Wavelet and Time-Domain Modeling of Multi-Layer VBR Video Traffic.pdfVIP

Wavelet and Time-Domain Modeling of Multi-Layer VBR Video Traffic Min Dai Dmitri Loguinov Department of Electrical Engineering Department of Computer Science Texas AM University Texas AM University College Station, TX 77843 College Station, TX 77843 min@ dmitri@ Abstract— Modeling of video traffic has always been an impor- dependencies between frame sizes in VBR traffic. In regard tant research area for the design, simulation, and performance to the first issue, several models have been proposed for the analysis of packet video networks and Internet streaming. In this frame-size distribution, including the lognormal [7], Gamma paper, we present a frame-level hybrid framework for modeling multi-layer VBR video traffic, in which the base layer is modeled [25], and various hybrid distributions (e.g., Gamma/Pareto [15] using a combination of wavelet and time-domain methods and or Gamma/lognormal [23]). Compared to the task of fitting the enhancement layer is linearly predicted from the based- a model to the frame-size distribution, capturing the ACF layer model. To accurately capture long-range dependent (LRD) structure of VBR video traffic is more challenging due to the and short-range dependent (SRD) properties of VBR base-layer fact that VBR video exhibits both LRD and SRD properties traffic, we use wavelets to model the distribution of I-frame sizes and a simple time-domain model for P/B frame sizes; [9], [16