Research On Localization Algorithm Based On Mobile Anchor For Wireless Sensor Network

Wireless sensor network (WSN) is an emerging and interdisciplinary researchdirection, which involves many disciplines’s knowledge. WSN has changed the way ofcommunication between people and nature by combining the physical world with thelogic information world. WSN can be widely used, because it is a self-organization anddata-centric network. Only the data binding position information is meaningful. Atpresent, positioning technology mostly use the static anchor node installing of GPSequipment to provide position information for unknown nodes, but the high densitydistribution of anchor nodes will increase the cost of positioning system. Using mobileanchor node instead of static anchor node is an effective method to solve the problem,so the localization algorithm of mobile anchor node is deeply researched in this paper.The positioning algorithm of using mobile anchor node still has manydisadvantages, such as positioning blind spots; the time of positioning implementationis too long. Some studies have proposed the use of incremental localization method, butthe cumulative error is a very serious problem. The author presents a new incrementallocalization algorithm based on meshing to improve the deficiency of the existingmethods; the positioning technology is mainly applied to large-area monitoring area,aiming to solve the problem of slow positioning speed and cumulative error. In order todecease the cost of localization and improve the position efficiency, this algorithmcombined mobile anchor and static anchor. The mobile anchor nodes move by plannedpath and broadcast their location information in the network. Unknown nodes wait forthe command of monitoring background after calculating their coordinates, and thendecide whether they need to convert to static anchor nodes, supporting other nodelocalization. In order to solve the problem of accumulation error from incrementallocalization, this paper optimized the positioning method and changed some knownnodes reasonably into static anchor nodes. In order to effectively reduce the cumulativeerror produced by network location, large-scale network is divided into some grids.According to the precision demand proposed by localization applications, simulationtools can tell us how to divided large network into grids of small area. Each grid isrelatively independent in the localization process.To illustrate the validity of the algorithm, this paper simulated and verified thealgorithm from many aspects, including the relationship of mesh number and position precision, as well as error analysis of nodes localization. According to the simulationresults, reasonable grid size can be selected for different precision demand. Error figureshows the difference of actual position and measurement position. In addition, thisalgorithm and MBAL algorithm are compared in aspects of positioning speed of theentire network and positioning error generated in the process of positioning. Thesimulation results show that the new incremental localization algorithm based onmeshing has faster location speed and smaller cumulative error for large scale network,and the advantage is more obvious if monitoring area is bigger.