The Node Localization Algorithm In Wireless Sensor Network

Nowadays, Wireless Sensor Network acts as a hotspot in the research of the network model. Although it is in a stage of academic theory research, its widespread application still has a lot of efforts needed to be made, but not only from the academic research level but also from specific practical application, the value and prospects of WSN are consistent. Sensor network has its distinctive characteristics of interdisciplinary research, involving the sensor technology, technology of network communication, wireless transmission technology, technology of embedded computing, distributed information processing technology, microelectronics manufacturing techniques, technology of the software programming, etc. The WSN research has a lot of branches:time synchronization technology in WSN, technology of network video in WSN, technology of the node deployment in the network, dynamic management of WSN technology, research on routing protocol in the WSN, congestion control research involving the WSN, localization of node technology in WSN, etc. In this paper, the research is based on the node localization algorithm in wireless sensor networks.Sensor nodes in the WSN have some characteristics:small volume, low cost which caused the node itself some restrictions on the manufacturing design:the power supply the total energy, their limited own communication ability and Node limited computing and storage capacity, etc. So to realize the node location, how to minimize energy consumption in the network communication should be considered at the same time, data transmission in the entire network processing nodes should consider their own practical ability. The localization of node algorithm based on WSN is mainly divided into two broad categories:Range-based localization algorithm and the Range-free localization algorithm. The frequently used Range-based method containing:AOA(Angle of Arrival)、TOA(Time of Arrival)、TDOA(Time difference of Arrival)and RSSI(Received Signal Strength Indicator). Relative to the Range-based localization algorithm, the Range-free is more appropriate for the practical application environment. In all aspects, the limitations in wireless sensor networks are as follow:network bandwidth, energy storage, memory ability, and hardware overhead cost, etc. As a result of the existence of these restrictions, the Range-based localization algorithm is impractical in some application scenarios. So, although the accuracy of Range-free localization algorithm is worse compared with the Range-based, but it still has its own research and practical application value. There are some classic Range-free algorithms:Centroid algorithm, DV-Hop algorithm and APIT algorithm etc. These will make introduction in this paper. In this paper, the main research work is about the Range-free localization algorithm.The work in this paper is as fllows:the background history of WSN is briefly described at first. Then the positioning problem in wireless sensor network (WSN) has made an explanation. For the two current research hotspot questions in WSN node localization algorithm, we have carried on the classification of instructions. In this paper, the main work is the Range-based localization algorithm. After explaining several classical localization algorithms, in this paper, we mainly discuss DV-HOP algorithm and location algorithm based on Voronoi diagram in-depth. After learning the two major problems of the original algorithm, through their own work and the corresponding simulation experiments, advanced algorithm are proposed.And through the relevant experiments, we prove that the advanced algorithm is more feasible, and in the related test scenarios, improve the positioning precision of localization algorithm better. Innovation of this paper is as follows:1. In the third chapter of this paper, I make a deep research about the classic DV-HOP positioning algorithm. The third step we put the original algorithm, establish the coordinates of the distance equation set, we are different from the original algorithm to the operation of the equation square on both sides, to do cross by subtracting any equation, then introduced through parameter replace the K value. Using the K value on the relationship between the unknown nodes coordinate to modify the original estimate node coordinates. When doing some simulation experiments, we prove the feasibility of the improved algorithm, based on the original classic algorithms, and verify the improved accuracy of at the localization algorithm.2. The thought of the Voronoi diagram in computational geometry is introduced into the study of localization algorithm. In order to reduce the range of position, we classify the regional of network at first. And then through the numerical sequence from unknown node to anchor node, we could get a minimum value of jump number. We think this value corresponding to the anchor node is the neighboring anchor nodes, and then we can determine the division of the unknown node area. Finally we can estimate the location of unknown node with the centroid algorithm. Although this algorithm is not applicable at some network topology. But in some the network environments, experiments have proved the advantages of the positioning performance of this algorithm.