Research On Target Localization Technology In Wireless Sensor Networks

Wireless sensor network (WSN) is a new kind of network technology. Because of its safety,dynamic, self-organization, it could be widely used in different areas including military,communication, industry, medical and so on. Target localization is an essential application ofWSN. This paper mainly focuses on the research and modifying Maximum Likelihood(ML)inthe model of energy.First of all, we apply the idea of grid information to ML, and research the BML. We use thebinary data for location which could not only decrease traffic load of network but also reduce thecomputational complexity. Next, we analyze the fault tolerance of BML as well as its correctionalgorithm——FTBML. FTBML could reduce influence of error in location for fully accountingof failures and errors in sensors. We also compare the performances of different algorithm underdifferent parameters through simulation.Secondly, we propose the CE-FTBML which is combined by two steps. Firstly, we use CEwhich is of low precision and complexity in whole area. Then, with the radius gained form CE,we utilize FTBML of higher precision in second search. We analyze precision and complexity ofCE-FTBML. By simulation, we could find that CE-FTBML has good precision and lowcomplexity under certain faulty probability of sensors.Finally, we also propose CE-FTBMLe which improves the strength acquisition of targetsignal in CE-FTBML. We could prove that the strength of target signal is highly precise throughsimulation. Hence, we could obtain the conclusion that CE-FTBMLe has a great value inapplication.