A multifractal traffic model with application in network traffic prediction proposal and evaluation / Um modelo multifractal aplicado a predição de trafego de redes-proposta e avaliação

AUTOR(ES)
DATA DE PUBLICAÇÃO

2005

RESUMO

Dynamic traffic control mechanisms can improve network resource utilization. The fractal traffic correlation structure can be exploited for designing dynamic network control mechanisms. Playing the most important role in the dynamic control mechanisms, traffic predictors can be accomplished from statistic traffic models able to describe the network traffic behavior. Firstly, this work proposes an extension to the widely used fractional Brownian traffic model, making it able to describe not only the self-similar, but also the multifractal traffic characteristics. In the following, assuming that the traffic statistic behavior can be modeled by the extended fractional Brownian traffic model, a novel traffic predictor is proposed. Experimental investigations show that the extended fractional Brownian traffic can match the multifractal traffic behavior, leading to modeling improvements and better prediction results

ASSUNTO(S)

redes de computação previsão telecomunicações - trafego telecommunication traffic fractais fractal computer networks forecasting

ACESSO AO ARTIGO

Documentos Relacionados