Research On Localization Algorithm Based On Underwater Wireless Sensor Networks

As one of the important facilities for marine exploration,marine data collection and natural disaster prediction,underwater wireless sensor networks(UWSNs)have been widely used in many related fields.Since the underwater sensor node localization is the basis of UWSNs' application in various related fields,the research of localization algorithms based on UWSNs has become one of the hot research directions in underwater field.The localization of underwater sensor nodes still faces challenges,such as the localization ratio of nodes,time synchronization,localization accuracy,and the mobility of nodes.This thesis carries out research on underwater node localization algorithm from two perspectives of large-scale network application and high-precision node localization.Firstly,this thesis introduces the network structure of the UWSNs with the function of each component,classification method based on UWSNs node localization algorithm,some classic node localization algorithm,underwater node localization challenges and localization algorithm evaluation criteria.Based on the overview of relevant content,two underwater node localization algorithms are proposed in this thesis,which respectively improve the node localization ratio of large-scale UWSNs and achieve high-precision underwater node localization in underwater acoustic propagation layer.A mobility-assisted localization scheme with time synchronization-free feature(MALS-TSF)for three-dimensional(3D)large-scale UWSNs is proposed in this thesis.The underwater drift of the nodes is considered in this algorithm,and the random walk model is used as the motion model of the nodes.In Phase I of the algorithm,anchor nodes within the monitoring area are planned and deployed according to the size of the monitoring area and the initial communication radius of the anchor nodes,and node localization is realized by minimizing the mean square error.The ordinary nodes that have been located in Phase I are converted into new reference nodes through confidence screening,and participate in the localization of ordinary nodes in Phase II.In Phase II,two-TOA algorithm is used to distance the ordinary node and the new reference node.The simulation results show that MALS-TSF does not need time synchronization and has a higher localization ratio.Compared with other algorithms,MALS-TSF reduces the density of anchor nodes,energy consumption of anchor nodes and localization cost of nodes and is suitable for 3D large-scale UWSNs.To realize the high precision localization of ordinary sensor nodes,a high precision underwater node localization algorithm based on particle swarm optimization(HPNLA-PSO)is proposed in this thesis.This algorithm improves node localization accuracy from two directions.Firstly,the acoustic compensation algorithm is used to correct the projection distance between the ordinary node and the anchor node in view of the layered phenomenon of underwater acoustic transmission.Secondly,an improved PSO algorithm combined with genetic algorithm is proposed to to realize high precision localization of underwater sensor nodes.In the experimental simulation,the advantages of HPNLA-PSO algorithm are verified from the node localization ratio,algorithm convergence,particle convergence effect,average error of node localization,average communication overhead and other aspects.The experimental results show that the proposed algorithm has high node localization accuracy for the three-dimensional static nodes under the phenomenon of underwater acoustic propagation layer.