您的位置:首页  > 论文页面

基于经验模式分解的Hurst参数估计器的MATLAB实现

发表时间:2008-05-31  浏览量:2755  下载量:1102
全部作者: 单佩韦,李明
作者单位: 华东师范大学信息科学技术学院
摘 要: 通过对局域网和广域网上大量突发网络流量的分析结果表明,网络流量普遍存在着自相似性和长相关性,其中赫斯特(Hurst)指数是表征网络流量统计特征的重要参数。经验模式分解算法是估计自相似过程Hurst指数的新方法,该算法与目前广泛使用的小波方法的区别在于,其本身具有高度自适应性的特点,并能够分解为一组满足指定余项误差的固有模态函数分量。本文介绍了一个在矩阵实验室(MATLAB)平台上利用面向对象的图形用户界面开发环境(GUIDE)仿真系统设计并实现的Hurst参数估计器。该系统基于经验模式分解算法,支持多种输入输出数据格式,用户可按给定分段点对自相似网络流量进行Hurst参数估计。
关 键 词: 参数估计;自相似过程;赫斯特指数;经验模式分解;矩阵实验室
Title: Realization of Hurst index estimator based on empirical mode decomposition by MATLAB
Author: SHAN Peiwei, LI Ming
Organization: School of Information Science and Technology, East China Normal University
Abstract: The measurement studies show that the burstiness of packet traffic in LAN as well as WAN is associated with self-similar and long-range dependency, and Hurst index is the key value of this model representing the statistics of traffic. Empirical mode decomposition(EMD) algorithm is a new method of estimating Hurst index of self-similar processes. The algorithm, different from the wavelet algorithm which is widely used, has high adaptability and can be decomposed to a set of intrinsic mode functions(IMF) with appointed residues error. This paper introduces a Hurst index estimator designed and implemented based on MATLAB using the object-oriented GUIDE simulation system. The system, based on the empirical mode decomposition algorithm, supports several format of data for import and export. It can also estimate the Hurst index of self-similar network traffic by given blocks and provides a friendly interface.
Key words: parameter estimating; self-similar processes; Hurst index; empirical mode decomposition; MATLAB
发表期数: 2008年9月第10期
引用格式: 单佩韦,李明. 基于经验模式分解的Hurst参数估计器的MATLAB实现[J]. 中国科技论文在线精品论文,2008,1(10):1119-1122.
 
0 评论数 0
暂无评论
友情链接