Distributed Mobile Node Localization Algorithms In Wireless Sensor Networks Research

As a new information acquisition and process platform, wireless sensor networks (WSN) can achieve complex large-scale monitoring and tracking tasks in the applied areas of poor environments and inaccessible terrains, so the nodes for localizing their own positions play an important role in the network and it is of great practical value. This thesis focuses on the study of range-based localization algorithm for mobile WSN based on the analysis of range-based approach and range-free approach.First of all, based on the study of a lot of related research literatures, the thesis sums up localization techniques and positioning systems for WSN at home and abroad. It introduces the performance evaluation standard and classification methods of localization systems and algorithms for WSN, and then, it analyses and discusses the principles and characteristics of recent representative localization algorithms.Secondly, this thesis proposes a high coverage rate of distributed mobile localization algorithm in WSN with the combination of "signal strength difference of leading beacon algorithm" (SSDLB) and "motion prediction localization algorithm" (MPL). Sensors are deployed with random uniform in two-dimensional applications, so, beacon nodes are fixed, while the unknown nodes randomly shift with random walk mobility model. During the process of positioning, beacon nodes send messages to the network and the unknown nodes receive messages sent from neighbor beacon nodes. Then, according to the number of neighbor beacon nodes (N), it uses different localization algorithm: when N>3 and N=3 (not collinear), it uses SSDLB algorithm; when N<3 and N=3 (collinear), it uses MPL algorithm. So, this method can not only reduce the computational error and cost caused by the traditional RSSI localization algorithm in the positioning way of transforming signal strength into distance, but also greatly improve the positioning accuracy and coverage of nodes. In order to improve positioning accuracy further, this method will choose the greatest received signal strength's beacon node as the leading beacon node in SSDLB algorithm and use Newton's two-dimensional interpolation to computer the previous moving speed and direction of the unknown nodes in MPL algorithm, which can make more accuracy of nodes position predicted.Finally, this research takes a series of simulated experiments by Matlab to verify the reasonableness of this kind of algorithm, and analyses the influence of various factors on the performance of localization algorithm. Therefore, the experimental results show that this algorithm can achieve a higher positioning accuracy and positioning coverage rate in condition of lower density of beacon node without requiring any additional hardware devices, and this positioning coverage rate can even be high to 100% with localization time process. So, to some degree, compared with traditional RSSI algorithm, it can greatly reduce the cost of communication and computation as well as save the network cost and improve the localization performance.