Based On The Pso Algorithm For Wireless Sensor Network Coverage Optimization Research

As the "Tip Nerve" of "Internet Of Things",Wireless Sensor Network is a wirelss、 self-organized、intelligent、information system which synthesize information sense and collection、management、transmission. Monitoring one of target areas is one of its typical applications. So it can collect some physical information that is useful to human beings. In fact,people cannot monitor the area in efficitive with correctly arranged except the user only can spread the sensor nodes in the target area and all of the nodes self-organize into network. With the limitily of the sensor nodes that comprise the basic element of WSN cause the target area cannot be coveraged completely. So if you want to use the WSN in widespread, the rate of coverage of the network should be guaranteed,which the question will affect the performance of the network and the quality of the service. WSN is a group-ment of network. Thus the coverage control problem obviously is provided with the group、 self-organizing feature. Therefore this paper is in order to provide some optimization scheme for the coverage control issue in the basic of group-intelligent algorithms. Here are some achieve jobs as follows:1、We analyze the theory of the Global Particle Swarm Optimization (G-PSO)and Local Particle Swarm Optimization(L-PSO), summarize the procedures、steps of the two algorithms, and design coverage control optimization experiment and simulate them in the environment of2.30GHZ and Matlab-R2012a (other experiments have the same condition).We obtain the results of the test and resolve the factors in G-PSO and L-PSO. Which this results can give us some useful proves for the methods to improve the performance.2、Aiming at the situation of the worst swarm in Global Swarm Particle Swarm Optimization, We propose the (The Worst Challenge Global Particle Swarm Optimization: WCG-PSO)algorithm. By the contrast in the same condition, it is proved to be more available.3、In consideration of the effections in the neighbors of the Local Swarm Particle Optimization,we present the Glowworm Local Particle Swarm Optimization (GL-PSO). we bring the attraction and lightness into the way of achieving neighbours. Under the same experiment, The GL-PSO achieves better results than the Local Particle Swarm Optimization’s.4、We propose the mixed algorithm of G-PSO and GL-PSO:GLG-PSO and compare these algorithms under the same experiment’s params. The GL-PSO is proved to be the best of all.The performance of the GLG-PSO isn’t better than the GL-PSO, but the mixed thought provides some genetics to the next advanced algorithm.5.. We realize the performance of the advanced Particle Swarm Optimization and combine it with the Shuffled Frog Leaping Algorithm-SFLA called GLPSO-SFLA. Because the SFLA has the global search and deeply local search. So we make full use of the SFLA’s strategies and utilize the GL-PSO to improve the learning ability of neighbours in the group of frogs. In the end, we compare the GLPSO-SFLA with G-PSO、WCG-PSO、L-PSO、GL-PSO、 GLG-PSO advanced algorithms.And we achieve the superiority of the new blue-print.It improves the rate of coverage in the rand spread Network which is self-organized with sensor nodes.