On the relevance of long term correlation in MPEG-1 video traffic

Variable Bit Rate, VBR, video is expected to become increasingly important with the large scale deployment of Broadband-Integrated Services Networks, B-ISDNs, over the next few years. Although the modeling of VBR video sources has recently received significant attention, there is currently no widely accepted model which lends itself to mathematical analysis. Furthermore, new video compression standards, such as the MPEG family, are emerging. On the basis of results of a detailed statistical analysis of a long sample of a movie encoded with the MPEG-1 algorithm, an analytically tractable model is developed and analyzed in detail. The model is able to capture both the distributional and temporal characteristics of this kind of traffic. The model was validated using a two-hour long sequence generated by the MPEG coding of the movie "Star Wars". We show that our model is a flexible tool to study network issues such as bandwidth allocation and statistical multiplexing.