Research On Localization And Matching Algorithm For Wireless Sensor Network

A wireless sensor network is a multi-hop ad hoc network,composed of a large number of sensor nodes with sensing,computing and processing capabilities.As a new technology for acquiring and processing information,wireless sensor networks are widely used in various fields.In many applications of wireless sensor networks,node localization technology is critical because the information collected by the sensor nodes is meaningful only when the location information of the nodes is accurately known.Therefore,it is self-evident to study how to improve the accuracy of the positioning algorithm in order to obtain accurate node position information.In addition,the requirements for localization technology are different due to different application requirements.Some applications need to match the estimated coordinates to the real coordinates in the case of a given node's true position.In this type of application,the accuracy of the matching is not only related to the accuracy of the positioning algorithm,but also related to the corresponding matching method.The higher the positioning accuracy,the better the matching method and the higher the matching accuracy.Therefore,an effective localization matching method is needed to improve the correct matching rate between the estimated coordinates and the real coordinates.In view of the above problems,this thesis proposes a better performance non-ranging positioning algorithm and a localization matching method,and analyzes the performance of the proposed method through simulation results.The specific research content is as follows:Firstly,a range-free localization algorithm NIHES based on hop is proposed.The corresponding coarse-grained quantization method NIHES-C and fine-grained quantization method NIHES-F algorithm are given.By analyzing the error introduction factors of DV-Hop algorithm,it is found that the error mainly comes from the error of the unknown node to the anchor node hop count estimation,and secondly because the distance estimation between the unknown node and the anchor node is biased,the farther away from the anchor node,The larger the distance estimation error is,and the multilateral localization method is more sensitive to the distance error,so these two factors reduce the positioning performance of the algorithm.Therefore,a hop quantization algorithm NIHES is proposed,and two methods of quantization are given.At the same time,a localization calculation method based on hop-based weighting is proposed.The experimental results show that the proposed algorithm can improve the localization performance,compared with DV-Hop,Improved DV-Hop,Multi Range DV-Hop and Virtual Hop algorithms in four different network topologies: random uniform,uniform mesh,hole type and C type,which proves the effectiveness of the proposed algorithm.Then a localization matching method based on nearest neighbor and bipartite graph is proposed,which is a process of localization and matching iterative loops.In this thesis,the NIHES-F algorithm is used to implement the node localization process,and then the nearest neighbor and bipartite graph methods are used to match the estimated position with the pre-installed actual position,and the intersection of the matching result sets obtained by the two matching methods is regarded as credible.The trusted node is upgraded to an anchor node to participate in a new round of localization matching.The iterative process ends when the adjacent two rounds of matching results are the same,and the result of the last bipartite graph matching is taken as the final matching result.Experimental simulations show that 98% of the location points are correctly matched in the scene with even nodes,and 89% of the location points are correctly matched in the non-uniform scene.