The Research And Verification On Key Technologies Of Wi-Fi Indoor Positioning With Altered Fingerprint

Wi-Fi fingerprinting based localization technique utilizes the signal strength of the access points to estimate users' location,it has a high applicability in complex indoor environments.With the widespread deployment of wireless local area networks,it has drawn a great deal of attention in the industrial and academic fields.The matching degree between the fingerprint database and the current signal environments is the key factor to accurate positioning.However,the fingerprints may change due to reasons such as the movement or power adjustment of the wireless access points.How to ensure accurate localization with altered fingerprints and reduce the maintenance of off-line databases has been a difficult and important issue in this field.In order to solve the problems mentioned above,the contributions of this thesis are as follows:Firstly,concerning about the traditional k-nearest neighbor positioning,the factors which affect its positioning accuracy are analyzed.Three proposed algorithms are proposed to filter the access points in the matching fingerprints.Two different location sets were used to test the improved algorithms,and the cumulative distribution function and average positioning error were given.In the test of both location sets,the average positioning accuracy of all the improved positioning algorithms are improved by more than 40% compared with the nearest neighbor algorithm,which validates the effectiveness of the improved algorithm and the stability of the positioning result.Secondly,the fast diversity detection algorithm is used to detect whether there is a change in the location area for the Wi-Fi fingerprinting,and a fingerprint reconstruction algorithm that adapts to the altered fingerprints is proposed.This algorithm applies the global threshold function processing to the altered AP recognition,removes the changed AP which is identified,reconstructs the fingerprints,and then performs positioning.Compared with the above improved algorithms,the fingerprint reconstruction algorithm reduces the average location error about 20% when the changed access points exist.It effectively avoids positioning errors in the presence of altered APs and improves positioning accuracy.In addition,this paper carries out a Wi-Fi indoor positioning system which adapts to changed fingerprints,including a fingerprints acquisition client,a positioning client which developed by Android,a positioning server,and a MySql database server.The fingerprints acquisition was performed,and the above theoretical algorithms were measured,furthermore,the test results are analyzed.The indoor positioning scheme proposed in this thesis effectively adapts to the change of Wi-Fi fingerprints.It can provide reliable indoor position information for location-based services such as information recommendation or query.In addition,the maintenance of off-line fingerprint database is relieved,and the identified altered access points provide dataset for the research on fingerprinting database update.