Optimisation of the performance of a rate-based congestion control system by using fuzzy controllers

Abstract

This article describes a congestion control protocol, based on the use of fuzzy controllers, for efficient utilisation of the capacity of the network. The basic control mechanism uses rate-based control, closely resembling the one specified by the ATM Forum for the Available Bit Rate (ABR) service category. Fuzzy logic is utilised in the network elements to enhance the required control decisions with the general aim of fair and optimal utilisation of the available bandwidth. The control protocol adapts to the changes of the load produced by CBR and VBR connections, and shares the available capacity of the network fairly between ABR connections. Simulations have been carried out to test the performance of the control system in the presence of several VBR and ABR connections with statistical traffic distributions and with varying buffer sizes of the network elements. Results of the tests show the adaptability and efficiency of the control system.