2,5] to name a few. A 3D design on Space/ Wavelength/ Time based on primes has been
concluded. The present design is based on a novel algorithm having no cross-correlation among
wave- lengths allocated to users having the same BIBD code and Space allocation based on Steiner
Triple Systems (STS). The present design can be used with any One-Dimensional (1D) Optical
Orthogonal Code having ‘λc = 1’. STS and Kirkman designs can also be used by choosing only
those triples which give ‘λc = 1’ [5, 6].
        The paper has been organized in such a way that for achieving desirable results, the
simulation setup of OCDMA network is outlined in section 2 and in the section 3 of this paper the
focus is on the performance analysis of setup. In section 4 the observed results, eye pattern diagrams
and result table are discussed. The section 5 concludes the salient observation about this work.



                                                   81
International Journal of Electronics and Communication Engineering & Technology (IJECET), ISSN 0976 –
6464(Print), ISSN 0976 – 6472(Online) Volume 4, Issue 5, September – October (2013), © IAEME

II.    PROPOSED DESIGN

        The setup of 24 users is here achieved by using six different wavelengths from third optical
window 1550 with channel spacing of 0.4 nm (1550nm-1552 nm) and on every wavelength four
different OOC codes are employing by using various codes generator. Continuous Wave (CW) laser
produces one or more optical signal outputs commonly used with external modulator to encode the
binary data signal upon CW laser source .CW laser is characterized by its power ,wavelength , line
width and phase. Optical WDM multiplexer (combiner) accepts multiple optical signals at its
input and generate a single stream which includes all the input WDM optical signals. Filter
performs the selection of the desired signal or wavelength. A signal whose filter output peak power
is not equal to the user specified range drop threshold will not be allowed to pass through the filter.
The ideal filter response is unity.
        There are many types of digital Filters like Fabry Perot, Gaussian, Raised Cosine,
Lorentzian, and Trapezoid. Carrier shifting is also performed here which allows to shift the carrier
frequency of the optical signal output from the filter, when the signal is filtered in multiple channel
representation, each signal is represented by its own frequency band centered about a mean
carrier frequency, and for setting the frequency this model is used. Amplifier model is used to
boost the signal intensity due to long distance single mode fiber. Figure 1 shows the simulation setup
for 24 users OCDMA network.




                        Figure 1. Simulation setup for 24 users OCDMA

       Fiber link is a topology defined as a number of fiber spans followed by optical amplifier user
can decide the total length of the link by changing the number of spans. This can be used in the
topology where specific level of optical attenuation is desired. There are various kind of codes can be
generated through specific code generators and modulators help of generate a continuous laser beam
accordingly. Amplifiers in optical link are used to achieve the compensation of the dispersion effect.

III. PERFORMANCE ANALYSIS

      For the transmission of data at transmitter side first of all data are applied to a binary wave
form generator which converts the binary data {0, 1} to the binary waveform. This signal is then

                                                  82
International Journal of Electronics and Communication Engineering & Technology (IJECET), ISSN 0976 –
6464(Print), ISSN 0976 – 6472(Online) Volume 4, Issue 5, September – October (2013), © IAEME

modulated with some WDM optical signal which is generated by multiplexing multiple continuous
wave laser input and generating the final output signal which include the entire input WDM optical
signal. The modulated output fed to the filter section where the wavelength signal whose filtered
peak power does not exceed or meet the user defined drop is not allow being passing through it.
The output of filter set as the carrier shifting model whose main purpose is to reset the carrier
frequency of the signals which were changed during the transmission setup and then amplified to
specific level this whole process is called encoding of given signal. Output of amplifier now enters
the 24/1 multiplexer which inhales inputs of 24 different users and transmits single multiplexed
output through single stream over long distance single mode fiber.
         In the filter section we change the filter types to different modes like Raised cosine,
Gaussian, Lorentzian, Trapezoidal and Fabry Perot for observing the various effects and results.
         The signal is now enters to the fiber link in which span and the length of the fiber is already
defined for particular system. This link is consisting of single mode fiber and amplifiers. The
output of fiber enters the optical splitter 1/24 which divides the one s ignal stream to 24 data
signals. Output of the optical splitter again enters an optical splitter which splits the signals to the
no of wavelength generated by continuous wave laser. For receiving the desired signal the
splitter's output enters the filter model which filter out the signal whose peak power exceed the
user specified drop. Filtered output is again passed through the carrier shifting model which sets
the carrier frequency of the signals this whole procedure is called decoding. Finally the signal is
now multiplex to 4/1 and then attenuated to the specified level of attenuation and received on to
receiver model.

