APPLICATION OF NEURAL-NETWORK IN ATM CALL ADMISSION CONTROL-BASED ON CELL TRANSFER STATE MONITORING WITH DYNAMIC THRESHOLD

Calls using different media which require different transfer quality w ill arrive at ATM networks. Therefore it is important to develop a met hod for allocating network resources efficiently to individual calls b y judging admission of calls. Various call admission control schemes h ave been already proposed, and these schemes assume that users specify values of traffic descriptors when they originate calls. However, it is sometimes difficult for users to specify these values accurately. T his paper proposes a new ATM call admission control scheme based on ce ll transfer state monitoring which does not require that users specify values of traffic descriptors in detail when they originate calls. In this proposed scheme, the acceptance or rejection of calls is judged by comparing the monitored cell transfer state value with a threshold prepared in advance. This threshold must be adjusted according to chan ges in the characteristics of traffic applied to ATM networks. This is one of the most serious problems in the control scheme based on the m onitoring of cell transfer state. Herein, this paper proposes neural n etwork application to the control scheme in order to resolve this prob lem and improve performance. In principle, the threshold can be adjust ed automatically by the self-learning function of the neural network, and the control can be maintained appropriately even if the characteri stics of traffic applied to ATM networks change drastically. In this p aper, the effectiveness of the application of a neural network is clar ified by showing the configuration of this proposed control scheme wit h the neural network, a method for deciding various parameter values n eeded to implement this control scheme, and finally the results of a p erformance evaluation of the control scheme. Inputs required by the ne ural network are also discussed.