Docstoc

VoIP Test & Measurement

Document Sample
VoIP Test & Measurement Powered By Docstoc
					VoIP Test & Measurement

        Why Test?
               VoIP Test & Measurement – Why Test?
               VoIP Test & Measurement – Why Test?



      Testing is Necessary
Competitive market requires high quality
– Retain customers
– Gain and maintain good reputation
Quality assurance, fewer truck rolls
Network reliability
– MTBF - Increase
– MTTR - Decrease
             VoIP Test & Measurement – Why Test?
             VoIP Test & Measurement – Why Test?



Testing Helps Retain Customers
Recruiting customers is expensive
Long-term retention is critical to ROI
Loss of voice customer may also mean loss
of the rest of the triple-play revenue
                VoIP Test & Measurement – Why Test?
                VoIP Test & Measurement – Why Test?


Potential Customer Perception
          Challenges
Confusion about unfamiliar technology
Concerns about emergency services
Fear that quality won’t be good
– Remembering experiences with immature
  technology
– Have experienced effects of jitter, delay and
  packet loss personally
              VoIP Test & Measurement – Why Test?
              VoIP Test & Measurement – Why Test?



         What is Quality?
Service quality must be at least as
good as PSTN (Public Switched
Telephone Network)
Speech quality must be
comparable to industry standards
Primary Line Reliability –
Consistent continuous testing
throughout the network
              VoIP Test & Measurement – Why Test?
              VoIP Test & Measurement – Why Test?



  Network Impact on Quality
Same physical network as HSD (High
Speed Data), with impressionable return
band
CPD (Common Path Distortion), impulse
noise, signal to noise, laser clipping,
response issues, etc.
Traditional testing must be stepped up a
notch
                 VoIP Test & Measurement – Why Test?
                 VoIP Test & Measurement – Why Test?



     Testing Finds Problems
Noise, distortion and interference still cause
problems
Causes data slowdowns, that in many cases
would go unnoticed by an HSD subscriber
– Retransmission of lost packets
– A temporary slowdown of data transmission may be
  unnoticed or accepted.
Poor installation craftsmanship is #1
reason for callbacks
                VoIP Test & Measurement – Why Test?
                VoIP Test & Measurement – Why Test?



               Truck Rolls
Truck roll costs >$130             Number of Tech Return Visits
                                        Innumerable
                                            2%
(Yankee Group)                          4 or more
                                           3%


34% of VoIP                         2 or 3
                                     11%


customers need on-                 1
                                  18%
site support (Decipher                                       0
                                                            66%
survey)
                 VoIP Test & Measurement – Why Test?
                 VoIP Test & Measurement – Why Test?



  Testing = Nervous System
Leprosy symptom: loss of sense of touch
– Results in not noticing injuries, not attending to
  infections, loss of affected tissue
Network health requires a
sense of touch
– Notice problems
– Find the causes
– Fix quickly
                              http://users.rcn.com/jkimball.ma.ultranet/BiologyPages/A/
                VoIP Test & Measurement – Why Test?
                VoIP Test & Measurement – Why Test?



General Reliability Objectives
Maximize MTBF (Mean Time Between
Failures)
– Tests reveal sources of common failures
– Results in natural focus of attention in those
  areas
Minimize MTTR (Mean Time To Repair)
– Tests help to locate problem source quickly
– Improves productivity
             VoIP Test & Measurement – Why Test?
             VoIP Test & Measurement – Why Test?



         Testing Helps…
To monitor and maintain quality
To retain customers and reputation
To maintain awareness of network health
To increase reliability
VoIP Test & Measurement

 Network Preparation & Maintenance
                 VoIP Test & Measurement – Why Test?
                 VoIP Test & Measurement – Why Test?



Enhanced Testing and Maintenance
 Before the service is rolled out
 – Align and test each node to assure operating
   margin for optimal quality
 Adequate digital video and HSD
 performance does not guarantee adequate
 VoIP performance
 – these services are relatively robust
 – able to withstand interference that causes
   packet loss and jitter
                  VoIP Test & Measurement – Why Test?
                  VoIP Test & Measurement – Why Test?


    Alignment and Maintenance
            Required
In order to provide the highest quality service and
mitigate service calls
The return path should be swept and balanced
Verify carrier-to-noise ratio operating margin
Ingress and impulse noise must be monitored for
at least a 24-hour period
– occurs at various times of the day
– has a wide variety of causes
               VoIP Test & Measurement – Why Test?
               VoIP Test & Measurement – Why Test?



           Return Sweep
Proper alignment of the system is critical to
service quality and reliability
Problems uncovered while sweeping can
also cause ingress
– Bad, or loose connections
– Cable damage
– Misalignment (can cause
  laser clipping)
               VoIP Test & Measurement – Why Test?
               VoIP Test & Measurement – Why Test?



