Docstoc

PowerPoint Presentation - for IEEE 802

Document Sample
PowerPoint Presentation - for IEEE 802 Powered By Docstoc
					                            Trace Based Streaming Video Traffic Model for 802.16m Evaluation Methodology Document

IEEE 802.16 Presentation Submission Template (Rev. 9)
Document Number:
   IEEES80216m-07_179
Date Submitted:
   2007-09-10
Source:
   Ricardo Fricks, Hua Xu
   email: hua.xu@motorola.com
   Motorola Inc.
   1501 West Shure Drive, Arlington Heights, IL 60004, USA
   Ronny (Yong-Ho) Kim, Kiseon Ryu
   email: ronnykim@lge.com
   LG Electronics Inc.
   LG R&D Complex, 533 Hogye-1dong, Dongan-gu Anyang, 431-749, Korea
Venue:
   Response to a call for comments and contributions on draft 802.16m Evaluation Methodology Document
Base Contribution:
   None
Purpose:
   Propose to adapt the trace based streaming video traffic model as mandatory model in draft IEEE 802.16m Evaluation Methodology
   Document.
Notice:
   This document does not represent the agreed views of the IEEE 802.16 Working Group or any of its subgroups. It represents only the views of the participants listed in
   the “Source(s)” field above. It is offered as a basis for discussion. It is not binding on the contributor(s), who reserve(s) the right to add, amend or withdraw material
   contained herein.
Release:
   The contributor grants a free, irrevocable license to the IEEE to incorporate material contained in this contribution, and any modifications thereof, in the creation of an
   IEEE Standards publication; to copyright in the IEEE’s name any IEEE Standards publication even though it may include portions of this contribution; and at the IEEE’s
   sole discretion to permit others to reproduce in whole or in part the resulting IEEE Standards publication. The contributor also acknowledges and accepts that this
   contribution may be made public by IEEE 802.16.
Patent Policy:
   The contributor is familiar with the IEEE-SA Patent Policy and Procedures:
                                     <http://standards.ieee.org/guides/bylaws/sect6-7.html#6> and <http://standards.ieee.org/guides/opman/sect6.html#6.3>.
   Further information is located at <http://standards.ieee.org/board/pat/pat-material.html> and <http://standards.ieee.org/board/pat >.
    Propose to adapt trace based streaming
    video model as MANDATORY model (1)
• Trace-based streaming video traffic model represents the
  key characteristics of streaming video application better than
  the proposed statistical models
   – Traffic model in the evaluation methodology document (EVM) should focus
     on capturing the accents of the application which posts special demand on
     the system performance,
   – In the streaming video traffic model case, the long rang dependency (LRD) is
     the key characteristic that needs to be captured, because high burstiness
     resulting from LRD posts high demand on both transport and buffering
     capability in the system
   – The statistical models proposed in EVM don’t exhibits the LRD characteristic
     of encoded video traces. (See slides 4-6)
• Trace-based streaming video traffic model can be easily
  used for design comparison
   – The trace based streaming video traffic model can be easily duplicated by
     other system simulators for design comparison
   – No additional statistical variance is introduced by the trace based traffic
     model itself, which make it easier to compare among results
    Propose to adapt trace based streaming
    video model as MANDATORY model (2)
• Trace-based streaming video traffic model is widely
  available and easy to implement
   – Trace based streaming video traffic model an be easily generated by every
     system simulator (simply read in the trace text file) with no ambiguity
   – It exhibits LRD in an hour trace where large sample size (over 100K
     samples) will be needed for a statistical model to exhibit LRD
   – There is a public library with 200+ video traces with multiple encoding
     characteristics (e.g., video quality and resolution) and from diverse video
     sources (e.g., movies, TV shows, teleconferences, remote learning,
     cartoons, sporting events).
• Trace-based steaming video traffic model can represent the
  general population of the streaming video
   – The trace based streaming video model consists of 12 traces carefully
     picked from five genre with different data rate, quantization value, hurst
     parameters, etc.
   – Users (in the simulation) shall pick a trace randomly from the 12 candidates
   – Further more, a random starting point in the trace shall also be picked by the
     user
  Effect of LRD on Network Performance
• Network performance degrades gradually with
  increasing LRD (self-similarity).
• The more self-similar the traffic, the slower the
  queue length decays.
• Aggregating streams of self-similar traffic typically
  intensifies the self-similarity ("burstiness") rather
  than smoothing it.
• The bursty behaviour may itself be bursty, which
  exacerbates the clustering phenomena, and
  degrades network performance.
• QoS depends on coping with traffic peaks - video
  delay bound may be exceeded.
Video Traffic Modeling

