Research On Localization Technology Of Mobile Anchor Nodes In Wireless Sensor Networks

With the wide application of wireless sensor networks, node localization technology has gradually become a hot research. For the application of wireless sensor networks, node position right directly determines the subsequent information acquisition, processing and transmission, estimating the appropriate node coordinates has important practical significance. A kind of algorithm is not likely to be applicable to any scene, therefore on the basis of summarizing the existing achievements, the article researches and analyses different scenarios about the node localization problem, and puts forward the corresponding solving algorithm, specific research content is arranged as follows:(1) On the basis of analyzing the advantages of the mobile anchor node instead of the static anchor node, mobile anchor algorithm is always combined with range-free algorithm in the environment where nodes' ID is known. This article firstly combines single mobile anchor node with APIT positioning algorithm to localize unknown nodes in the sparse environment, and overcomes the drawback that static APIT positioning algorithm should be used in dense distributed environment and has in-to-out false. In this algorithm a single anchor node moves along the preset trajectory in the region, unknown nodes record virtual anchor nodes' coordinate within communication, then exhausting all triangular combinations constituted by virtual anchor nodes, using the angle-sum method to determine the roughly positions of unknown nodes. Finally, the centroid of the inscribed triangle circle represents the optimal coordinate of node.(2) In the actual node positioning process, the number of nodes and ID in the unknown region is often unknown. Aiming at the nodes' localization problem in this environment, this article introduces the RSSI quantization model, and uses the split clustering algorithm to cluster the mixed RSS sample sequence from unknown nodes at resident localization, the number of clusters corresponds to the number of unknown nodes, then grids the unknown region. According to the best RSS value of the virtual anchor node clustering, the circle cross search is carried out, and initial coordinates of unknown node are obtained by using the eight neighbor maximum method. At last the particle swarm optimization algorithm obtains the optimal node coordinates.(3) With the rapid development of the Internet of things, the dynamic sensor networks have been paid more and more attention. Taking it into account that static network localization algorithm should be frequently called and can't meet the actual demand. Therefore, aiming at the problem of localization of mobile unknown nodes involved in the practical application environment, this article designs a mobile anchor node localization algorithm based on Monte Carlo Box. RSSI quantitative model is introduced into this algorithm, anchor node further reduces the anchor box area based on the received RSS value, and the two Newton interpolation method is used to predict the node's moving speed for narrowing the sampling box, and it adaptively estimates the corresponding number of samples. Finally, it uses the idea of crossover and mutation in genetic algorithm to improve the sampling success rate. Experimental results show that the improved MCB algorithm has high positioning accuracy.In this paper, the designed mobile anchor node localization algorithm successfully solved the node localization problem in three different scenarios, and improved the nodes' accuracy based on the original algorithm. The simulation results show that the improved algorithm can accurately locate the nodes, and has a wide range of application prospects.