Research On Wireless Sensor Network Location Algorithm Based On Particle Swarm Optimization

Node localization technology is not only one of the most important main support technology of wireless sensor network(WSN), but also one of the research difficulties. In most practical applications, it is crucial to acquire accurate absolute positions of nodes or relative positions between nodes. In order to achieve the purpose of positioning precisely, on one hand, the distance evaluated error between nodes or distance-measuring error should be reduced, on the other hand, the localization algorithm should be improved.Firstly, this paper represented the development history of WSN technology and summarized research status of domestic and overseas. Then, range-based technology between WSN’s nodes and range-based localization algorithms were studied and compared. At last, the particle swarm optimization(PSO) algorithm was studied, improved and simulated under the range-based localization model.In this paper, the localization algorithms of WSN based on the PSO algorithm were studied. The main researches are the following two points:Firstly, in WSN with anchor nodes, to solve the shortcomings of traditional RSSI ranging model, this paper adopted an improved RSSI ranging model and proposed a kind of localization method named chaotic PSO algorithm based on weight decreases linearly(W_CLSPSO).The comparison between W-CLSPSO algorithm based on anchor nodes selection strategy and W-CLSPSO algorithm without anchor nodes selection strategy proved that the feasibility of anchor nodes selection strategy in the experiment of this paper. The simulation results of W-CLSPSO algorithm, CLSPSO(chaos particle swarm optimization algorithm) algorithm, PSO algorithm and WLS(weighted least squares) algorithm under the condition of different noise figure and number of anchor nodes show that the W-CLSPSO algorithm’s localization accuracy and stability is superior to the other three algorithms.Secondly, aim at anchor-free WSN, this paper put forward a kind of new localization method cooperated with an improved PSO algorithm and Taylor algorithm based on TDOA localization model. In different noise figure, the average position error and mean square error of PSO algorithm, AsyLnPSO(particle swarm optimization algorithm based on asynchronous learning factor) algorithm, SAAPSO(adaptive weight particle swarm optimization algorithm based on asynchronous learning factor) algorithm and Min-Max+Taylor algorithm, SAAPSO algorithm, SAAPSO+Taylor algorithm were compared respectively. The simulation results show that the improved algorithm proposed in this paper has higher positioning accuracy, smaller cumulative error and better stability.