The Research On Node Positioning Technologies In Wireless Sensor Networks

Wireless sensor networks (WSNs) are generally used to monitor and collect varieties of characteristic where the networks cover. With the rapid development of software and hardware of WSNs, recent years, WSNs are translating from academic research to actual use. WSNs have a broad application prospects and can be widely used in many application areas, such as industry and agriculture produce, environment monitor, military affairs and so on. In all these applications, node localization is of extreme importance. Without the position information, the data collected by WSNs will be meaningless. In addition, node localization is also the foundation of WSNs' other theories, such as location-based route and tracking. Therefore, study on the node localization is of great significance to WSNs' theory research and actual use.The thesis first makes a brief introduction to the conception, characteristic and architecture of WSNs, and then mainly focuses on the fundamental algorithms of node localization and their basic ideas. On the basis of summing up the advantages and disadvantages of various algorithms, the thesis proposes an Environment Aware Node Distance Measure Method (EAM). EAM uses RSSI to measure distance. In EAM, the whole covering area of the sensor network is divided into several sub regions by the relationship of beacon nodes, and then EAM can figure out the parameters of the propagation model in each sub region with the cooperation of beacon nodes. In each sub region, EAM uses the corresponding propagation model to transform RSSI to distance. In this way, EAM can improve the precision of the distance measure. In order to verify the performance of EAM, we carry out a simulation in Matlab. Simulation results show that, comparing with the common methods, EAM can effectively improve the distance measure accuracy about 15%-30%.In EAM, the unknown node selects the corresponding sub region's signal propagation model as its signal propagation model. In the process, the region selection of unknown node has a considerable influence on the distance measure accuracy and localization accuracy. Because of this, this paper gives four region selection methods and makes an analysis of region selection in mathematical methods. Finally, the thesis gives a general principle of region selection.Among the localization algorithms using RSSI,obstacles in the networks have a considerable influence on the distance measure accuracy. By proposing the EAM, it becomes possible to analyze the obstacles in some special cases. In this paper, we analyze the obstacles in three cases. Through the analysis of obstacles, we can further improve the measure accuracy of the EAM approach.At last, the thesis implements an EAM-based node localization system and discusses its design in detail. We analyze and compare these two systems' localization error in the lab environment, which use EAM and common distance measure method. Experiment results show that the localization accuracy can be improved about 20% by adopting EAM-based system. And the average location error is less than 30%, which can satisfy most location requirements in wireless sensor networks.