Research On Low-overhead And Scalable Indoor Localization Technology Based On The Fusion Of UWB/WiFi

Location-based services have gradually penetrated into people's daily life,and indoor localization is the core technology.Nowadays,due to wide distribution of infrastructure and the rapid development of mobile devices,WiFi fingerprint technology has gradually become one of the most mature technologies in the indoor localization field.However,the high offline overhead and the lack of scalability hinder its large-scale implementation and development.On the other hand,due to Ultra Wide Band(UWB)has the advantage of fast data transmission,it can achieve centimeter-level positioning accuracy through ranging method.However,additional hardware devices need to be carried during positioning,and the positioning accuracy is reduced under non-line-of-sight conditions,which prevent it from large-scale development.In view of this,this thesis focuses on the positioning technology based on the fusion of UWB and WiFi.It can achieve scalability and reduces the overhead of offline database construction.Firstly,a low-overhead localization system with hybrid UWB/WiFi technology is proposed,which aims at reducing the overhead of offline fingerprint collection.Secondly,an automatic offline database construction method based on UWB is proposed.With the assistance of UWB technology,this method can automatically construct the offline fingerprint database.Therefore,there is no need to set up reference points in the target area in advance,which greatly enhances the system scalability.Finally,this thesis designs and implements a real-time indoor trajectory tracking system,and the effectiveness of system is verified in a real-world environment.The contributions of this thesis are outlined as follows:Firstly,this thesis proposes a low-overhead localization system with hybrid UWB/WiFi technology.In the offline phase,a timestamp matching algorithm is proposed,which combines coordinate and Received Signal Strength Indicator(RSSI)in hybrid UWB/WiFi environment.This method accelerates the construction of WiFi fingerprint database.Besides,in order to improve the positioning accuracy,a wireless Access Point(AP)weight assignment method based on different distances is proposed.In the online phase,the weighted K nearest neighbor algorithm is used to obtain the final positioning result.At last,this thesis verifies the effectiveness of the system by conducting experiments.Secondly,this thesis proposes an automatic offline database construction method based on UWB technology,which can automatically complete the construction of offline fingerprints database.In this method,a heterogeneous data fusion algorithm is proposed to obtain the initial fingerprint data.In order to address the problem of redundant initial fingerprint data,a fingerprint data optimization algorithm is proposed.In the online positioning phase,for improving the localization accuracy,this thesis proposes an online dual-domain constraints positioning method.Finally,this thesis verifies the effectiveness of the system by conducting experiments.Thirdly,this thesis designs and implements a real-time indoor trajectory tracking system,and implements a system prototype in real scene.The function of the system is to track the indoor trajectory in real time,and it is mainly composed of a data collection terminal and an upper computer terminal.The data collection terminal is responsible for data collection,and the upper computer terminal displays the pedestrian trajectory in real time.