   Installation RSVP Testing
Separate hand-held tester or optioned
installation analyzer
Verifies return path loss
– Generates return path test signal
– Ranges to determine appropriate launch level
– Compares with limit for pass / fail result
Tests return band noise/ingress
              VoIP Test & Measurement – Why Test?
              VoIP Test & Measurement – Why Test?



         Linear Distortions
Upstream measurements are more relevant
Generate 16 QAM carrier and have
someone read (analyze) it in the headend
Analyze the pre-equalization information
returned from the CMTS (DOCSIS 2.0) to
obtain in-channel frequency response,
micro-reflections, group delay and equalizer
stress
              VoIP Test & Measurement – Why Test?
              VoIP Test & Measurement – Why Test?



          Ingress Causes
Loose or corroded (probably old) F-
connectors
Loose hard-line connectors and splices
Loose faceplates in feeder passives (taps,
line couplers and splitters)
Bad drops
                VoIP Test & Measurement – Why Test?
                VoIP Test & Measurement – Why Test?



         Ingress Mitigation
Leakage monitoring and maintenance –
where there’s a leak there’s probably
ingress
Train installers and test
 installations
– Tight connections with good
  connectors
– Verify drop integrity with leakage
  detector
              VoIP Test & Measurement – Why Test?
              VoIP Test & Measurement – Why Test?



   Drop Passive Intervention
Return band step
attenuation for low
value taps
Return band traps for
basic-only subs?
                  VoIP Test & Measurement – Why Test?
                  VoIP Test & Measurement – Why Test?



     Ingress Troubleshooting
I-Stop Probe
– Taps Unused KS Ports for
  Bi-Directional Measurements
– Compatible with SLMs
  Reverse and Forward Path
  analyzers
– Protects Instruments From Damage Caused By Line
  Power
– Locate ingress sources without removing reverse
  modules or diplexers and without disrupting forward or
  reverse service
              VoIP Test & Measurement – Why Test?
              VoIP Test & Measurement – Why Test?



     Using the I-Stop Probe
Screw the probe into a KS port on the
distribution tap
Connect the field
analyzer to the test
port
                      VoIP Test & Measurement – Why Test?
                      VoIP Test & Measurement – Why Test?



       Using the I-Stop Probe
Press the button on the
side of the probe
– If the ingress decreases by
  4-6 dB when the button is
  depressed, the source of
  the ingress is farther from
  the node than you are
– Ingress that doesn't
  decrease is entering the
  system nearer to the node
  than you are.
The I-Stop Probe has little
or no visible effect on
forward path signals.
                                Headend spectrum display.
               VoIP Test & Measurement – Why Test?
               VoIP Test & Measurement – Why Test?



    Return Band Monitoring
Addition of VoIP service requires better
hygiene
Constant monitoring means an automated
system is essential
SNMP trap management with controllable
limits, alarms and selective notification is
commonly used
               VoIP Test & Measurement – Why Test?
               VoIP Test & Measurement – Why Test?



    Monitoring System Value
Useful before and after initiating service
– Used to prepare the node for certification
– Then to actually certify the node
– Then to maintain the node and as an aid to
  locate problems before they impact the
  subscriber
Goal: >40dB “carrier-to-crud” ratio
                VoIP Test & Measurement – Why Test?
                VoIP Test & Measurement – Why Test?



 Monitoring System Selection
Seeing all forms of ingress is primary
concern
– relatively stable forms of CPD
  and ingress
– very transient ingress and
  impulse noise
Monitoring analyzer must be very fast
Must be able to discern interfering signals
within band of legitimate system carriers
                VoIP Test & Measurement – Why Test?
                VoIP Test & Measurement – Why Test?



 Monitoring System Operation
Most monitoring systems use frequency
scanning return band analysis
– limits set to detect ingress and noise problems
When limit is crossed, “trap” or alarm is
triggered
Limits are configurable, and alarm is
scalable, to gauge response urgency
                      VoIP Test & Measurement – Why Test?
                      VoIP Test & Measurement – Why Test?



                Impulse Noise
Over-the-air electromagnetic interference (EMI)
Bursts of random noise
Originates from household appliances
–   hair dryers
–   Blenders
–   drills
–   microwave ovens
Dirty power-line insulators and fluorescent
lights can cause additional disturbances in
both the forward and return bands.
These signals enter the coax in the same way
conventional RF ingress does
              VoIP Test & Measurement – Why Test?
              VoIP Test & Measurement – Why Test?



               Catch a Transient
Impulse noise, and intermittent ingress can
be difficult to catch with conventional
analysis schemes
Looking at one node when ingress occurs
at another
Looking at one frequency, when ingress
occurs at another
               VoIP Test & Measurement – Why Test?
               VoIP Test & Measurement – Why Test?



