Research On Indoor Localization Algorithm Based On Channel State Information

With the rapid development of wireless technology and the continuous increase of indoor activity time,indoor location-based services have increasingly penetrated into all aspects of people's daily life and work.Due to the wide coverage of WiFi signals and low cost without additional hardware equipment,WiFi-based indoor localization technology has become a research hotspot in the field of indoor localization and has important research value.Traditional WiFi localization technology generally adopts the method based on received signal strength indicator(RSSI),but there are problems such as poor signal stability and low localization accuracy.Channel state information(CSI)contains rich subcarrier information and is more robust.Therefore,this paper conducts a systematic research on the indoor localization algorithm based on WiFi signal CSI,aiming to improve the localization accuracy.The main research work of this paper is as follows:(1)In order to solve the problems of low indoor localization accuracy and poor portability of equipment,the smartphone is used instead of the personal computer(PC)to obtain fine-grained CSI,and the CSI amplitude and phase difference data fusion method is adopted,and a smartphone-based active fingerprinting localization algorithm is proposed.The data is corrected by signal processing technology,and the optimal data selection method is designed to further improve the localization accuracy.In order to cope with the noisy wireless signal environment,the DBSCAN clustering algorithm is used to remove outlier sample points to reduce environmental interference.Then the principal component analysis(PCA)method is applied to extract the most effective features and improve the computational efficiency.Finally,the support vector machine multi-classification algorithm is used to estimate the target position.Experiments verify that this algorithm effectively improves the localization accuracy.(2)In order to solve the problems of inconvenience of carrying the device and heavy workload of the fingerprint algorithm,the impact of target movement on CSI data is analyzed,and a device-free localization algorithm based on Doppler frequency shift is proposed,which can achieve target localization without offline training.Firstly,the presence of persons is detected through changes in the CSI data of multiple wireless links.The pair of antennas information from the same wireless network card is used to eliminate the random phase shift,and the accurate Doppler frequency shift information caused by the walking of the person is extracted and correlated with the walking direction.Then,through time-frequency analysis of the CSI data of multiple links,the moving direction and walking duration of the target corresponding to each link are obtained to achieve target localization and target walking trajectory prediction.Finally,video surveillance is achieved through the linkage of the localization system and the camera.Experiments verify the effectiveness of this algorithm.This paper conducts extensive experiments in multiple real indoor environments to test the localization accuracy and effectiveness of the two proposed algorithms.By comparing the effects of different parameters and methods on the performance of the algorithms,the performance of the two localization algorithms proposed in this paper is evaluated in many aspects.Experimental results show that the localization accuracy of the two algorithms proposed in this paper are both greater than 90%,and the average localization error is less than 0.6 m,which can effectively improve the localization accuracy.There are 47 pictures,4 tables and 88 references in this thesis.