Research On The Range-free Localization Algorithm In Wireless Sensor Network

Localization technology is a key technology in Wireless Sensor Network(WSN), which provides position information of node for the WSN, and it has great significance for target tracking, data fusion, etc. Therefore, localization technology has became one of hot topics to study.In the existing localization algorithm, because the range-free localization algorithm have advantages which are low hardware requirements and low energy consumption, it is more suitable for resource-constrained Wireless Sensor Network. But the localization error of range-free localization algorithm is large, and most of them are used under two-dimensional environment. This article aim to study the range-free localization algorithm. The main work as below:Analyzing the basic theory and several different classification methods localization algorithm for Wireless Sensor Network. And introducing three calculation method of node localization, such as trilateration, triangulation, maximum likelihood estimation. At last several range-free localization algorithm is analyzed in detail, it provides a starting point for following research.As the deficiency of centroid location algorithm and DV-hop algorithm, a rang-free and hybrid location algorithm has been proposed in the paper. This algorithm use the DV-hop algorithm to get the rough distance between different nodes and rough estimate coordinate for weighed centroid location algorithm. By twice-weighed centroid computation, it can get more precise coordinate of unknown node. located.Simulation results show that, this algorithm improve the precision of node localization compared with two original location algorithm.Object of study is DV-Hop algorithm which using in three-dimensional environment, and analyze the reasons why it has errors. Correcting the average hops distances of anchor. Besides, we use a simplified particle swarm algorithm replace the maximum likelihood method. The simulation results show that the positioning accuracy of the new algorithm has great improvement.As to this article, it is mainly for forest fire prevention and environmental monitoring applications. The methods and used in this paper also has certain significance for other applications.