Adaptive estimation of parameters of multiplicative cascade-based network traffic model in wavelet domain

摘要

An adaptive and recursive approach to the network traffic modelling taking into account multifractal characteristics of the network traffic flows is proposed. To this end, equations for some statistics of the scaling coefficients and for the model multipliers in order to develop a new algorithm that enables the adaptive estimation of parameters of a multiplicative cascade in the wavelet domain are presented. Simulations and comparisons with other three important models in the literature were carried out in order to verify the performance of the proposed approach.