Weibull分布的自相似流量及其对排队性能影响的仿真分析

摘要

深入研究自相似流量,对网络性能的评估优化是至关重要的.通过对Weibull分布的自相似性进行理论分析及仿真,表明自相似性源于信源的属性,并不是单一流量叠加的结果.在NS2下,采用叠加Weibull分布的ON/OFF信源的方法实现了一种自相似流量的生成,并将其应用于GI/GI/1/K/FIFO排队模型网络性能的仿真分析中,补充了现有文献的仿真结论.%Thorough analysing the self-similar traffic is crucial to the estimation and optimization of the network performance. For Weibull distribution,in this paper it has been proved by means of both theoretical analysis and computer simulation that its self-similarity originates from the property of traffic sources rather than the mere aggregation. The thesis presented an implementation of forming self-similar traffic achieved by the aggregation of Weibull distributed ON/OFF sources with NS2 as well as its application in a further simulative analysis on the network performance of the GI/GI/1/K/FIFO queuing model,and this complemented the simulation conclusions of existing references.