Predictive congestion control of ATM networks: multiple sources/single buffer scenario

Proposes a predictive congestion control methodology for the available bit rate (ABR) service class in an ATM network for the case of multiple nodes with single buffer scenario. An adaptive controller is developed to control traffic where sources adjust their transmission rates in response to the feedback information from the network nodes. Specifically, the dynamics of the ABR sources with the buffer is modeled as a nonlinear system and an autoregressive moving average based (ARMAX) adaptive controller is designed to predict the explicit values of the transmission rates of the sources so as to prevent network congestion. Stability analysis of the closed-loop system is presented. Simulation results are provided to justify the theoretical conclusions