Network Congestion Control Method Based On Genetic Immune Particle Swarm Optimization

At present, with the rapid and extensive development of Internet, network congestion slowly become a important factor that restrict the development of Internet, the network appeared more and more with the QoS requirements of the services, and the services set a higher request to the QoS, therefore, in order to promote netwok healthy and stable development, ensure the normal work of other QoS mechanism, it is a important direction in the current network research by appropriate ways of prevention and control of congestion, also is homework to do. This thesis is given in detail the analysis of network congestion control method based on hybrid genetic immune particle swarm optimization, the main contents are as follows:（1） Proposed the idea of solving the network congestion control problems by genetic immune particle swarm optimization algorithm on foundation of analysis of the network congestion and the network congestion control mechanism. Finally we analyse the simulation tool:NS₂and its application to network modeling and congestion control.（2） This paper analyzes the QoS routing algorithm of quality of network service detailedly by analysis of several types of topology model, an optimization mathematics model of QoS routing, which provides the conditions for network routing optimization, network congestion prevention and control, is proposed on the basis of analyzing the QoS and QoS routing.（3） Genetic Algorithm（GA）and Immune Algorithm （IA）is introduced into Particle Swarm Optimization algorithm（PSO）, which a genetic immune particle swarm optimization algorithm is given. Introducing crossover and mutation mechanism of genetic algorithm, and identifying and selecting idea of immune algorithm, improves the probability rate of particles’ fitness, and ensure that the diversity of particles will not be affected, so that the new individual can not only jump out of the local advantages but also can keep excellent individual characteristics, to avoid the excessive concentration of the individual. （4） Through the analysis of the parameters of the network path and the index of the path optimization, we proposed a network congestion control method based on hybrid genetic immune particle swarm optimization. The method plans the path of network load, which makes network load balancing and resource consumption as aim function to avoid network congestion by balancing the load and minimizing network resource consumption to meeting multifold indexes of bandwidth and delay. The results of simulation shows the effectiveness and the reliability of this method.