Fine Tuning of Fuzzy Token Bucket Scheme for Congestion Control in High Speed Networks

In ATM networks, due to the dynamic nature of the bursty traffic, severe network congestion may occur. Therefore, policing as the preventive congestion control is required for properly guaranteeing QoS. The leaky bucket control mechanism, which is the most widely and only implemented one, has some drawbacks. In this paper the principles of fuzzy logic are deployed to tackle the problem of congestion control. This paper implements the fuzzy logic based token bucket scheme in which token generation rate is controlled by Fuzzy Logic. This work studies how to implement token rate predicting and policing in a high speed (ATM) network using fuzzy token bucket. The fuzzy token rate predictor facilitates token rate predicting and dynamic policing. It is able to achieve lower delay, less cell loss and higher throughput by Fuzzy token bucket scheme. A simulation study was performed for the fuzzy logic predictor using MATLAB-Simulink Simulator. The simulation results show that the fuzzy logic predictor gave reasonable token rate prediction compared to a non-fuzzy Token bucket scheme.