Research On WIFI Fingerprint Location Optimization Algorithm

The rapid development of information technology and popularization of intelligent mobile terminal products improve the requirements for location-based service applications.The traditional positioning technology represented by GPS(Global Position System)has few advantages on positioning indoor or in the environments which lack of hardware support.And due to high physical shielding or high hardware cost,many other positioning methods,like infrared positioning system and ultrasonic positioning,are limited.With the widely deployment of Wifi Access point,the WiFi fingerprint location has a great potential application whether indoor or outdoor.However,with the rapid development of network,the increase of Internet users' datas,as well as the dynamic changes in the deployment location of WiFi access points,bring wifi fingerprint location the problem that how to deal with the big new data.In order to solve the problems caused by the new data,density-based clustering algorithm is studied in this paper.Integrated the incremental characteristics of DBSCAN(Density-Based Spatial Clustering of Applications with Noise)algorithm,an incremental HDBSCAN(Hierarchical Density-Based Spatial Clustering of Applications with Noise)clustering algorithm based on density adaptive property of the HDBSCAN is proposed.Given the appearance of new samples,the processing steps are refined based on the traditional incremental DBSCAN algorithm from two aspects,core point neighborhoods and update set.The algorithm devised in this paper effectively uses the historical information of clustering and reduces the dependency on parameters of the density based incremental clustering.The simulation results show that incremental clustering algorithm designed in this paper improves the clustering positioning performance in terms of the computational cost and the positioning time.Based on the study of incremental clustering algorithm,this paper further studies how to position accurately with less fingerprint information data to restore the user's historical trajectory.Drawing on the method of KF-CS(Kalman Filtered Compressed Sensing),the user mobile behavior is modeled and a dynamic compressed sensing based WiFi fingerprint positioning algorithm is devised.Dynamic CS localization algorithm go through multiple iterative computation of Kalman filtering and compressed sensing to reduce the mean square error below the set threshold.And the simulation results show that compared with the traditional CS based WiFi fingerprint positioning algorithm,the dynamic compressed sensing based WiFi fingerprint positioning algorithm can improve the positioning accuracy and reduce the time required for position recovery.