| Particle Swarm Optimization (PSO) is a new kind of evolutionary computation and was originally introduced by Eberhart and Kennedy in 1995. It has been proven to be a powerful global optimization method. PSO has been widely applied in function optimization, neural network training, etc. But PSO is a newly emerging optimization method, and there was more research work that should be substantiated. So it is very significant to seek more powerful and improved algorithms based on PSO to solve concrete engineering problems.As for my work and academic achievements, it is composed of two subjects. The first subject is that net model is converted to mathematic model on the basis of analyzing the problem of QoS routing in the Ad Hoc network, and simulated by basic PSO.The experiments have shown that the algorithm has great probability & effectiveness.This content is reflected in the 2nd, 3rd and 4th chapter. The second subject is the Adative and Asymmetric Fully Informed PSO (AAFIPSO) applied to QoS routing optimization in Ad Hoc network,reflected in the 5th and 6th chapter.First, some basic Ad Hoc network conceptions are summarized in this thesis, and some hot points and issues in this area are concluded, then some study points in this thesis are given. A simulation and evaluation scheme is put for test and evaluation expediently, and some simulation models and their adaptability were expounded.It is a hotspot and a difficulty to put forward PSO in the light of the topological structure. In view of the shortcoming that weighting is fixed of Asymmetric Fully Informed PSO, the AAFIPSO is proposed. The adaptive inertia weighting function, which reflects the exploration and exploitation of the particle during the evolution, is rebuilt. And AAFIPSO is used in routing optimization for Ad Hoc Network based on link existent time. The experiments have shown that the algorithm can converge faster than traditional methods, and had better foreground in Ad Hoc network for routing. |
PDF Full Text Request