Intelligent Optimization Algorithms And Their Application In ATM Network Flow Control

Integrated Service Digital Network (ISDN) is considered as the tendency of the communication technique in the future. Asynchronous Transfer Mode (ATM) is adopted as a typical technique by the International Telecommunication Union (ITU). The loss of data and the congestion of information are main factors that affect the quality of service. Available Bit Rate (ABR) service is the only type of traffic that can be controlled by the feedback mechanism. The control and management of ABR service have become a focus and attracted a lot of people in the control and communications field. The research on congestion control of ATM network has great significance in both theory and practice. To control ATM network better and solve the congestion problem of controllable flow, the following research works have been done:Firstly, based on the feedback mechanism of ABR service, the model of single bottleneck node is given. Based on this, the PID control method is introduced into the network control and the PID controller is designed. Furthermore, the sufficient condition of the stability of closed loop is also given.Secondly, to apply PID controller better, the Genetic algorithm (GA) and Particle Swarm Optimization algorithm (PSO) have been used to tune the parameters of the PID controller. The controllers based on the optimization control ATM network flow better, reduce the probability of congestion and improve the utilization of network resource.Thirdly, to overcome the disadvantage of easy trapping into local optimal value for the general PSO algorithm, two modified methods of PSO algorithm are presented. One is to increase the diversity by change the moving direction of a part of particles; the other is to combine the algorithm with chaos. These two modified PSO algorithms have strong abilities to search global optimal value, and are able to overcome the shortcoming of the general PSO algorithm which is easy to trap into local optimal value.At last, the PID controller tuned by intelligent optimization algorithm proposed above is applied to explicit ABR service congestion control. This controller restrains oscillation in a certain extent, improves steady and dynamic performance of system, and makes the system have some robustness performance at the same time.