Analysis, Modeling and Generation of Self-Similar VBR Video Traffic-英文文献.pdf

Analysis, Modeling and Generation of Self-Similar VBR Video Traffic Mark W. Garrett Walter Willinger Bellcore 445 South St. Morristown NJ 07960 Abstract allocate and regulate bandwidth assigned to the service. Short periods of congestion may be handled gracefully through the We present a detailed statistical analysis of a 2-hour long em- use of prioritization(layered coding) [PVW91], congestion no- pirical sample of VBR video. The sample was obtained by tification [GARR93], and intelligent scheduling algorithms at applying a simple intraframe video compression code to an the switch [CLAR92]. However, good design and analysis of action movie. The main findings of our analysis are (1) the a network requires an understanding of the traffic itself. This tail behavior of the marginal bandwidth distribution can be paper presents a long trace of VBR video with statistical analy- accurately described using “heavy-tailed” distributions (e.g., ses and simulation results which address issues of both source Pareto); (2) the autocorrelation of the VBR video sequence modeling and network performance. decays