Research On Indoor Wi-Fi Positioning Based On Beamforming

Along with the development of wireless communication techniques,the popularization of the intelligent terminals and the increasing demand for the indoor location based service(LBS),indoor wireless localization technology has received more extensive attention in recent years.Wireless local network(WLAN)based indoor localization technology has advantages of low cost,portability,wide range of use and becomes a research hotspot in the field of indoor wireless localization sensing.This thesis studies different indoor localization technology and indoor localization system,analyzes the research status at home and abroad,compares the advantages and disadvantages of different indoor localization methods,and finally carries out research work based on beamforming-based indoor Wi-Fi localization.This article studies indoor channel modelling,indoor localization algorithm and introduces the Angle of Arrival(AOA)estimation method based on beamforming to optimize indoor localization.The specific contents are as follows.First,this thesis studies indoor channel modeling based on the deterministic model and the empirical model.For indoor channel modeling based on the deterministic model,this thesis studies the basic principle of indoor channel modeling based on ray tracing,and the Received Signal Strength(RSS)data in the indoor environment is acquired through the input modeling of Wireless Insite simulation software to build the offline fingerprint database.For indoor channel modeling based on empirical model,this thesis studies the indoor channel modeling based on log-path loss model and distance separation model based on IEEE 802.11 n,and obtaining indoor offline RSS data by MATLAB for simulation.Compared the deterministic model with the empirical model,it is found that the method of indoor channel modeling based on ray tracing is better,and it is more in line with the characteristics of multipath fading in indoor scenes.Then,this thesis studies indoor localization algorithm based on propagation model and fingerprint database model.For the indoor positioning algorithm based on propagation model,the log-path loss model is estimated by least squares fitting,and the distance measured from the target point to the access point(AP)is estimated by the RSS,which is measured by the target point.In order to solve the problem that the log-path loss model estimates the distance of "singular point" and the distance difference between adjacent target points and the same AP is within a certain range,this thesis proposes a method based on singular point correction and improvement of the adjacent point threshold correction localization algorithm.The performance of the algorithm has been greatly improve,which corrects the average positioning error from 33.16 m to 2.11 m.Then,for the indoor location algorithm based on fingerprint database,compared with the traditional neighbor algorithm,the feature information of the data in the fingerprint database is not ful y utilized.This thesis proposes an improved fingerprint localization algorithm based on random forest regression model.The algorithm trains offline fingerprint database by random forest regression model,and reduces the average positioning error form 1.84 m to 1.30 m,and analyzes the influence of the positioning results by simulation.Finally,this thesis analyzes the large positioning error in the localization algorithm based on the fingerprint database,due to the smaller RSS which leads to the inconspicuous fingerprint characteristics,and then affecting the positioning performance.To solve this problem,an improved fingerprint localization algorithm based on beamforming AOA estimation is proposed by introducing angle estimator,which reduces the average positioning error to 0.80 m.on the basis,based on the continuous motion characteristics of the target,a quadratic localization algorithm based on neighborhood threshold optimization is proposed,which reduces the average positioning error to 0.62 m.