Hierarchical neuro-fuzzy call admission controller for ATM networks

Abstract

In this work, a hierarchical neuro-fuzzy call admission controller for ATM networks based on the GARIC architecture is proposed. The controller contains a neural network as a critic, using the reinforcement learning scheme, and three fuzzy sub-networks, controlling cell loss ratio, queue size and link utilization in the ATM multiplexer. The final decision of the call admission controller is obtained as a weighted combination of the decisions generated by the fuzzy sub-networks. In order to study the performance of the proposed controller, it is simulated under various variable bit rate traffic patterns and the results obtained are evaluated in terms of network utilization. Introduction of an initial knowledge base to improve training times is discussed and the results with and without the knowledge base are given. Finally, methods to enhance the performance of the proposed controller are mentioned.