Research On Localization Algorithm Of Wireless Sensor Network Based On DV-Hop

As a new type of intelligent network with communication,computing and sensing capabilities,wireless sensor network can achieve dynamic monitoring of target area data.Currently,wireless sensor network has been widely used in key areas such as national security,aerospace,precision agriculture,and smart homes.The DV-Hop(Distance Vector Hop)algorithm,as a classical range-free positioning technique,has attracted the attention of researchers since it was proposed.In this paper,DV-Hop algorithm is used as the research object,and different improvement strategies are proposed for the problems in the localization process,respectively,and the main works are as follows:(1)To address the low positioning accuracy of the original DV-Hop,an improved DV-Hop localization algorithm based on weighted iteration and nearest neighbor first is put forward.Firstly,the initial average hop size of each beacon is obtained by replacing the unbiased estimation with the minimum mean square error criterion,and each beacon is assigned different weights according to the per-hop error,and the average hop size is iteratively improved by using the weighted minimum mean square error criterion.Secondly,the calculation process of the average hop size of unknown nodes in the original DV-Hop is deleted,and the inter-node distance estimation method is reconstructed with the average hop size of beacon nodes as the starting point.Finally,considering the influence of beacons with different distances on the unknown nodes,the nearest neighbor first idea criterion is used to locate the beacon nodes in groups,and the potential errors in the solution process are eliminated by the equation rotation strategy.The simulation experiments prove that the proposed algorithm has better localization performance compared with the original DV-Hop and existing improved algorithms.(2)To improve the localization performance of the original DV-Hop,an enhanced DV-Hop localization algorithm based on the spring model and reliable beacon set is proposed.Firstly,the effect of the correction coefficient from the target beacon node to the reference beacon node on the target path is analyzed from the perspective of the spring model in physics.By taking the Angle between the reference path and the target path as the weight,the correction coefficient of the reference path is weighted and summed to obtain the correction coefficient from the unknown node to the target beacon node,and the distance correction between the unknown node and the target beacon node is realized.Then,the RANSAC algorithm is fused to extract the reliable beacons,and the least squares method is used to fit the coordinates of the reliable beacons and determine the optimal position of the unknown nodes in combination with the network connectivity.The experimental results show that the algorithm has good localization accuracy and localization stability.(3)To promote the positioning performance of DV-Hop algorithm in anisotropic networks,an improved DV-Hop localization algorithm incorporating gray wolf algorithm is proposed.In order to solve the problem of large estimation distance error caused by obstacles,the binary gray wolf algorithm is introduced to realize the "0-1" encoding of beacon nodes,and the dominant set of beacons for localization is constructed based on the size of the gray wolf position adaptation value.In addition,the continuous gray wolf algorithm is used to replace the least squares method to solve the set of distance equations composed of dominant beacons to further improve the localization accuracy of unknown nodes in anisotropic networks.Simulation experiments demonstrate that the algorithm exhibits relatively excellent localization performance in all three classical anisotropic networks.