摘要： 针对表征自相似网络流量统计特性的赫斯特(Hurst)指数，讨论一种基于经验模式分解的Hurst指数估计算法。该算法通过对自相似网络流量数据进行自适应分解，得到一组满足指定余项误差的固有模态函数分量，由其能量对数化函数与Hurst指数之间的线性拟合，估计出Hurst指数。实验表明，该算法能对自相似网络流量的Hurst指数进行自适应估计。

关键词: 自相似, 赫斯特指数, 经验模式分解

Abstract: This paper discusses a new method based on the Empirical Mode Decomposition(EMD) algorithm to estimate the Hurst index that is an important statistical parameter of self-similar network traffic. The algorithm can adaptively decompose self-similar traffic into a series of Intrinsic Mode Function(IMF). By using the relationship between the energy of IMFs and the Hurst index, it can adaptively estimate the Hurst parameter of self-similar traffic. Experimental results show that this algorithm can adaptively estimate the Hurst index of self-similar traffic.

Key words: self-similar, Hurst index, Empirical Mode Decomposition(EMD)

中图分类号: 

• TP393