Improved DV-Hop Localization Algorithm For Wireless Sensor Networks

Wireless sensor network, consisting of numerous micro-sensor nodes (hereinafter nodes) deployed in the monitoring region, is a self-organizing multi-hop network system which is formed through wireless communication. The nodes collaborate with each other to perceive, collect and process the information of the perception object in the network coverage areas and send it to the observer. The wireless sensor network, according to its own characteristics, has a wide range of application prospect in environmental monitoring, military and high-risk areas. In the wireless sensor network, the nodes'location information is essential to the sensor network monitoring activity. The position an incident occurred and the nodes'location information we accessed are critical information contained in the sensor node monitoring messages which will become meaningless without nodes'location information. Therefore, the research on the node location algorithm of the wireless sensor network is of great theoretical and practical significance.Considering for the costs, it is impossible for every node of the wireless sensor network to have the precise GPS location capability. Generally, in practical wireless sensor network applications, only a small number of nodes have the precise GPS location capability, whereas the remaining nodes estimate its position through specific location algorithms. Based on the classic self-location algorithm DV-Hop (distance vector-hop) algorithm in the wireless sensor networks, this paper proposed an algorithm which makes an improvement on the estimation of the distances between adjacent nodes as well as the calculation of the distance from the target node to the node anchor it had done.This thesis proposed an improved algorithm based on the DV-Hop (distance vector-hop) algorithm which is a classic self-location algorithm in the wireless sensor network. The improved algorithm computes the distance of unknown nodes and beacon nodes, using the angle∠ABC formed of three neighbor nodes A, B and C in the path of unknown nodes and beacon nodes, and given distance data of among the nodes. This angle∠ABC is estimated with an overlapping degree of B'neighbor node sets collecting with A and C, respectively. The experimental results show that the improved algorithm appears to be more accurate in the location compared to DV-Hop algorithm.