Non-cooperative WiFi Target Location Technology Based On Mobile Single Station

Due to the development of information technology and the increasing demand for location services,indoor location systems have attracted more and more attention from researchers.In the existing Wireless Local Area Network,the indoor location system of WLAN is mostly an active location mode,so the active participation of the target is required.For important applications such as rescue and anti-terrorism,when the target's location information is to be collected,it becomes very difficult.In addition,most indoor location technologies based on WLAN often need to utilize multiple Access Points(APs)can achieve the target localization,but it is often difficult to satisfy using multiple APs in the actual environment,if only a single AP can not complete the localization.In order to solve the above problems,this paper studies a non-cooperative Wi Fi target location technology based on mobile single station.The direction finding of non-cooperative Wi Fi targets is performed on the platform of Software-Defined Radio(SDR)and the moving distance of the direction finding platform is estimated by Pedestrian Dead Reckoning(PDR)algorithm,and then constructs a geometric model to actively complete the target position estimation.The research contents of this paper are as follows:Firstly,the angle-time coupling signal receiving model is constructed by using Channel State Information(CSI)of multi-carrier.The energy in the direct path direction is studied by using the relationship between subcarrier CSI and multipath delay,and then the Angle of Arrival(AOA)of incoming wave is estimated,which guarantees the real-time performance of the location system.The IEEE 802.11 Saleh-Valenzuela(S-V)channel model is used to analyze the impact of multipath signals on the energy spectrum peaks of direct signals under indoor environment.The existing SDR platform is used to build a third-party direction finding system,and the direction finding of non-cooperative Wi Fi target devices is realized.Secondly,in order to improve the real-time of the direction finding system,the amplitude characteristics of the Channel Impulse Response(CIR)in the Line-of-Sight(LOS)and Non Line-of-Sight(NLOS)environments are studied,and a LOS/NLOS environment recognition algorithm based on CIR is proposed,the amplitude features of the CIR are trained and classified by using the Random Forest(RF)machine learning algorithm,and direction finding and location of the target are realized in the LOS environment.In order to estimate the moving distance of the direction finding platform during location,this paper studies the PDR algorithm based on Micro-Electro-Mechanical System(MEMS),in which the accelerometer information is utilized to detect and estimate pedestrians' gait and step length and the Extended Kalman Filter(EKF)is used to fuse nine-axis sensor data to estimate the heading angle.Finally,this paper designs the AOA/PDR mobile single-station location algorithm,and implements the location system in software.At the same time,a test platform is built to verify the performance of the proposed algorithm.The experimental results show that the recognition rate of the proposed LOS/NLOS environment recognition algorithm reaches 96% and 97% respectively.The angle error of the mobile single-station location system is 11.5 degrees with 67% confidence,the estimated delay is within 0.5 seconds,and the location error with 67% confidence is 2.1 meters.