Research On Csi And Pdr Fusion Indoor Positioning Algorithm Based On Ekf

In recent years,with the popularization of mobile smart devices and the development of wireless communication technology,there is great demand for indoors and outdoors positioning services.In outdoor environment,global navigation satellite system(GNSS)can provide users with high-precision positioning through signal enhancement technology,but in complex indoor environment,because the GNSS signal is blocked,it can not meet the needs of indoor high-precision positioning,so the development of indoor positioning technology has become a research hotspot.In view of the limitations of two commonly used indoor positioning methods,which are fingerprint positioning based on channel state information(CSI)and PDR positioning based on pedestrian dead reckoning(PDR),this paper proposes a CSI and PDR fusion indoor positioning algorithm based on extended Kalman filter(EKF).The main contents and conclusions of this article are as follows:(1)The traditional fingerprint location method based on CSI only uses CSI amplitude information as fingerprint information,which leads to the low accuracy of fingerprint location.In this paper,the CSI phase information and amplitude information are combined as the location fingerprint information of CSI fingerprint location to improve the indoor location accuracy.In this paper,linear transformation is used to reduce the CSI random phase shift caused by hardware acquisition defects,and the feasibility and effectiveness of using CSI phase information as fingerprint information is verified by phase correction algorithm(2)In order to solve the problem that the traditional fingerprint location method based on Received Signal Strength(RSS)has a large positioning error,this paper proposes a CSI passive fingerprint location algorithm based on K K-Nearest Neighbor(KNN)classifier.Compared with RSS,CSI has a higher degree of channel information and can accurately reflect the environmental information.Finally,the algorithm is compared with the traditional fingerprint location method based on RSS and CSI amplitude information.The results show that in the same experimental environment,the CSI passive fingerprint location method based on KNN improves the indoor location accuracy by 55% and 15%respectively compared with the former two methods.(3)Aiming at the problem that the positioning system using single positioning technology has poor applicability and can not meet the needs of indoor high-precision positioning.In this paper,an indoor localization algorithm based on the fusion of CSI and PDR is proposed.The CSI localization module uses the CSI passive fingerprint localization algorithm based on KNN,and the PDR localization module uses the multiple gait on-line estimation Monitoring PDR positioning algorithm can reduce CSI positioning fluctuation and PDR accumulated error and improve system positioning accuracy by using EKF's unique position trajectory constraint.Finally,the algorithm is compared with the passive fingerprint location methods based on multiple gait on-line estimation Monitoring PDR and CSI.The results show that the average error of CSI and PDR based on EKF is the smallest,and the positioning accuracy is improved by 54% and 21% respectively.