Research On Localization Algorithm For Wireless Sensor Networks Based On Local Topology

Wireless sensor network(WSN)involves multiple technologies including wireless communications,sensor technology,distributed information processing,embedded technology and microelectronics,and is widely used in various fields of transportation,military,medical,environmental protection,etc.During application,determining the location of an event is one of the key issues that call for solution after an event is detected.Apart from determining event location,location information also has functions such as target tracking,network management,routing,etc.Therefore,designing an efficient WSN location algorithm is an indispensable part of wireless sensor network management.This paper studies on the localization algorithm of wireless sensor networks.Main research works of this paper are:(1)By respectively analyzing and summarizing the advantages and disadvantages of WSN location technology from three aspects: machine learning,ranging and non-ranging,it provides a firm foundation for designing WSN localization algorithms with high precision and low power consumption.(2)Based on the research of LE-LPCCA localization algorithm,a locality preserving distributed localization algorithm LE-DLPCCA is proposed by introducing local topology and distributed features.Simulation results show that when the training sample ratio is 70%,the positioning accuracy can reach 86%,and the energy consumption is greatly reduced at the same time;meanwhile,the lifetime of the whole wireless sensor network can also be prolonged,and the modeling speed is 8 times higher than before.(3)This paper analyzes the layout of WSNs with manifold features,introduces local topology information and non-beacon nodes information,studies the localization in semi-supervision framework,and proposes a mobile localization algorithm based on Laplacian mapping LP-LapRLS.The algorithm not only improves the generalization ability of the predicted model,but also increases the speed of modeling.Compared with similar algorithms,the experiment proves that LP-LapRLS has higher positioning accuracy and stability.When the training set proportion is 60%,the positioning accuracy can reach 84%.(4)Based on the research of wireless sensor network architecture and protocol stack,this paper designs and realizes WSN positioning simulation platform by VC ++ in VS2010 integrated environment.Then the location effects of the LE-DLPCCA and LP-LapRLS algorithms are compared and analyzed.Positioning accuracy of the LE-DLPCCA algorithm is about 2% higher than that of the LP-LapRLS algorithm.However,the LP-LapRLS algorithm is robust to outliers and has the highest modeling efficiency in localization algorithms.