论文部分内容阅读
在研究通信网络的流量控制及网络性能时，业务源模型是一个十分重要的组成部分。传统的业务模型大多基于Ｐｏｉｓｓｏｎ模型或其改进形式，假定业务突发长度呈负指数分布。近年来对大量的ＷＡＮ、ＬＡＮ、及ＶＢＲ业务、ＩＳＤＮ中传输的业务等多种业务的测量表明：在这些业务中普遍存在着自相似性（或称为长相关性）。该性质显著影响宽带网的流量控制及排队分析，因而已经受到越来越多的关注。本文综述了近几年在自相似业务模型研究方面所取得的主要成果，包括自相似业务的数学描述、建模方法、排队分析等。
The business source model is a very important component when studying traffic control and network performance in communication networks. Traditional business models are mostly based on the Poisson model or its improved form, assuming that the burst length of the business is negative exponential distribution. In recent years, the measurement of a large number of services such as WAN, LAN, and VBR services, and services transmitted in ISDN has shown that there is widespread self-similarity (or long-term correlation) among these services. This nature has a significant impact on traffic control and queuing analysis of broadband networks and has received more and more attention. This article summarizes the major achievements made in the research of self-similar business models in recent years, including the mathematical description, modeling methods and queuing analysis of self-similar business.