The Hurst parameter characterizes the degree to which a time series is long range dependent (LRD). The value of this parameter can be used as an input to algorithms for bandwidth allocation, buffer sizing and congestion control. However, for these algorithms to be effective over the long run they must change their actions when the value of the Hurst parameter changes. We demonstrate a new technique which uses a wavelet decomposition to detect a change in the Hurst parameter. Our technique tests the variance structure of the wavelet coefficients at multiple scales and uses changes in variance to signal a change in the value of the Hurst parameter. The efficacy of the proposed technique is demonstrated by comparing its performance to that of another recently proposed method for change detection. The performance tests were conducted using artificially generated data sets which contain changes in the Hurst parameter of known position, magnitude and sign.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.