Video traffic modeling based on RBF networks

The basis of traffic and congestion control in the ATM network lies in the traffic modeling. Due to the variety of services and high transmitting speed, the traditional analytical methods become intractable. As an alternative, a radial basis function (RBF) neural network has been adopted for video traffic modeling. In this paper, an LBG algorithm based and Hestenes singular value decomposition (SVD) method has been proposed to select the centers of the hidden layer neuron and to calculate the weights of the output layer synapse in the RBF network. To evaluate the proposed approach, movie data have been utilized in several experiments.