          DSP Advantage
Catches transients that other
methods miss
80 Spectrum “snapshots” a second
“Sees” entire spectrum at once,
digitizes it, and converts it to the
frequency domain using fast
Fourier transform.
              VoIP Test & Measurement – Why Test?
              VoIP Test & Measurement – Why Test?



          DSP Example




Time Domain                       Frequency Domain
              VoIP Test & Measurement – Why Test?
              VoIP Test & Measurement – Why Test?



Monitoring System Speed Test
Compare 3 different monitoring system’s
ability to capture a 15 ms. pulse
VoIP Test & Measurement – Why Test?
VoIP Test & Measurement – Why Test?


           Monitoring System
             Speed Test
            DSP-based Monitoring System
                •   Note display shows both the
                    impulse at 24 MHz and its
                    harmonic at 48 mHz.



             Analog Scanning Analyzer System




             Analog Scanning Analyzer System
                VoIP Test & Measurement – Why Test?
                VoIP Test & Measurement – Why Test?



Occupied Spectrum Challenge
What about ingress at the same
frequency as existing (TDMA)
carriers?
Spectrum analyzer min-hold will
eliminate the existing carrier, but
takes time and will also minimize the
ingress; also subject to scanning
delay
TrafficControl mode, captures
transients that occur at occupied
channels.
                  VoIP Test & Measurement – Why Test?
                  VoIP Test & Measurement – Why Test?


        “Traffic Control” Mode
Captures interfering transient signals that appear
beneath legitimate carriers
Spectrum snapshot acquisition is not
synchronized with the TDMA signals, so
snapshots will be taken when data signal is not
present in the spectrum
– Without Traffic Control, appears to be present 100% of
  the time
– With the feature activated only snapshots taken with
  data carrier blanked appear on the display
– Assumes ingress level lower than expected carrier level
                    VoIP Test & Measurement – Why Test?
                    VoIP Test & Measurement – Why Test?



      NOC / Field Integration
In addition to monitoring the return
band, the instrument must integrate
with field analyzers
Provide a view of ingress spectrum
at the NOC for comparison with field
test point.
Provide response and level
information
– Distribution alignment
– Installation verification
             VoIP Test & Measurement – Why Test?
             VoIP Test & Measurement – Why Test?


Why Certify the Node?
A maintenance challenge
Increased use of return band increases
importance of reliability
– VOD
– High Speed Data
– VoIP
Return path certification process
verifies viability for services
             VoIP Test & Measurement – Why Test?
             VoIP Test & Measurement – Why Test?



What is Node Certification?
The return path is monitored for an
extended period for ingress or impulse
noise
– Extended period ensures that worst case
  time-of-day is included
– Node certification can be tied to a
  minimum acceptable percentage of
  unavailability
               VoIP Test & Measurement – Why Test?
               VoIP Test & Measurement – Why Test?


 Stepped Up Traditional Plant
    Maintenance Testing
Traditional plant maintenance testing is
essential to provide quality voice service
Sweeping
Leakage
Ingress analysis
Return path monitoring
VoIP Test & Measurement

     RTP and Ping Tests
                VoIP Test & Measurement – Why Test?
                VoIP Test & Measurement – Why Test?



VoIP Test Implementation Timing
Pre-Infrastructure
– Planning stage, no VoIP service equipment
  installed
– Node hardening, preparation, certification
– VoIP RTP tests (standard on field analyzer)
Infrastructure-Based
– Test system part of VoIP service infrastructure
– Active monitoring, scheduled testing, field
  troubleshooting and installation support
              VoIP Test & Measurement – Why Test?
              VoIP Test & Measurement – Why Test?



Pre-Infrastructure VoIP Testing
Prepares system for VoIP service offering
by pre-testing
RTP (Real-time Transport Protocol) tests
for packet loss, jitter, and delay, with MOS
Inexpensive solution, with low-cost headend
based server and standard 860 DSPi
firmware
               VoIP Test & Measurement – Why Test?
               VoIP Test & Measurement – Why Test?



    VoIP Installation Testing
Testing is critical to ensure quality in a very
competitive market – expectations must be
met in order to succeed
Testing provides:
– Spot check of parameters that impact quality
– Ensures headroom
– Sets baseline for trend analysis
– Enables troubleshooting
              VoIP Test & Measurement – Why Test?
              VoIP Test & Measurement – Why Test?



Pre-Infrastructure VoIP Test Process
  Test server installed in the NOC
  near the PSTN handoff point
  Call set up using proprietary
  signaling
  Communication takes place
  using RTP (Real-Time Transport
  Protocol) common to VoIP signaling
  protocols
             VoIP Test & Measurement – Why Test?
             VoIP Test & Measurement – Why Test?



             RTP Test
Communication established
Communication occurs using RTP over
UDP (User Datagram Protocol) at pre-
negotiated port
Server and analyzer
synchronize test parameters
and clocks
Results are calculated and compared to
limits – Pass/Fail results
  VoIP Test & Measurement – Why Test?
  VoIP Test & Measurement – Why Test?



Sample Results
               VoIP Test & Measurement – Why Test?
               VoIP Test & Measurement – Why Test?



Sample Results – Drill Down




  Downstream                          Upstream
              VoIP Test & Measurement – Why Test?
              VoIP Test & Measurement – Why Test?



 Mean Opinion Score (MOS)
Subjective voice quality score based on the
perception of a random group
of people listening to speech
over a communication system.
Group of males and females rate the quality
of test sentences read
Each person rates from 1 to 5
MOS is average: 1 (worst), 5 (best)
                         VoIP Test & Measurement – Why Test?
                         VoIP Test & Measurement – Why Test?


      Mean Opinion Score (MOS)
Rating   Definition Description

                   A perfect speech signal recorded in a quiet
  5      Excellent
                      booth
                   Intelligent and natural like long distance
  4       Good
                      telephone quality
                   Communication quality, but requires some
  3        Fair
                      hearing effort
                   Low quality and hard to understand the
  2        Poor
                      speech
  1        Bad      Unclear speech, breakdown
               VoIP Test & Measurement – Why Test?
               VoIP Test & Measurement – Why Test?



             MOS Reality
Top rating of 5.0 is unattainable in reality
Codec’s introduce some degree of quality
loss
– G.711 codec,
  max MOS achievable is 4.4
– G.729 codec,
  max MOS achievable is 4.2
                VoIP Test & Measurement – Why Test?
                VoIP Test & Measurement – Why Test?



               More MOS
“Toll quality” conversation requires an
MOS of at least 4.0.
Satisfactory conversations require an
MOS of at least 3.6
Here’s another way to look at it:
– 4.0 to 5.0 = desirable voice quality
– 3.6 to 4.0 = generally acceptable
– <3.6 = not recommended
                VoIP Test & Measurement – Why Test?
                VoIP Test & Measurement – Why Test?



    How MOS is Calculated
E-Model used to
determine R-factor
– Formula takes
  various impairment
  factors into account
R-factor used to
determine MOS                From “Voice Over IP Performance Monitoring,”
                             Cole, R. G. and Rosenbluth, J. H., AT&T
                             Laboritories, Middletown, NJ, Computer
                             Communication Review
              VoIP Test & Measurement – Why Test?
              VoIP Test & Measurement – Why Test?



MOS Test Waveform Samples
                      Packet Loss
  Codec    0%             1%        5%
           MOS=4.19      MOS=3.85   MOS=2.90
  G.711
           MOS=3.57      MOS=3.40   MOS=2.67
  G.723
           MOS=3.75      MOS=3.42   MOS=2.40
  G.729

                      Noise (WBN)
  Codec   29 dBrn       40 dBrn   44 dBrn
           MOS=4.08      MOS=3.39    MOS=2.94
  G.711
              VoIP Test & Measurement – Why Test?
              VoIP Test & Measurement – Why Test?



         Latency (Delay)
Causes Echo and Talker Overlap
When delay is > 50 msec, echo becomes a
problem; echo cancellation is required
Talker overlap is significant when one-way
delay is > 250 msec
                    VoIP Test & Measurement – Why Test?
                    VoIP Test & Measurement – Why Test?



              Latency (Delay)
Accumulation (or Algorithmic) Delay
– Codec dependent
– Time to collect a frame of voice samples for processing
Processing (or Encoding) Delay
–   Codec dependent
–   Processor execution time
–   Algorithm type
–   Concatenation – encapsulating small packets into a
    single larger frame
                VoIP Test & Measurement – Why Test?
                VoIP Test & Measurement – Why Test?



          Latency (Delay)
Network delay
– Caused by the physical medium and protocols
– Also affected by jitter buffer
Polling delay
– Headend polls customer locations
– Delay while voice packets wait for the next poll
– Equipment minimizes by anticipating next poll,
  and queuing up packets
               VoIP Test & Measurement – Why Test?
               VoIP Test & Measurement – Why Test?



          Latency (Delay)
Echo
– Present in POTS network, but acceptable with
  delay less than 50 msec
– In VoIP networks delay is always > 50 msec,
  requiring echo cancellation (digital filter)
              VoIP Test & Measurement – Why Test?
              VoIP Test & Measurement – Why Test?



         Latency (Delay)
Can seriously impair communication
Usually a by-product of switching and
routing
Must be less than 300 msec. round trip
                 VoIP Test & Measurement – Why Test?
                 VoIP Test & Measurement – Why Test?