IV.   SIMULATION RESULTS

        The simulation work has been carried out using OPTSIM where 24 user, refer figure, have
been setup with 6 different wavelengths for 4 different optical orthogonal codes which generate
overall input spectrum. The various wavelengths are taken from optical window 1550 nm at channel
spacing of 0.4 nm (1550.0 nm - 1552.0 nm). The value of input and output response in terms of BER
and Q factor is varying for different types of filters. Figure 2 shows graphical representation of input
signal for the parameters power in watts vs. wavelength in meters, whereas it shows the spectrum of
six wavelengths.




                           Figure 2. Wavelength spectrum (input signal)

                                                  83
International Journal of Electronics and Communication Engineering & Technology (IJECET), ISSN 0976 –
6464(Print), ISSN 0976 – 6472(Online) Volume 4, Issue 5, September – October (2013), © IAEME

       The total average of power received after transmission through the single mode optical fiber
and the decoder is different for the different filters. For the Gaussian filter the BER for the receiver 1
is 2.247×e-18 and for the user 24 it 1.09×e-17. The value of Q2dB is 18dB for user 1 and 17 dB for
user 24. Figure (3-17) show graphical representation of various eye patterns through plot of signal in
watts vs. time in seconds. Figure 3 represents the eye diagram at transmission side. Figure 4
represents the eye diagram at user 1 and Figure 5 represents the eye diagram at user 24 at receiver
side.




                          Figure 3. Gaussian filter response (transmission side)




                       Figure 4. Gaussian filter response at user 1 (Receiver Side)




                      Figure 5. Gaussian filter response at user 24 (Receiver Side)

                                                   84
International Journal of Electronics and Communication Engineering & Technology (IJECET), ISSN 0976 –
6464(Print), ISSN 0976 – 6472(Online) Volume 4, Issue 5, September – October (2013), © IAEME

       After implementing the Fabry Perot filter in same OCDMA setup the values of responses are
changes immediately. The BER for user 24 is 4.69×e-18 and for user 1 is 6.81×e-20. The quality factor
value is lies between 18dB-19dB. Figure 6 represents the eye diagram of Fabry Perot filter at
transmission side. Figure 7 represents the eye diagram at user 1 and Figure 8 represents the eye
diagram at user 24 at receiver side.




                       Figure 6. Fabry Perot filter response (transmission side)




                     Figure 7. Fabry Perot filter response at node 1 (Receive side)




                     Figure 8. Fabry Perot filter response at node 24 (Receiver side)

                                                   85
International Journal of Electronics and Communication Engineering & Technology (IJECET), ISSN 0976 –
6464(Print), ISSN 0976 – 6472(Online) Volume 4, Issue 5, September – October (2013), © IAEME

       The setup response after implementing raised cosine filter is changed to 1.14×e-17 BER for
user 1 and 2.75×e-15 BER for user 24. The values of Quality Factor (Q2dB) are lies between the
17dB-18dB. Figure 9 represents the eye diagram at transmission side. Figure 10 represents the eye
diagram at user 1 and Figure 11 represents the eye diagram at user 24 at receiver side.




                       Figure 9. Raised Cosine filter response (transmission side)




                   Figure 10. Raised Cosine filter response at node 1 (Receiver side)




                   Figure 11. Raised Cosine filter response at node 24 (Receiver side)



                                                   86
International Journal of Electronics and Communication Engineering & Technology (IJECET), ISSN 0976 –
6464(Print), ISSN 0976 – 6472(Online) Volume 4, Issue 5, September – October (2013), © IAEME

T       he value of BER is 2.84×e-14 for user 24 and 1.90×e-16 for user 1 after implementing the
Trapezoidal filter in OCDMA setup which shows a statically change. The value of quality factor is
18dB for user 1 and 17 dB for user 24. Figure 12 shows the eye pattern for Trapezoidal filter at
transmission side. Figure 13 and Figure 14 shows the eye pattern for Trapezoidal filter at receiver
side for user 1 and user 24 respectively.




                       Figure 12. Trapezoidal filter response (transmission side)




                    Figure 13. Trapezoidal filter response at node 1 (Receiver side)




                    Figure 14. Trapezoidal filter response at node 24 (Receiver side)

                                                   87
International Journal of Electronics and Communication Engineering & Technology (IJECET), ISSN 0976 –
6464(Print), ISSN 0976 – 6472(Online) Volume 4, Issue 5, September – October (2013), © IAEME

        The setup response after implementing Lorentzian filter is changed to 2.11×e-5 BER for user
24 and 8.37×e-7 BER for user 1 .The values of Q2dB are lies between the 12dB-13dB. The response
of this filter is very low compare to all other. Figure 15 represents the transmission side eye pattern.
Figure 16 and Figure 17 for the user 1 and user 24 respectively at receiver side.




