The Improvement Strategy Of Localization Algorithm To Sensor Network

With the advances in Micro-Electro-Mechanical Systems, wireless communicationsand low power embedded technologies, wireless sensor networks have been developedgreatly. Specially as the core for Internet of Things, the wireless sensor networks havebeen widely used. Localization technology which is one of the key techniques takesgreat significance on the efficiency of the application of wireless sensor networks. Thecorresponding action can be taken only when the position information of the event isknown, which can be implemented by knowing the position information of all thesensor nodes.This thesis firstly describes the background and the significance of the localizationin wireless sensor networks. Then it gives an overview on the current state of the art.Moreover, some classical range-free localization algorithms and systems are introduced.All of the above are based on reading a large amount literature at home and abroad.The error of the unknown nodes is large by using the average hop distance which iscomputed by taking some of beacon nodes into account in traditional DV-Hop algorithm.For the problems in traditional DV-Hop introduced above, a node deployment strategyis proposed in the algorithm earlier period as DV-Hop algorithm in the application ofnetworks where nodes are uniformly distributed; And a distance correction strategy isproposed in the algorithm medium period since the average hop distance has differentinfluence on the distance computation between hops; Hybrid particle swarmoptimization is proposed in the algorithm later period as trilateration or multilateraltionscomputation unknown node coordinate Low Precision.Simulations are made by Omnet++and Matlab to verify the efficiency of theimproved algorithms. The results of experiments indicate that the accuracy of the threeimproved algorithms is higher than DV-Hop, especially in Hybrid particle swarmoptimization, which can fulfil the demand of practical applications.