Neurocomputação na Estimativa do Parâmetro de Hurst

Danielo GomesNazim AgolmineNelson L. S. da FonsecaJosé Neuman de Souza

In this work, an Hurst parameter estimator by using Artificial Neural Networks (ANNs) has been developed. The traffic for this analysis is generated by an algorithm and the results are compared with three others estimators, in order to evaluate the accuracy of the proposal. It is also investigated the influence of some elements of the neural nets' learning - hidden neurons and activation function - in the final performance. Simulation results show that the neural network approach can be used on the Hurst parameter estimation for data traffic.

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: