Research On High Precision Indoor Positioning Algorithm Based On WLAN

With the popularization and development of wireless network technology,people’s demand for location information in daily life is increasing day by day.Indoor positioning technology has a broad application prospect in the fields of logistics transportation,earthquake rescue and shopping navigation.Based on WLAN network,this paper studies indoor fast and high precision positioning technology.In the offline stage,a fast fingerprint database construction algorithm is designed.The scheme is divided into three fingerprint database optimization algorithms: fingerprint database filling,AP selection and reference point clustering.The Fuzzy C-Means clustering algorithm is introduced to fill the fingerprint database to ensure its integrity and improve the information entropy of the fingerprint database.An AP selection strategy based on Inter Quartile Range is used to simplify the fingerprint database and reduce the scale of the fingerprint database.The improved Self-Organizing Map algorithm is proposed to realize the clustering of reference points and divide sub-regions.The experimental results show that the offline fingerprint database can lower the complexity of fingerprint database and reduce the locating time.In the online stage,a high-precision fusion positioning algorithm is designed,and the WKNN algorithm is improved from three aspects:reference points selection,characteristic distance compensation and weight function.A reference points selection strategy based on relative density is studied to flexibly select the number of reference points involved in positioning.Two kinds of reference points are proposed to determine different types of neighborhoods and the neighborhoods are trained and determined by polynomial fitting model.The optimal value of the distance compensation factor is obtained by the experiments and optimization algorithm.Different weight functions are evaluated experimentally,and Gaussian function is selected as the weight function.Experimental results show that the improved WKNN algorithm effectively improves the positioning accuracy,and the average positioning error of the fusion algorithm is(1.07±0.12)m in different positioning environments.The fusion algorithm is suitable for complex localization scene and has practical application prospect.28 Figures,17 Tables,66 References.