Research On Key Technologies Of WLAN Location Fingerprint Indoor Positioning Algorithm

With the rapid development of wireless communication technology and the wide application of various intelligent terminals,The demand for high-accuracy indoor location services is growing dramatically.In the field of medical treatment,earthquake relief,public social media,transportation,navigation and scene monitoring,it has shown a huge market prospect.WLAN-based fingerprinting positioning technology are relatively simple to construct positioning environment with low cost,with the application in the vast majority of indoor positioning needs,has become a hot technology in the current research on indoor positioning.In this thesis,WLAN-based fingerprinting positioning technology is studied.In the beginning,the common errors and their causes in WLAN-based fingerprinting positioning technology are discussed,and the corresponding solutions are summarized.This thesis analyze WLAN-based fingerprinting positioning technology from off-line and online,this thesis makes in-depth analysis and research from four aspects of AP positioning performance difference analysis,fingerprint database construction,database clustering,block matching and matching localization.In order to reduce the workload of establishing the position fingerprint database and to improve the positioning accuracy,an improved algorithm for location fingerprint localization based on AP signal distribution variance is proposed.Because of the different distribution position of AP and the difference of AP itself,different AP have different effects on the positioning effect,this method distinguishes the contribution sizes of different AP in the positioning process.From the simulation experiment,the improved localization algorithm based on AP signal variance not only shortens the establishment time of fingerprint database,but also improves the accuracy of positioning.Secondly,the application of K-means clustering method in WLAN-based fingerprinting positioning technology on WLAN is analyzed,this thesis summarizes the advantages and disadvantages of the clustering algorithm in the positioning process and put forward the corresponding improvement plan.Finally,because the K-means clustering method exists in the class boundary fuzzy problem and we can not accurately select the reference point located at the boundary of adjacent class,this directly results in lower positioning accuracy at the boundary of adjacent class.To solve this problem,a method of indoor location fingerprinting based on twice K-mean clustering is proposed.Through the re-processing of the first clustering results to reduce the error caused by the unreasonable selection of the reference point at the boundary of adjacent class,and improves the accuracy of the positioning.