Estimation of Network Traffic Hurst Parameter Using HHT and Wavelet Transform

It has been demonstrated that both wide and local area network traffic are statistically self-similar. The Hurst index is the only parameter to characterize the self-similarity. The real-time normal network data stream should accord with network traffic´s statistical self-similarity and its long-range dependence (LRD) features correspondingly, which can be judged by the value of Hurst parameter. Wavelet transform is a common method used to estimate self-similar parameter. However, the wavelet analyses can not eliminate the influence of non-stationary signal´s periodicity and trend term. In view of the fact that Hilbert-Huang transform (HHT) has unique advantage on nonstationary signal treatment, a refined self-similar parameter estimation algorithm is designed in this paper through the combination of wavelet analysis and Hilbert-Huang transform and a set of experiments are run to verify the improvement in the accuracy of parameter estimation.