TITLE:
Another Important Parameter’s Research on Estimating Self-Similarity
AUTHORS:
Peng Luo, Juan Wang
KEYWORDS:
Self-Similarity, Hurst Parameter, Parameter cf, Long-Rang Dependence
JOURNAL NAME:
International Journal of Communications, Network and System Sciences,
Vol.9 No.8,
August
22,
2016
ABSTRACT: It is convincingly demonstrated by numerous
studies that the self-similarity of modern multimedia network traffic is
presented by Hurst parameter (H). The specific performance is that the similar
degree is higher along with the increase of H when H is between 0.5 and 1.
However, it is doubtable that whether the complicated process of
self-similarity can be described comprehensively by the parameter H only.
Therefore, another important parameter cf has been proposed based on the
discrete wavelet decomposition in this paper. The significance of the
parameters is provided and the performance of the self-similarity process is
described better.