A source model for VBR video traffic based on M/G/∞ input processes

Statistical evidence suggests that the autocorrelation function of a compressed-video sequence is better captured by ρ(k)=e^-β√(k) than by ρ(k)=^-β =e^{-β log k} (long-range dependence) or ρ(k)=e ^-βk. A video model with such a correlation structure is introduced based on the so-called M/G/∞ input processes. Though not Markovian, the model exhibits short-range dependence. Using the queueing performance under “real” video traffic as a reference, we study via simulations the queueing performance under two video models: the M/G/∞ model and the fractional ARIMA (F-ARIMA) model (which exhibits LRD). Our results indicate that the M/G/∞ model is much more accurate in predicting the actual queueing performance than the F-ARIMA model. Furthermore, only 𝒪(n) computations are required to generate an M/G/∞ trace of length n, compared to 𝒪(n²) for a F-ARIMA trace