ABSTRACT
Abstract Variable Bit Rate (VBR) MPEG traffic is expected to be one of the major
traffic sources for high speed networks. Compressed video traffic (MPEG) exhibits complex patterns which vary from one stream to another. One of the most important reasons for these fluctuations (or patterns) in the overall bit rate are the scene changes within a video stream. The aim of this paper is to propose a novel scene-based model considering scene change characteristics. We analyze the scene changes within an MPEG stream in more detail including scene change identification techniques. Compared to Markovian models, we show that the proposed model (scene-based ) provides a good approximation of the statistical characteristics behavior of the real MPEG stream. We also show that the model is capable of capturing the Long Range Dependency (LRD) feature of such traffic. To examine the appropriateness and limitations of these models, simulation experiments are conducted to study the performance over an ATM multiplexer using the generated (synthetic) sequence based on the presented models and the real traffic.