                     Jitter
Packets arrive at destination out of timing or
sequence
Jitter buffer is used to enable re-ordering of
packets
Increasing size of jitter buffer threatens delay
Should be less than twice packetization rate
Jitter buffer overflow causes packet loss
                  VoIP Test & Measurement – Why Test?
                  VoIP Test & Measurement – Why Test?



                Packet Loss
Frames dropped under peak loads and
congestion
– Time-sensitive, uncorrectable by retransmission
Mitigation techniques:
– Interpolation – replaying the last packet received; not
  effective for bursty packet loss
– Redundancy – send redundant information; soaks up
  bandwidth and adds delay
– Voice coder – lower bandwidth coder used with
  redundancy; improves bandwidth usage, but not delay
              VoIP Test & Measurement – Why Test?
              VoIP Test & Measurement – Why Test?



            Packet Loss
Can be caused by network congestion, jitter
buffer overflow, or ingress
Random packet loss is less noticeable than
“bursty” packet loss
Target <1%, which is less
noticeable when loss is
random; >4% renders
service unusable
                      VoIP Test & Measurement – Why Test?
                      VoIP Test & Measurement – Why Test?



 Packet Loss Impacts MOS
Because of the real-
time nature of VoIP
traffic, very little
damage to the data
stream can be repaired
– 1% packet loss will
  noticeably degrade
  phone service
– 4% packet loss renders
  phone service unusable.
                  VoIP Test & Measurement – Why Test?
                  VoIP Test & Measurement – Why Test?



    Measurement Expectations
Test Parameter
Recommended
Results
           Packet Loss       Delay (round     Jitter
                             trip)
Maximum    <3%               <300 msec

Recommended <1%              <200 msec        2X
                                              packetization
Better     0.1%              <60 msec
VoIP Test & Measurement

     Network-based Testing
                    VoIP Test & Measurement – Why Test?
                    VoIP Test & Measurement – Why Test?


Infrastructure-Based Test System Overview
Test Management Server
 – MSO or regionally distributed
 – Provides centralized control,
    measurement database and analysis
Probe
 – RAN or headend
 – Initiates tests at various locations
    as directed by Test Management Server
Installation Responder
 – Home
 – Hand-held design for field use
                     VoIP Test & Measurement – Why Test?
                     VoIP Test & Measurement – Why Test?


Service Level Assurance System Characteristics
Integrated server-based test system provides continuous,
comprehensive view of VoIP network performance
Enables end-to-end testing initiated from the field test point
Automated subscriber service activation tests
Automated subscriber VoIP loop maintenance testing
Gateway probe for PSTN legacy trunk testing
Multi-service troubleshooting (Voice, Fax, Data, Video)
Automated – web based – future <Video over IP> proof
                     VoIP Test & Measurement – Why Test?
                     VoIP Test & Measurement – Why Test?


Core System Server Components
Test Plan Manager - manage and automate test plans
Test Plan Browser - view information about test plans
(properties, occurrence and test results).
Test Plan Generator - generate test plans with test
numbers, routes, trunks, perform load balancing between
destination test numbers and import from LCR (Least Cost
Routing) tables.
Decision Support System - generates web reports and
allows users to apply filters in relation to test plan, date &
time, countries, test origin and destinations.
QoS Analyzer and Gateway - generates and sends SLA
related service threshold violation notifications to the Fault
Management OSS or specific user pagers or cell phones.
                  VoIP Test & Measurement – Why Test?
                  VoIP Test & Measurement – Why Test?



   Core VoIP Test Objectives
Pro-actively Test end-to-end perceived quality
and IP QoS with intrusive Probes
– Echo, Delay, Noise, RTP Statistics, IP Latency, etc.
Aggregate results into Global Quality Indicators
– MOS, R-factor, E-model Reports
Automate Service Level Alarms generation
– SNMP Traps based on critical threshold violations
  should be sent to centralized Fault Management
  System
                     VoIP Test & Measurement – Why Test?
                     VoIP Test & Measurement – Why Test?



   Test Call Configuration Steps
1. Add destination phone number to Test Server
   database if necessary
2. Add additional network tones, such as Ring,
   Busy, Dial, etc., to Test Server database if
   necessary
3. Construct Test Plan
  •   Test Type, Origin port, Destination port / phone
      number, Carrier ID and number of test calls.
4. Generate Test Plan
5. Run Test Plan
                           VoIP Test & Measurement – Why Test?
                           VoIP Test & Measurement – Why Test?



