Wireless Sensor Network(WSN) is an important part of the information technology revolution,has been widely used in military, medical care, environmental monitoring, target tracking,and other fields, is an important means of information acquisition.In many practical applications, the wireless sensor network node location information is the basis of its provide various application service. Since the nodes in wireless sensor network are of large size and small volume and energy limited,GPS(Global Positioning System) positoning device,which is of high energy consumption and large volume and high cost is limited when applied to the large-scare WSN.For large-scale wireless sensor network, the main node localization algorithm is divided into two types at present : Range-based node localization and Range-free node localization. Range-based node localization alogrithm first measure of the unknown node to the node distance by Radio Signal Strength Indicator(RSSI) and Time Difference of Arrival(TDOA), and ranging technology, list equations, then solve nonlinear equations by the maximum likelihood estimation method and trilateration method, the final positioning of the unknown node. The other is Rage-free node localization algorithm mainly in the wireless sensing net deployment of special communication protocol, through the network connectivity to estimate the distance between nodes, thus completing the process of positioning. more typical algorithm mainly include Distance vector-hop(DV-HOP) and centroid algorithm, etc.Particle swarm optimization algorithm is a new kind of swarm intelligence optimization algorithm, it is mainly used for solving nonlinear optimization problems, Because of its simplicity, easy implementation,fast convergence speed and advantages of less adjustable parameters, PSO is widely used in nonlinear function planning, path planning problem, Economic forecasting, neural network training areas and so on. But the particle swarm optimization algorithm in solving complex multimodal nonlinear optimization problems, easy to fall into premature convergence. Aiming at this disadvantage, in this paper we propose an improved scheme, named LH-DMPSO(Low_High Dual Mutation Particle Swarm Optimization). With the introduction of the dual mutation factors to follow the best particle and the worst particle, the scheme presented a dynamic mutation strategy based on the average particle distance to maintain the swam diversity, and to overcome the premature convergence issue.In the traditional RSSI-based WSN node localization algorithm, error will be introduced when use RSSI to measure the distance.Traditionally, we first has the traditional error analysis of node localization algorithm in detail, then positioning problem is transformed into a nonlinear optimization problem, the improved particle swarm optimization algorithm LH- DMPSO can be solved for WSN node localization. Finally, tests results shows our algorithm on performance is outperform than the traditional maximum likelihood algorithm and the standard particle swarm optimization algorithm in positioning accuracy. |