                        Figure 15. Lorentzian filter response (transmission side)




                     Figure 16. Lorentzian filter response at node 1(Receiver side)




                    Figure 17. Lorentzian filter response at node 24 (Receiver side)

                                                   88
International Journal of Electronics and Communication Engineering & Technology (IJECET), ISSN 0976 –
6464(Print), ISSN 0976 – 6472(Online) Volume 4, Issue 5, September – October (2013), © IAEME

       The Results carried out by the simulation setup is represented in Table 1. In this paper we
observed that by changing the filter in setup of OCDMA the values of BER and the quality factor
changes and effect of same can be observed with the help of various eye pattern diagrams.

                           Table 1: Results of the OCDMA 24 user’s network
     S.        Type of Filter Used In                 BER                    Q2dB
     No.               Setup                User 1        User 24     User 1      User 24
      1           Gaussian Filter         2.247×e-18      1.09×e-17   18 dB        17 dB
      2         Raised Cosine Filter       1.14×e-17      2.75×e-15   18 dB        17dB
                                                  -20            -18
      3          Fabry Perot Filter        6.81×e         4.69×e      19 dB        18dB
      4          Trapezoidal Filter        1.90×e-16      2.84×e-14   18 dB        17dB
                                                   -7             -5
      5          Lorentzian Filter         8.37×e         2.11×e       13dB        12 dB

V.      CONCLUSION

       This paper focuses on the OCDMA network simulation module and setup design of its end
nodes with 24 users. Here, results obtained from the simulation setup by using different filters like
Gaussian, Fabry Perot, Raised Cosine, Trapezoidal and Lorentzian by considering BER, Eye Pattern
and Quality factor. It has been concluded that the simulation result of the Fabry Perot filter is with
minimum distortion a n d e r r o r whereas the Lorentzian filter gives most distorted output. Hence
Fabry Perot filter e n v i r o n m e n t i s t h e most e f f e c t i v e e f f o r t f o r t h e evaluation of the
performance of OCDMA system. Here the work concludes that the data which are transmitted at
the transmitter side with various codes through this method can be received successfully with
minimum error ratio at the receiver side.

REFERENCES

  [1]      Fredrick Vanhaverbeke and Marc Moeneclaey, “Sum Capacity of equal-power users in overloaded
           channels” IEEE Trans. Commun., Vol. 3,No. 2,pp 228-233,Feb 2005
  [2]      Sharon Goldberg, Varghese Baby, Ting Wang and Paul R “Source-Matched Spreading Codes for
           Optical CDMA” IEEE transactions on communication, vol.55,no. 5.pp. 850-853,May2007.
  [3]      Mahdi Karimi and M. Nasiri-Kenari, “An internally coded TH/OCDMA scheme for fiber optics
           communication system and its performance analysis- part 1: using optical orthogonal codes” IEEE
           transactions on communication. Vol. 55, no. 2, pp333-344, Feb. 2007.
  [4]      M. Ravi Kumar et. Al ,“ A New Multi Wavelenght - Optical code division multiple access code
           design based on balanced incomplete block design” IEEE transactions on communication vol.55 no.
           2, Feb 2007.
  [5]      M.Irfan Anis, Naveed Ahmed and Saifuddin, “Design and performance analysis of OCDMA system
           using different filters” IEEE transactions on communication, Feb 2009
  [6]      Ivan Glesk et.al, “Evaluation of OCDMA system deployed over commercial network Infrastructure”
           IEEE Tu.C5.4 ICTON, pp1-4, April 2011
  [7]      Parambir Singh, Manoj Kumar, Anurag Sharma, “Design and Performance investigation of multiuser
           OCDMA network” IJSER, Vol. 4,Issue 7,pp 2549-2552,July 2013.
  [8]      Faîçal Baklouti and Rabah Attia, “Numerical Suppression of Linear Effects in an Optical CDMA
           Transmission”, International Journal of Electronics and Communication Engineering & Technology
           (IJECET), Volume 3, Issue 3, 2012, pp. 112- 121, ISSN Print: 0976- 6464, ISSN Online: 0976 –6472.
  [9]      J.Ravindrababu and Dr.E.V.Krishna Rao, “Performance Analysis and Comparison of Linear Multi-
           User Detectors in DS-CDMA System”, International Journal of Electronics and Communication
           Engineering & Technology (IJECET), Volume 3, Issue 1, 2012, pp. 229 - 243, ISSN Print:
           0976- 6464, ISSN Online: 0976 –6472.

                                                        89

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:0
posted:10/24/2013
language:Latin
pages:10