A hybrid histogram and neural based call admission control for VBR video traffic

We propose a hybrid neural network (NN) approach to estimate cell loss rate of variable bit rate (VBR) video traffic for call admission control (CAC) in an ATM environment. Existing CAC algorithms, which are mostly based on the on-off model, do not appear to apply well to VBR video traffic. In reality, VBR video sources are not two-state on-off sources. Recently, a histogram based model for video traffic behavior was proposed which is able to overcome most of the deficiencies in conventional approaches and can handle VBR video traffic in various traffic situations. However, it has some problems: unable to guarantee cell loss rates for short burst periods; overestimation of cell loss rates during call set up etc. We therefore propose a NN based hybrid approach, where NN is used to refine the evaluation result by applying the knowledge gained of the actual performance of the histogram scheme. We show how performance data derived from histogram based approach can be used as training data in the NN training scheme to produce even better results than the pure histogram based approach, while still retaining the merits of it