         Test Plan Scheduling Steps
1.       From Schedule Executions window, configure schedule
         parameters: time (start now, start on
         month/day/year/time, every day/week/month) and Run
         Times
2.       Optional parameters
     –     email notification of test plan completion,
     –     email of exported test results upon test plan completion
     –     real-time SLA parameter evaluation (Critical, Major, Minor,
           Warning) according to customer requirements (e.g.: Echo,
           Noise, Delay, MOS, etc)
     –     real-time SNMP alarm forwarding to external Fault Management
           System
3.       Generate Schedule – schedule is executed
                 VoIP Test & Measurement – Why Test?
                 VoIP Test & Measurement – Why Test?



   Test Results and Analysis
Test results are stored in test server database
after each test call is completed, and can be
viewed in real-time
Results can be accessed using a web browser,
and can be exported (.csv format)
Standard report templates, multiple views
(Summary, Detailed and Test Call View), as well
as Service Level Classes (QoS Metric thresholds)
allow the user to display results as per their needs
No manual calculation of any QoS Metrics is
required
      VoIP Test & Measurement – Why Test?
      VoIP Test & Measurement – Why Test?



Sample Results Screen
      VoIP Test & Measurement – Why Test?
      VoIP Test & Measurement – Why Test?



Sample Results Screen
             VoIP Test & Measurement – Why Test?
             VoIP Test & Measurement – Why Test?



    Measurement Systems
Active vs. passive systems
Installation Support
Loopback tests
Trunk testing
RTP streaming / monitoring
Network segmentation
Transponder loopback tests
                     VoIP Test & Measurement – Why Test?
                     VoIP Test & Measurement – Why Test?



       VoIP Service Active Monitoring
       Hand-




PP30
        held




                                                             DQ Web GUI
          E100

Phone




                         E100
                 PP500                                            DQ R7
 Fax                                                              Server

                                                                  LAN
 PC

                                                     T1

 TV                                                 E100 PP6000
                           VoIP Test & Measurement – Why Test?
                           VoIP Test & Measurement – Why Test?


                                                Day-Of-Install
                  Test Setup, Results,
                 Analysis all on DSP 860
                                                   Tests
       Hand-




PP30
        held




                  Screen and Keyboard                              DQ Web GUI
          E100

Phone




                               E100
                       PP500                                            DQ R7
 Fax                                                                    Server

                                                                        LAN
 PC

                                                           T1

 TV                                                       E100 PP6000
                            VoIP Test & Measurement – Why Test?
                            VoIP Test & Measurement – Why Test?


               RTP / Codec MTA Loopback
                         Tests
       Hand-
        held




PP30


                                                                    DQ Web GUI
          E100

Phone




                                E100
                        PP500                                            DQ R7
 Fax                                                                     Server

                                                                         LAN
 PC

                                                            T1

                                                           E100 PP6000
                   Loopback Voice Test
 TV



                  -Continuity Test (108)
                   -Speech Test (PESQ)
                    -Delay + DTMF Test
                           VoIP Test & Measurement – Why Test?
                           VoIP Test & Measurement – Why Test?

                 Legacy “Bellcore” Trunk Testing
       Hand-




PP30
        held




                                                                   DQ Web GUI
          E100

Phone




                               E100
                       PP500                                            DQ R7
 Fax                                                                    Server

                                                                        LAN
 PC

                                                           T1

 TV                                                       E100 PP6000




                                                   PSTN Trunk
                                                  Testing (100,
                                                 102, 105 & 108)
                               VoIP Test & Measurement – Why Test?
                               VoIP Test & Measurement – Why Test?



               RTP Streaming / Monitoring
       Hand-




PP30
        held




                                                                       DQ Web GUI
          E100

Phone




                                   E100
                           PP500                                            DQ R7
 Fax                                                                        Server

                                                                            LAN
 PC

                                                               T1

 TV                                                           E100 PP6000


                   RTP Streaming Tests
                 -Packet Loss, Jitter, Delay
                     -R Factor Analysis
                      VoIP Test & Measurement – Why Test?
                      VoIP Test & Measurement – Why Test?



           Segmenting to/from the Hubs
       Hand-




PP30
        held




                                                              DQ Web GUI
          E100

Phone




                          E100
                  PP500                                            DQ R7
 Fax                                                               Server

                                                                   LAN
 PC

                                                      T1

 TV                                                  E100 PP6000


       Troubleshooting from the HUB
Call Quality, Speech, Echo, Delay, Noise, Fax
                          VoIP Test & Measurement – Why Test?
                          VoIP Test & Measurement – Why Test?


                 Transponder Loopback Tests
       Hand-
        held




PP30


                                                                  DQ Web GUI
          E100

Phone




                              E100
                      PP500                                            DQ R7
 Fax                                                                   Server

                                                                       LAN
 PC

                                                          T1

                                                         E100 PP6000
                  Loopback RTP Test
 TV



                   -Continuity Test
                 -Speech Test (PESQ)
                 -Delay + DTMF Test
                           VoIP Test & Measurement – Why Test?
                           VoIP Test & Measurement – Why Test?



