Study On Localization Algorithm Based On RSSI Measurement In Wireless Sensor Network

In recent years,with the rapid development of researching on Wireless Sensor Networks,node positioning technology,which as one of the supporting technologies of Wireless Sensor Networks,has attracted more and more attention,and become a hot issue around the scholars.The nodes' position information is the prerequisite for many functions and applications can be realized in Wireless Sensor Networks.Therefore,it is important to research on node localization technology.This paper analyzed the current research status of wireless sensor network node positioning technologies and algorithms,introduced the existing localization algorithms in the two aspects of ranging-based and ranging-free,analyzed localization algorithms in algorithm characteristics,scope of application,the positioning precision.Based on research and analysis of RSSI ranging technology,two high precision positioning algorithms are proposed.The main contents of this paper are as follows:An improved node location algorithm based on simulated annealing is proposed.First,the algorithm calculates the area where the unknown nodes may exist according to beacon nodes' communication range,and set the centroid of the area as the initial coordinates of the unknown nodes.Then,according to the variance of RSSI ranging,sets the weight of beacon nodes to optimizing the cost function of simulated annealing,to reduce the effect of beacon node which has large ranging error on nodes positioning.By the experimental analysis,under the same conditions,compared with the Simulated Annealing Localization algorithm,the algorithm can reduce the positioning error effectively.This paper proposes a weighted probabilistic localization algorithm based on RSSI.Concerning the problem that path loss exponent for each link is different,the algorithm using Gaussian noise simulates path loss exponent error to construct the ranging probability model.According to the beacon nodes' communication range,divide the area where the unknown nodes exist into a number of grids.The beacon nodes' weights are determined according to their signal strengths.Then,grids'confidence degree are calculated according to beacon nodes' weights and corresponding probability models,the coordinates of the grid which has greatest confidence degree as unknown nodes' location.Finally,according to the network topology information eliminate flip ambiguity.Simulation experiments show that,under the same conditions,compared with the Weighted Least Squares localization algorithm,the proposed algorithm is closer to the real environment and has higher positioning accuracy.