自相似业务量产生机理与模型精度分析  

Analysis on generation mechanism of self-similar traffic and its accuracy of model

作　　者：陈炳锋[1] 徐从富[1] 何俊 

机构地区：[1]浙江大学计算机学院,杭州310027 [2]浙江省辐射环境监测站,杭州310012

出　　处：《计算机工程与应用》2009年第3期101-103,共3页Computer Engineering and Applications

基　　金：国家自然科学基金(No.60402010);国家高技术研究发展计划(863)(No.2007AA01Z197)~~

摘　　要：对通过叠加满足Pareto重尾分布的ON/OFF业务点产生自相似业务量的方法进行了较系统的分析。在介绍了这种自相似业务量实现方式的理论基础后,给出了一种检验业务点叠加数量与模型精度之间关系的研究方法。仿真结果表明:当叠加ON/OFF业务点数量为44时,自相似业务量Hurst参数的相对平均误差分别为6.8%(H=0.6)和5.0%(H=0.9);当叠加ON/OFF业务点数量为132时,自相似业务量Hurst参数的相对平均误差为1.25%(H=0.6)和0.85%(H=0.9)。In this paper,the self-similar traffic generated by multiplexing the ON/OFF points which meet the Pareto heavy-tail distribution is discussed.The corresponding theory of the method above is introduced and then the research method on the relationship of the number of the points and the precision of the self-similar traffic model is given.Simulation results show that the relative average error of the Hurst parameter in the self-similar model is 6.8% when H=0.6 and 5.0% when H=0.9 in the case that the number of ON/OFF points is 44.While corresponding data change to 1.25% when H=0.6 and 0.85% when H=0.9 respectively in the case that the number of ON/OFF points reach to 132.

关 键 词：自相似 重尾分布 PARETO ON/OFF 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]