Study on Self-adaptive Systematic Double Sampling method for Self-similar Network Traffic  

作　　者：Liu Yuanzhen Liu Yuan Li Xiaohang 

机构地区：[1]So Yangtze Univ, Sch Informat Engn, Wuxi Jiangsu 214122, Peoples R China

出　　处：《China Communications》2007年第2期86-89,共4页中国通信（英文版）

摘　　要：This paper proposes anew variation of systematic sampling, called Self-Adaptive Systematic Double Sampling(SSDS). This algorithm can fully consider the self-similar and heavy-tailed distribution characteristics of network traffic and estimate Hurst parameter correctly.The experiments on real Internet traces indicate,compared with traditional sampling methods,the new method advances the accuracy and practicability of the sampling measuring system obviously,and can achieve simplicity,adaptability and controllability of resource consumption.This paper proposes a new variation of systematic sampling, called Self-Adaptive Systematic Double Sampling(SSDS). This algorithm can fully consider the self-similar and heavy-tailed distribution characteristics of network traffic and estimate Hurst parameter correctly. The experiments on real Internet traces indicate, compared with traditional sampling methods, the new method advances the accuracy and practicability of the sampling measuring system obviously, and can achieve simplicity, adaptability and controllability of resource consumption.

关 键 词：SELF-SIMILARITY Hurst PARAMETER DOUBLE sampling 

分 类 号：TP393.06[自动化与计算机技术—计算机应用技术]