Research On WSN Localization Algorithm Based On RSSI Weighted Centroid And GASA Optimization

Wireless sensor network as a new science and technology,now has been a lot of attention.It will be a lot of new science and technology fusion.In addition it has a strong wireless communication capabilities,making the application in various fields are very extensive.For wireless sensor network node location information is very important,so has led to a major node location is a technology node location for research also received more and more attention.Firstly,the wireless sensor network localization algorithm has been developed for a simple analysis of their respective characteristics and shortcomings are briefly analyzed.The current location algorithm can be divided into two types according to the localization process,the need for distance measurement,they are measured based positioning algorithm based on distance and location algorithm based on distance measurement.RSSI is a typical algorithm based on distance measurement,because it can use the model to signal loss in the communication process to calculate the distance between nodes so that it does not require additional hardware through to ranging,thus becoming a hot research.This paper proposed a wireless sensor networks localization algorithm which is based on the weighted centroid RSSI and GASA optimization to improve the situation of node localization in wireless sensor network where the positioning error problems frequently occur.In ord to locate the own position of unknown nodes,it is assumed that there is a certain proportion ofthe known position of the anchor nodes in wireless sensor networks. Since, the distance between unknown nodes and anchor nodes can be calculated out by RSSI weighted centroid algorithm, an unknown location parameter mathematical model can be established. Then, the most optimal solution to obtain a position of unknown nodes is achieved with the help of GASA optimization algorithm. The simulation results show that when the number of anchor nodes is 30, the average positioning error of this algorithm is fewer than 10%, which is lower than that of RSSI weighted centroid algorithm which is about 20%-25.5%. In addition, the more nodes are, the less positioning errors are.