突发业务流的TES建模方法

TES Modeling Method for Traffic with Burstiness

高速通信网络存在大量诸如可变速率(VBR)视频这类具有突发性的业务流。突发性,在数学上主要通过业务流的到达时间间隔的边缘概率分布和自相关函数来描述。TES(Transform-Expand-Sample)是一种建模静态随机过程的非参数化技术,能准确匹配边缘概率分布并很好地近似自相关函数。本文简要介绍了TES方法的基本原理,再详细讨论突发业务流的TES建模及其软件实现。对MPEG视频业务的建模的仿真结果表明,TES模型能够很好地表征网络业务流的突发性。

A large number of traffic streams with burstiness, such as VRB (variable bit rate) video traffic, exist in high-speed communications networks. The mathematical underpinnings of burstiness are complex. Two main factors contributing to traffic burstiness are the marginal probability distribution and autocorrelation function of traffic interarrival times. TES (Transform-Expand-Sample ) is a nonparametric technique that enable a user to model stationary stochastic processes with an exact match of the marginal probability distribution and a good approximation of the autocorrelation function. In this paper, after brief introduction to the fundamental principle of TES methods, TES modeling and its software development of traffic streams with burstinessa are discussed in details. The simulation results on modeling of MPEG-compressed video traffic demonstrate that TES models are able to characterize the burstiness of network traffic streams.