• Hurst parameter
• long-range dependence
• parameter estimation
• telecommunications networks
• wavelet decomposition
• packet traffic
• time-scale analysis

Network traffic exhibits fractal characteristics such as self-similarity and long-range dependence. Several estimators of fractal parameters have been developed, but few consider the possibility of tracking the time evolution of those parameters. Some fractal-aware network algorithms such as effective bandwidth estimation, admission control or traffic prediction could improve their performance if an accurate description of the time evolution of traffic fractality were developed. The aim of the present dissertation is to study the temporal evolution of fractal traffic parameters via a wavelet-based analysis. In particular, three different wavelet transforms are presented: Discrete Wavelet Transform (DWT), Wavelet Packet Transform (WPT) and Stationary Wavelet Transform (SWT). The dynamic estimation of the fractal parameters of a locally monofractal process is strongly dependent on the choice of the variance change points algorithm to be applied to the wavelet coefficients series. Namely, we discuss the following two procedures: Iterated Cumulative Sum of Squares (ICSS) and Schwarz Information Criterion (SIC).