The application of fuzzy logic in real-time traffic control in ATM networks

A fuzzy logic-based system is presented to deal with the real-time traffic control problem in ATM networks. We have combined the modified leaky bucket (LB) technique with the moving window mechanism to identify the traffic parameters. The fuzzy leaky bucket (FLB) modifies the token rate according to the peak rate and the burst time which characterize the source behavior. Simulation results show that our system is transparent to sources that respect the parameter values negotiated at the call setup and is effective in detecting source violation with low response time.