Jitter Analysis of CBR Streams in Multimedia Networks
Abstract
The performance of real time applications such as video and voice streams relies on packet delay jitter. Large delay jitter causes buffer overflow or underflow at the receiver end and the user encounters interrupts. The delay jitter is mainly due to the perturbation of background traffic in the bottleneck router. Fulton and Li 12 provide an analytical approximation for the first-order and second-order statistics of delay jitter. However, their analysis is based on a Markovian model of the background traffic, which is not quite suitable for Internet traffic and requires lots of computational effort. We propose an efficient method to predict the jitter variance of a CBR (constant-bit rate) connection based on the wavelet model of the background traffic. The wavelet analysis extracts the statistical properties of background traffic and the analysis result can be used to predict an upper bound for the jitter variance of the CBR connection. time scales and multifractal at small time scales. The traditional Markovian traffic model is unable to capture this multifractal behavior well and is not a proper model for performance analysis 19 20. In order to predict the jitter more efficiently and accurately, we applied wavelet analysis to characterize the background traffic and propose an upper bound for the jitter variance of a constant-bit rate connection. The arrangement of this paper is as follows. In the next section, we briefly introduce the wavelet analysis for traffic and show the multifractal behavior of a real traffic trace. In section 3, the background traffic is characterized by the Logscale diagram. An approximation of queue length distribution is derived from properties of wavelets
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 01, 2003
- Accession Number
- ADA439338
Entities
People
- Jia-shiang Jou
- John Baras
Organizations
- University of Maryland