Research On Indoor Location Technology Based On Bluetooth Assisted PDR Algorithm

With the rapid development of new media technology represented by the internet,the Location Based Service(LBS)has become more and more popular in the daily life.In the past decade,LBS has been widely used in various applications such as map navigation,personnel location and logistics tracking.However,on the one hand,in the indoor environment,because of the complexity of the internal structure of the building,Global Positioning System(GPS)cannot provide stable and reliable positioning service for users.On the other hand,according to incomplete statistics,people spend most of their day indoors.Therefore,a high-precision,low-cost and easy-to-promotion indoor positioning technology has become an urgent need.Based on the above analysis,this paper focuses on the positioning technology based on the received signal strength(RSS)of Bluetooth and pedestrian dead reckoning(PDR).Then,a Bluetooth-assisted PDR algorithm for indoor positioning is proposed.The main research contents are as follows:(1)This paper presents a parameter adaptive PDR algorithm(APDR).Compared with the traditional PDR algorithm,APDR improves the gait detection and heading estimation.In the aspect of gait detection,a threshold adaptive gait detection method is proposed.Firstly,the influence of the glitch noise on the accelerometer is eliminated by moving average filter.Then the coherence of the front and back steps is used to dynamically change the threshold.Finally,gait detection is carried out based on the dynamic threshold.In the aspect of heading estimation,a heading fusion method with adaptive parameters is proposed,which not only integrates the gyroscope and the electronic compass,but also references the heading information of the previous moment.Specifically,the moving average filtering is used to reduce the random error and noise of the electronic compass.Then,the weight of the above three parameters is adjusted in real time by using the adaptive method.Finally,the final heading is obtained according to the weight.The experimental results show that the proposed APDR algorithm has average positioning errors of 1.38 m and 1.18 m respectively in two different experimental environments.Compared with MPDR(PDR using magnetometer-based heading),GPDR(PDR using gyroscope-based heading)and SmartPDR,the positioning accuracy is improved by 44.49%,39.99% and 25.76%,respectively.(2)In order to solve the problem of accumulated error of APDR,this paper proposes a fusion indoor location algorithm based on Bluetooth assistant APDR(B-APDR algorithm).The key idea is to use Bluetooth RSS strongest beacon method to get locations whose positioning error is in a certain range,and these locations are called as key points.Then,B-APDR uses the key points before and after to eliminate APDR's cumulative error and adjust APDR's parameters.Specifically,firstly,the initial location of the smartphone is given;secondly,the smartphone collects Bluetooth RSS while runs APDR positioning;when the RSS is greater than the preset threshold,B-APDR employs the proposed key point identification method to identify the indoor key points;finally,B-APDR corrects the positioning result and adjusts the step length parameters of APDR.The experimental results show that the proposed B-APDR algorithm has average positioning errors of 0.92 m and 1.00 m respectively in two different experimental environments.Compared with the APDR algorithm,the positioning accuracy of the B-APDR algorithm is improved by 33.33% and 15.25%,respectively,and it can eliminate the cumulative error of the APDR effectively.