Voice & Call Quality Measurements
Call Connectivity
 – Post Dial Delay
 – Call Completion Ratio
 – Call Connection Cause Values
Speech Quality
 –   Mean Opinion Score
 –   Echo
 –   Noise
 –   Delay
 –   Distortion and Clipping
 –   DTMF Transparency
 –   Loss
Modem/Fax Quality
VoIP Transmission Quality
Additional Metrics
             VoIP Test & Measurement – Why Test?
             VoIP Test & Measurement – Why Test?


    Measurement System
         Overview
Network-based VoIP test system
Management server
Deployment of probes
Tests performed
VoIP Test & Measurement

      Installation Testing
                VoIP Test & Measurement – Why Test?
                VoIP Test & Measurement – Why Test?



       Installation Objective
Take advantage of subscriber visit to make
sure everything is in good shape
– Ensure that the home is properly wired
– Avoid a possible service truck roll in the future
– Keep the potential for ingress from that home to
  an absolute minimum
                  VoIP Test & Measurement – Why Test?
                  VoIP Test & Measurement – Why Test?




Installation Steps
Drop inspection
Leakage
Levels
Cable modem range
& register
EMTA installation
– Call connection and voice
  quality tests
Subscriber education
Phone call
     VoIP Test & Measurement – Why Test?
     VoIP Test & Measurement – Why Test?


Network Test Diagram
                               VoIP Test & Measurement – Why Test?
                               VoIP Test & Measurement – Why Test?



Test Process – Step 1                               DQR7
                                                     DQR7
                                                    Server
                                                     Server




                                                      Connect PP30 to home
                                                         telephone outlet


                PP30
                             RJ11
                                                                PowerProbe

                      EMTA
                                    RF
                                                      Network
860 DSPi
           Ethernet
                         VoIP Test & Measurement – Why Test?
                         VoIP Test & Measurement – Why Test?


Test Process – Step 2                              DQR7
                                                    DQR7
                                                   Server
                                                    Server



                                            Browse to DQR7 Server,
                                                authenticate, enter node
                                                and number to call


                   PP30
                                RJ11

                                                             PowerProbe
                                       RF
                         EMTA                    Network
   860 DSPi
              Ethernet
              VoIP Test & Measurement – Why Test?
              VoIP Test & Measurement – Why Test?


Test Initiation
                           VoIP Test & Measurement – Why Test?
                           VoIP Test & Measurement – Why Test?



Test Process – Step 3                                 DQR7
                                                       DQR7
                                                      Server
                                                       Server




                                            PowerProbe initiates
                                               calls as directed by
                                               DQR7 Server
                   PP30
                                RJ11                              PowerProbe

                                       RF
                         EMTA
                                                    Network
   860 DSPi
              Ethernet
          VoIP Test & Measurement – Why Test?
          VoIP Test & Measurement – Why Test?



PowerProbe 30 – Audio Sample
        Speech & DTMF (PP30)
                 VoIP Test & Measurement – Why Test?
                 VoIP Test & Measurement – Why Test?



Test Process – Step 4                            DQR7
                                                  DQR7
                                                 Server
                                                  Server




                                             PP30 answers calls, and
                                                converses with
                                                PowerProbe

                    PP30
                                 RJ11
                                                             PowerProbe

                          EMTA
                                        RF     Network

    860 DSPi
               Ethernet
                       VoIP Test & Measurement – Why Test?
                       VoIP Test & Measurement – Why Test?



Test Process – Step 5                                 DQR7
                                                       DQR7
                                                      Server
                                                       Server

Results displayed
   on 860 screen



                             PP30
                                          RJ11


                                                                PowerProbe
                                                 RF
                                   EMTA
                                                      Network
            860 DSPi
                        Ethernet
          VoIP Test & Measurement – Why Test?
          VoIP Test & Measurement – Why Test?



       Test Results




Pass / Fail                   Drill-Down
              VoIP Test & Measurement – Why Test?
              VoIP Test & Measurement – Why Test?



              Test Suite
Speech and DTMF Test
– DTMF - Dual Tone Multi-Frequency (touch
  tones)
– PESQ (ITU P.862)
– Voice Quality Evaluation System (VQES)
– Echo & Noise Agent
– Call Connection Statistics
– DTMF Transparency
– Fax Connection
                            VoIP Test & Measurement – Why Test?
                            VoIP Test & Measurement – Why Test?

