WAVELET ANALYSIS OF LONG-RANGE-DEPENDENT TRAFFIC

A wavelet-based tool for the analysis of long-range dependence and a r elated semi-parametric estimator of the Hurst parameter is introduced, The estimator is shown to be unbiased under very general conditions, and efficient under Gaussian assumptions, It can be implemented very e fficiently allowing the direct analysis of very large data sets, and i s highly robust against the presence of deterministic trends, as wed a s allowing their detection and identification. Statistical, computatio nal, and numerical comparisons are made against traditional estimators including that of Whittle. The estimator is used to perform a thoroug h analysis of the long-range dependence in Ethernet traffic traces, Ne w features are found with important implications for the choice of val id models for performance evaluation, A study of mono versus multifrac tality is also performed, and a preliminary study of the stationarity with respect to the Hurst parameter and deterministic trends.