Research On Indoor Positioning Algorithm Based On RSSI

With the development of modern urban construction,large-scale indoor space continues to increase,and people spend more and more time indoors.How to accurately obtain the position of indoor personnel,businesses,and goods has become an urgent need for people.At present,indoor positioning mainly uses wireless sensor networks as the research direction.The existing positioning algorithms have large error and cannot accurately obtain positioning information while WKNN has relatively small positioning error,but it is still limited.In order to reduce the indoor positioning algorithm error,a positioning error optimization algorithm based on WKNN is proposed.And in order to reduce the positioning cost,a positioning cost optimization algorithm based on WKNN is proposed.Based on WKNN positioning error optimization algorithm is to select and group the K fingerprints according to the beacon node after WKNN,and calculate the Euclidean distance between the K RSSIs in each group and the RSSI of unknown nodes respectively.Then,according to the euclitic distance to calculate K fingerprints weight and use it to correct unknown node RSSI.After,according to the logarithmic normal distribution model to calculate the distance between unknown node and beacon node.Finally,using the maximum likelihood estimation method to estimate the unknown node coordinates.This paper selects 200*200m environment suitable for large shopping malls and large underground parking lots to compare the positioning error of WKNN and the positioning error optimization algorithm.The simulation results show that the positioning error of the optimization algorithm is lower than that of WKNN,which improves the positioning effect and is suitable for the pursuit of high positioning accuracy.Based on WKNN localization cost optimization algorithm is to reduce the positioning cost in the same environment which uses 13 beacon nodes to rearrange 200*200m.According to the 13 RSSIs values of unknown nodes,three reliable beacon nodes are selected and the fingerprint database is updated according to the reliable beacon nodes.Then,the RSSI ratio of unknown node is obtained as the reliability of beacon nodes to modify the fingerprint database RSSI.Finally,WKNN is adopted to estimate the unknown node.The simulation results show that the positioning error of the cost optimization algorithm is slightly higher than that of WKNN in the 200*200m environment,but reduce the positioning cost of at least 55% by reducing 23 beacon nodes.The cost optimization algorithm is suitable for situations where the positioning accuracy is moderate and the cost is strictly controlled.