The Research Of Localization Algorithm On Wireless Sensor Network Three-dimensional Surface

In recent years, as wireless sensor networks be used in military and civilian more widely, people are more concerning the research in wireless sensor networks. And the node positioning technology is one of the most important base technology of WSN’s application, The study of node positioning technology also be more widely, it’s applicable scenario has extended from two-dimensional to three-dimensional surface plane, and from the based of fixed anchor node to the mobile anchor node. Among of these, the three-dimensional surface positioning is the most closest to the practical application of wireless sensor networks, but the researchs about it are relatively few, and those research always lack of consideration to the actual complex terrian, such as the impact of the convex or concave surface to the positioning errors and node localization rate.This paper proposes the impact of three-dimensional surface topography to node localization based on the two-dimensional localization algorithm, The main impact to nodes and node positioning accuracy rate are the rugged terrain and the impact of uneven distribution to node localization anchor nodes and node energy consumption. To solve these problems, This paper first presents a hierarchical decomposition positioning algorithm of the three-dimensional surface irregularities. Firstly, it hierarchical the network based on the height, In order to reduce the iterative positioning error in the relatively larger application scenario. Then subnetting within each layer according convex or concave node, and the nodes with similar degree and adjecented each other will be divided into a subnet. In the end, locate the coordinates of the nodes in each subnet and merged into a whole network.The algorithm will improve a lot in locating and positioning error rate than the rest of the algorithm, but the computational complexity and power consumption increase is small. So this paper also proposed to locate the unknown node with the moving path planning based on critical nodes, the algorithm is improved based on the convex decomposition positioning algorithm in the three-dimensional surface. Mainly includes the following aspects :(1) The algorithm maps the network nodes to a two-dimensional plane, it will determine whether the network needs to be divided by make out if there are some areas be coincided, Divide the network into multiple subnets if they are coincided in order to improve the rate of node localization.(2) The algorithm will find all convex or concave nodes as key nodes based on the definition of the concave or convex node of two-dimensional plane in the subnet. And locating key node as known node using a mobile anchor node, then use the known node localize the rest of unknown node in order to avoiding the influence of convex node on positioning error.(3) The algorithm use mobile anchor nodes to replace the anchors, that can be adapted to a variety of complex terrain in order to avoiding some nodes cannot be positioned and positioning error iteration more larger on account of the density of anchor nodes are not enough or deployed uneven.In the end, this paper will show that both of them have much improvement on positioning error, positioning rate and energy consume of nodes based on experiment. And the second algorithm has significantly improved than the first one in all respects.