    Uncorrelated Random Variables                              Old school surveys:
         Renewal Processes

             Poisson Processes
             Bernoulli Processes
             Phase-Type Renewal Processes


    SRD: Short-Range Dependent Processes

         Markov Processes

             DTMC: Discrete-Time Markov Chains
             CTMC: Continuous-Time Markov Chains
             SMP: Semi-Markov Processes
             MRGP: Markov Regenerative Processes
             MAP: Markovian Arrival Process


         Markov Modulated Processes

             IPP: Interrupted Poisson Processes
             IBP: Interrupted Bernoulli Processes
             IFP: Interrupted Fluid Processes
             MMPP: Markov Modulated Poisson Processes
             MMBP: Markov Modulated Bernoulli Processes
             MMFP: Markov Modulated Fluid Processes
                                                               New modeling trend using traces:
         Regression Models

             MA: Moving Average
             AR: Autoregressive
             DAR: Discrete Autoregressive
             ARMA: Autoregressive Moving Average
             ARIMA: Autoregressive Integrated Moving Average
             TES: Transform-Expand-Sample

    LRD: Long-Range Dependent Processes

         Self-Similar Models

             fARIMA: Fractional ARIMA
             fGN: Fractional Gaussian Noise
             fBM: Fractional Brownian Motion
             Aggregation of High-Variability ON/OFF Sources
Video Traffic Modeling

    Uncorrelated Random Variables
                                                                      802.16m streaming video model:
         Renewal Processes

             Poisson Processes
             Bernoulli Processes
             Phase-Type Renewal Processes


    SRD: Short-Range Dependent Processes                      AR(2)
         Markov Processes

             DTMC: Discrete-Time Markov Chains
             CTMC: Continuous-Time Markov Chains
             SMP: Semi-Markov Processes
             MRGP: Markov Regenerative Processes
             MAP: Markovian Arrival Process
                                                                      ... but long-range dependence is
                                                                      intrinsic to VBR encoded streams
         Markov Modulated Processes

             IPP: Interrupted Poisson Processes
             IBP: Interrupted Bernoulli Processes
             IFP: Interrupted Fluid Processes
             MMPP: Markov Modulated Poisson Processes
             MMBP: Markov Modulated Bernoulli Processes
             MMFP: Markov Modulated Fluid Processes

         Regression Models

             MA: Moving Average                                       and there is evidence that LRD
             AR: Autoregressive
             DAR: Discrete Autoregressive                             processes can negatively affect
             ARMA: Autoregressive Moving Average
             ARIMA: Autoregressive Integrated Moving Average          multiplexing performance
             TES: Transform-Expand-Sample

    LRD: Long-Range Dependent Processes

         Self-Similar Models

             fARIMA: Fractional ARIMA
             fGN: Fractional Gaussian Noise
             fBM: Fractional Brownian Motion
             Aggregation of High-Variability ON/OFF Sources
               Silence of the Lambs
• Besides the statistical characteristics of the same video stream will
  depend of many factors: GoP size, quantization, etc




     Silence of the Lambs (H.264/AVC, G16B1, CIF 352x288, 30fps, ~30 min.)
     Quantizer (QP)                                 10          16          48
     I frames     mean (bytes)                 22,723     12,478         373
                  standard deviation (bytes)    92,969     64,578      2,027
     P frames     mean (bytes)                 11,663      4,320            70
                  standard deviation (bytes)    59,255     35,236          652
     B frames     mean (bytes)                  2,726         855           21
                  standard deviation (bytes)    35,138     15,415           62
     Hurst parameter (R/S)                        0.97       0.94       0.82

                  Long-Range
                  Dependence
               Properties:

               Encoder: H.264 Full
               Variable Bit Rate (VBR)
Long-Range     Frame Size: CIF 352x288
Dependence     GoP Size: 16
               No. B Frames: 1
               Quantizer: 10




               Properties:

               Encoder: H.264 Full
  Long-Range   Variable Bit Rate (VBR)
  Dependence   Frame Size: CIF 352x288
               GoP Size: 16
               No. B Frames: 1
               Quantizer: 16
              Properties:

              Encoder: H.264 Full
              Variable Bit Rate (VBR)
 Long-Range   Frame Size: CIF 352x288
 Dependence   GoP Size: 16
              No. B Frames: 1
              Quantizer: 48




Short-Range
Dependence
    (?)       Properties:

              Encoder: MPEG4 Rel2
              Variable Bit Rate (VBR)
              Frame Size: (?) 320x240
              GoP Size: 12
              No. B Frames: 8
              Quantizer: (?)

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:5
posted:12/3/2011
language:English
pages:9