Research And Implementation Of Indoor Positioning System Based On Multivariate Fingerprints Fusion And Pedestrian Dead Reckoning

With the popularity of wireless positioning technology,location based service(LBS)shows a broad market prospects.At present,WLAN fingerprint-based indoor fingerprint positioning is the mainstream of indoor positioning technology.Due to the time-varying nature of WLAN RSS in the indoor environment,the fingerprint positioning accuracy based on WLAN RSS is limited.With the popularity of a variety of sensors in smart phones,as well as the emergence of a variety of positioning methods based on smart phones,this article for the indoor environment,how to improve the positioning accuracy of the problem,from improving the fingerprint positioning accuracy and integration of positioning accuracy the following studies:Research on Fingerprint Location Mechanism of Fingerprint and Magnetic Field Intensity Fusion.In this paper,an algorithm based on AdaBoost is proposed for WLAN Fingerprint and magnetic field Fingerprint Fusion.Based on the analysis of WLAN RSS and magnetic field strength,WLAN RSS and magnetic field intensity are used as binary fingerprints in fingerprint library to improve the spatial difference and temporal stability of fingerprints.AdaBoost machine learning algorithm to match the fingerprint,calculate the final positioning results.Experiments show that the fusion fingerprint positioning mechanism compared to a single WLAN RSS fingerprint positioning mechanism can improve the positioning accuracy.Research on Fusion Location Mechanism of Fingerprint Location and Pedestrian Dead Reckoning(PDR)Location.In this paper,a fusion mechanism of fingerprint positioning and PDR localization is proposed.Based on the proposed method,the fingerprint localization method is used to fuse the fingerprint localization and PDR positioning,and the fusion method is obtained in the process of fusion.Statistical gain for various location information.Experiments show that the fusion mechanism after the integration of a single fingerprint positioning mechanism can improve the positioning accuracy.Design and Realization of Fingerprint Location and Fitting Location and Fitting Location.The positioning system uses C/S architecture.Client running on the Android platform,to achieve a fingerprint collection,PDR positioning and other functions.Positioning server-side running on the Windows platform,to achieve a fingerprint positioning,particle filtering and other functions.And finally through the software college 8-layer real environment verification,the system and fingerprint positioning system can effectively improve the positioning accuracy.