ProgCon: um Sistema para Prognóstico de Congestionamento de Tráfego de Redes

Jorge Castro e SilvaMarcilia CamposPaulo Cunha

The behavior of the current networks traffic is chaotic, possessing statistical properties which include complex temporal correlations. In this scenario, a problem not yet solved is related to traffic congestion. A non stationary characteristic, short and long-range temporal dependences and the present noises in the traffic give opportunity to use new techniques that allow the use of simple models for its analysis. This work proposes a mechanism for forecasting possible congestions through the identification of the tendency component in time series that represent the network real traffic, using discrete wavelet transform. The proposed mechanism is efficient since can detect congestion in the traffic network.

Caso o link acima esteja inválido, faça uma busca pelo texto completo na Web: Buscar na Web

Biblioteca Digital Brasileira de Computação - Contato:
     Mantida por: