Predictive fuzzy explicit rate allocation (PFERA) for traffic control in ATM networks

We propose an efficient and stable algorithm employing fuzzy logic for available bit rate (ABR) traffic in asynchronous transfer mode (ATM) networks. Large time-delay incurred in the feedback path and the statistical variation in the link capacity at the ATM node can cause buffer saturation or unfairness. The queue length of the ABR is estimated one step ahead using the delayed source rates. The source rates are regulated by the proposed fuzzy logic controller which has learning capabilities. The stability of the overall control system is analyzed and the performance of the proposed mechanism is evaluated.