VBR video traffic modeling and synthetic data generation using GA-optimized Volterra filters
International Journal of Network Management
| Hi-index | 0.00 |
The major driving force behind the emerging ofbroadband integrated networks is traffic from videosources. Since this traffic will be a substantial portion ofthe overall traffic carried by the integrated networks, it isextremely important that adequate source models arefound. The networks' dimensioning depends on thestatistical properties of traffic sources. For video sourcesthe Moving Picture Experts Group (MPEG) compressionscheme was proposed several years ago and it has becomethe defacto standard for video compression since then.However, even with the huge reduction of bits that MPEGcompression provides, it does not smooth the video traffic.Indeed the MPEG compression algorithm guarantees thatthe MPEG stream will be bursty.The aim of the work outlined in this paper was findinga suitable model for the MPEG coded video traffic. Thepaper concentrates on a variety of models and theircapabilities to model adequately the statistical propertiesof MPEG video. In particular, the Transform-Expand-Sample (TES), the Autoregressive (AR), theAutoRegressive Moving Average (ARIMA) and the firstorder statistics histogram models were investigated interms of both accuracy and complexity. A short discussionregarding the tradeoff betwee the accuracy and thecomplexity of these models is also provided.