                 PESQ (ITU P.862) Test
                                    • VAD / CODEC / PACKET LOSS
Objective Voice QoS                     Frame Muting Ratio (Analog perceived Packet Loss)
Measurements                            Speech Clipping Ratio
• Speech MOS                            Clipping Events

    PESQ-LQ Mean Opinion Score          Total, Maximum & Average Clipping Duration
     (5-1 MOS Rating) – ITU P.862       Proportion of Front-End & Back-End Clipping (VAD
                                        Issues)
         (4.59 – 1.16)
                                        Front-End & Back-End Clipping Events
    Speech Activity
                                        Average Front-End & Back-End Clipping Duration
• VF Measurements
                                        Proportion of In-Between clipping (Network Issues)
    Speech Power
                                        In-Between Clipping Events
    Loss
    Wideband Noise                      Average In-Between Clipping Duration

    Noise Gain                          Hang-Over Events
    Signal-to-Noise Ratio               Average Hang-over Time
                            VoIP Test & Measurement – Why Test?
                            VoIP Test & Measurement – Why Test?



   Voice Quality Evaluation System (VQES)
Objective Voice Quality                Perceived Voice Quality
Measurements                                  (specifically for Voice over IP)

   • Echo Path Loss (dB)                   VQES Mean Opinion Score (5-1
                                           MOS Score)
   • Echo Path Delay (ms)
                                           Percentage of Unusable, Difficult,
   • Speech Power (dBm)
                                           Irritating Calls (%)
   • C-Message Noise (dBrnC)

   • Wideband Noise (dBrn)

   • Speech Distortion

   • Roundtrip Delay (ms)
                           VoIP Test & Measurement – Why Test?
                           VoIP Test & Measurement – Why Test?

               Echo & Noise Test
Objective Voice Quality
                                  Network Performance Indicators
Measurements
                                      • Call Completion Ratio
   • Echo Path Loss (dB)
                                      • Call Loss Ratio
   • Echo Path Delay (ms)
                                      • Answer Seizure ratio (ASR)
   • Speech Power (dBm)
                                      • Network Effectiveness Ratio
   • C-Message Noise (dBrnC)
                                      (NER)
   • Wideband Noise (dBrn)
                                      • Answer Bid Ratio (ABR)
   • Roundtrip Delay (ms) -
   Approximation based on EPD
                   VoIP Test & Measurement – Why Test?
                   VoIP Test & Measurement – Why Test?



  Call Connection Statistics
Call Connection Timers
   • Ring Duration (s)
   • Post Dial Delay (s)

Connectivity Status
   • Call Setup Time (s)
   • Off-Hook
   • Call Duration (s)
   • Call Disposition Codes (Up to 15)
                         VoIP Test & Measurement – Why Test?
                         VoIP Test & Measurement – Why Test?


         DTMF Transparency Test
                          Digit Detection   Power Measurements
                           DTMF Digit 1     DTMF Power 1 (dBm)

                           DTMF Digit 2     DTMF Power 2 (dBm)
DTMF Performance
                           DTMF Digit 3     DTMF Power 3 (dBm)
• DTMF Detection Ratio
                           DTMF Digit 4     DTMF Power 4 (dBm)
• DTMF Avg. Power          DTMF Digit 5     DTMF Power 5 (dBm)

                           DTMF Digit 6     DTMF Power 6 (dBm)

                           DTMF Digit 7     DTMF Power 7 (dBm)

                           DTMF Digit 8     DTMF Power 8 (dBm)

                           DTMF Digit 9     DTMF Power 9 (dBm)

                           DTMF Digit 0     DTMF Power 0 (dBm)

                           DTMF Digit *     DTMF Power * (dBm)

                           DTMF Digit #     DTMF Power # (dBm)
                              VoIP Test & Measurement – Why Test?
                              VoIP Test & Measurement – Why Test?


             Fax Connection Test
Call Connection                    Negotiated Parameters and Far-end Capabilities
                                       Binary File Transfer
    • Off-Hook
                                       Maximum Baud Rate
    • Call Duration (s)
    • Call Disposition Code            Data Compression

    • Hang-up Codes                    Error Correction

    • Remote Station ID                Page Length
                                       Page Width
                                       Scan Time per Line
                                       Vertical Resolution

                                   Network Performance Indicators
                                       • Call Completion Ratio (CCR)
                                       • Calls at Maximum (Max.) Speed Ratio
                                       • Call Disposition Code Ratios
                                             -Transmission Speed Ratios
                VoIP Test & Measurement – Why Test?
                VoIP Test & Measurement – Why Test?



                    PP30
Simple, comprehensive test
verifies digital phone service
quality
Pass/fail results eliminate
measurement interpretation
mistakes
Drill-down for detailed
troubleshooting information
Cost efficient and complete test
solution
               VoIP Test & Measurement – Why Test?
               VoIP Test & Measurement – Why Test?



     Installation Test Review
Installation testing as a feature of the
network-based test system
Step-by-step process
Display examples
Review of tests performed
VoIP Test & Measurement – Why Test?
VoIP Test & Measurement – Why Test?



 Thanks!

				
DOCUMENT INFO