A simple model for MPEG video traffic
Hai Liu, Nirwan Ansari, Y.Q. Shi · 2002
A simple approach to model MPEG coded video traffic is proposed. The idea is to first decompose the MPEG video traffic into three components, corresponding to I, P, and B frames, each modeled by a self-similar process. These processes are then structurally modulated in a manner similar to how the frames are grouped into the GOP (Group of Pictures) pattern. The new model is shown to be able to capture the statistical properties (both LRD and SRD) of the MPEG video traffic.