Research On Localization Algorithm Of Underwater Acoustic Sensor Network Assisted By Autonomous Underwater Vehicle

In recent years,underwater acoustic sensor network has been widely concerned with the continuous increase of underwater exploration.Underwater node localization is a hot topic in current research,and location-based applications have a good prospect.However,due to the complexity of underwater environment,the node localization error is relatively large.Therefore,improving the localization accuracy becomes the main problem of UASN node localization.In order to solve this problem,this thesis proposes two underwater node localization algorithms for underwater acoustic sensor networks.The main work of this thesis is as follows:(1)In order to improve the localization accuracy of nodes without additional hardware,this thesis proposes an improved CWDV-Hop algorithm based on DV-Hop.The algorithm is mainly divided into three steps.Firstly,in order to make full use of the distance information between anchor nodes,this thesis defines the concept of distance-weighted factor and average hop distance of single-node for the first time,and then modifies the average hop distance of anchor nodes.The distance information between anchor nodes is known,and the influence of extremely curved path on the average hop distance can be reduced to a certain extent.Secondly,the hyperbolic localization algorithm is used to instead of the least square method to estimate the location of unknown nodes,which reduces the coordinate loss caused by the equation subtractionin of least square algorithm.Finally,the coordinates of unknown nodes are further optimized by the chicken swarm optimization.By setting the fitness function and corresponding parameters,the optimal location of nodes can be obtained by updating the individual location of chicken groups.The simulation results show that CWDV-Hop algorithm is superior to the comparison algorithms in different parameters.The localization accuracy of CWDV-Hop algorithm is improved by about 53.6% compared with traditional DV-Hop algorithm.(2)In order to reduce the network energy consumption and localization error,and improve the coverage of underwater acoustic sensor network,an autonomous underwater vehicle assisted localization algorithm named KVDV-Hop is proposed.This algorithm is divided into three steps.Firstly,after fully considering the distribution of nodes,the idea of clustering is used to determine the location of virtual anchor nodes.The typical K-means algorithm is improved and it can automatically give the center and number of clustering.Secondly,the underwater robot has the ability of precise navigation and localization.It can be regarded as a mobile anchor node,and broadcasts the packets of virtual anchor node during the moving process to assist the node localization of underwater acoustic sensor network.In this thesis,the path planning problem based on the underwater robot is transformed into the traveling salesman problem,and the ant colony algorithm is used to determine the optimal path of the underwater robot.The movement of underwater robot not only reduces the energy consumption of the network,but also improves the network coverage.Finally,the unknown node can determine its own coordinates by receiving virtual anchor node information.Matlab simulation results show that the algorithm is 41.5% higher than DV-Hop algorithm on the aspect of localization accuracy.