Wireless Sensor Network Node Localization Algorithm

The information of node's location in wireless sensor network is vital, and it is the premise of the network applications in environmental monitoring, target recognition and tracking, also node localization is the core technology of wireless sensor network, so the study on technology of node localization is very meaningful. With the wireless sensor network technology advances, it has a wide application in real life, the research for more anti-jamming,more simple algorithm and less power consumption node localization technology has an important theoretical and practical value.This paper introduces the core supporting technology of wireless sensor network, research status of node localization at first, and then introduces the theoretical basis of node localization algorithm.At the same time,the positioning algorithm is classified from many angles, performance evaluation and several important factors which affects performance are analyzed briefly. Finally, after focused on analyzing the two typical localization algorithms-the centroid location algorithm and DPE location algorithm, we pointed out the existing problems.Based on these problems, two improved methods are proposed. Algorithm simulation shows that the improved algorithm obviously has a better performance than the original algorithm, and the main tasks of this paper as below:Firstly,the paper proposed three improved Centroid Algorithm.The centroid localization algorithm based on non-range postioning and has the advantages of low algorithm complexity and high positioning precision. But the algorithm has drawbacks of requiring more anchor nodes,more caculation and have high cost.Considering this defect, this paper proposes there improved ideas-artificially laying out partial initial anchor nodes,weighted centroid algorithm and iterative weighting method. Because localizition accuracy of node in edge position through centroid localization algorithm is not high, in initialization, an anchor node is laid in each corner in the distribution area in order to improve the precision. Moreover, in the centroid algorithm, due to the closer position result in greater impact of computing nodes, so giving each node the weight can reduce the error.For the node density is low in the network, upgrading the node which has been positioning can solve the sparse anchor node problem. When upgraded anchor node in positioning the unknown node, its weight should be lower than the original anchor,and thereby reducing the cumulative effect of errors. By successive iteration loop the algorithm can locate all the unknown nodes.The simulation show that the algorithm remains higher positioning accuracy under the condition of low anchor node density. Secondly, the author make an improvement to the DPE algorithm. DPE algorithm is a localization algorithm based on range positioning, it belongs to the incremental algorithm, common algorithm requires at least three anchor node in node positioning, DPE algorithm can be used when a node has two anchor nodes around itself. The improved DPE algorithm algorithm is mainly improved by restricting the incremental direction to solute the problem of multiple solutions while only two anchor nodes. This improved algorithm uses triangulation method to measure the angle between the nodes and it combines the distance to calculate the coordinates of unknown nodes. Overall,it needs fewer anchor nodes and low cost, it can be used to solve the problem that the node in edge area is unable to locate or has a low positioning precision. The simulation shows that the improved algorithm positioning accuracy is relative taller.