Research On Localization Of Mobile Wireless Sensor Network In Local Coordinate System

Wireless sensor network(WSN)is a cyber-physical system which integrates wireless communication and microelectronic mechanical systems.It has a broad application prospect in many fields such as military,environmental monitoring and industrial control,etc.In many applications,the location information of sensor nodes is of great significance,so localization technology and localization algorithms are important contents of WSN research.Traditional stationary sensor networks have disadvantages,for example,the localizability of network depends on the early deployment,the inability to dynamically allocate resources.With the rapid development of new energy and image recognition technology,mobile WSN has obvious advantages in target tracking,resource allocation,task scheduling,etc.Therefore,the localization algorithms for mobile WSN have been researched by numerous scholars.In this thesis,the distributed localization of mobile WSN nodes based on bearing-only measurements in local coordinate system in two-dimensional space is studied.The localization problem of mobile nodes is transformed into the conversion relationship problem between fixed coordinate systems in this thesis.On this basis,a distributed solution is designed for the cooperative localization of mobile free nodes and the self-localization of isolated free nodes in the case of a single mobile anchor node.the main content includes the following parts:(1)This thesis studies the cooperative localization problem of mobile WSN,puts forward the sufficient condition for free nodes to localize each other in local coordinate system and fully proves it.After that,the cooperative localization method is further analyzed.Based on these,the gradient descent algorithm for free nodes to localize each other and the distributed cooperative localization algorithm of WSN are designed,and the convergence of the algorithms is proved.Finally the effectiveness and robustness of the cooperative localization algorithm is verified and analyzed by numerical simulation.(2)The localization problem of the sensor nodes without neighbor free nodes is deeply analyzed in this thesis.An iterative self-localization algorithm was designed,and the convergence of the algorithm is proved.On this basis,the neighbor information in the case of neighbor free nodes is added to the self-localization algorithm,so that the algorithm can realize the self-localization regardless of whether there are neighbor free nodes.Finally,under different communication and measurement topologies,the simulation experiments were designed to compare the localization performance between self-localization algorithm and cooperative localization algorithm.The robustness and effectiveness of the self-localization algorithm are verified and the differences between the two algorithms are analyzed by simulation.