Research On Node Localization Optimization Algorithm In Wireless Sensor Networks

Wireless Sensor Networks(WSNs)is a system composed of intelligent sensor nodes deployed in the monitoring area,which has broad prospects in many fields.Node localization technology is one of the important technologies in WSNs.In order to provide the events information,the nodes need to locate itself in advance.Therefore,node localization technology has great value.This thesis focuses on the node localization algorithm in WSNs and proposes two localization optimization algorithms.Distance Vector-Hop(DV-Hop)is a range-free localization algorithm,which is simple to use and has attracted much attention.Aiming at the problem of low localization accuracy caused by the error due to hop counts and hop distance,this thesis proposed a new algorithm topological similarity distance vector hop algorithm(TSDV-Hop),which used topological similarity to correct the hop counts.Firstly,TSDV-Hop defined topological similarity.A theorem that the distance between two nodes was inversely proportional to the topological structure similarity was proved.Then a hop-correction equation which could correct the hop count between neighbor nodes was designed.Secondly,in order to make the hop count between nodes more accurate,TSDV-Hop introduced the hop-correction parameters and the best hopcorrection parameters were obtained through simulations.Finally,the levy adaptive improved bird swarm algorithm was utilized to optimize the coordinates of unknown nodes by constructing a new objective function fitness.TSDV-Hop corrected the drawback of large hop error in DV-Hop algorithm without additional hardware.Simulation results show that TSDVHop has better localization accuracy.In mobile anchor node location,one or several anchor nodes are utilized to move along the planned path in the monitoring area to locate unknown nodes.Aiming at the problem that the existing path planning methods do not fully consider the distribution of nodes in the network,resulting in low localization efficiency and lengthy paths,this thesis proposed a new localization algorithm path planning for a single mobile anchor node(PSMAN)based on the dynamic path planning of a single mobile anchor node.Firstly,PSMAN introduced the topological similarity of TSDV-Hop algorithm to improve the hop count between nodes.The improved hop count between nodes was used to define the average hop value which reflected the density of nodes.The nodes were classified by the relationship between nodes and the average hop value.The table of class center node ID was obtained.Then the mobile anchor node traversed the network along the class center node table,and the optimal mobile scheme was obtained by ant colony algorithm.Finally,in order to solve the flip ambiguity caused by the coplanar virtual anchor nodes,a path planning scheme combining straight lines and broken lines was proposed.And the node localization was completed using received signal strength indicator(RSSI)algorithm and the least square method during the node movement.Experimental results show that the PSMAN algorithm has great localization performance.