A neural network based ATM call admission controller for multiple service classes with different QoS

This paper proposes a new approach to adaptive call admission control based on a neural network for multiple service classes with different quality of service (QoS) in the ATM-based broadband integrated services digital networks. We extend Hiramatsu's (1990) neural network based on-line controller for the single QoS by constructing multiple pattern tables based on each service's acceptance or rejection at the call set-up requests. We consider call admission control of two service classes and employ two different cell loss rates as QoS for two service classes having different traffic characteristics. The cell loss rate for each service class is simultaneously controlled by considering the target cell loss rate of each class and the trunk capacity. Computer simulation results show the effectiveness of our adaptive call admission controller for two service classes with different QoS.