Research On Key Technologies Of Wireless Positioning And Navigation

With the development of society,location information has become important in life.For example,in wireless sensor networks(WSNs)or mobile Internet applications,the accuracy of location information and location-based services can be effectively improved through wireless positioning and navigation technology.Therefore,it is important to study the key technologies of wireless positioning and navigation.This paper mainly studies DV-Hop(distance vector-based)in WSNs applications and smart phone-based E2E(end to end)indoor navigation in mobile Internet applications.The main works of this paper are as follows:1)Focusing on the hop-distance ambiguity problem of DV-Hop,this paper proposes a distributed gradient descent based localization refinement algorithm(GDLR).Firstly,GDLR proposes the hop size based pseudo ranging method to estimate the distance between unknown nodes and their neighbors.Then,the refinement process is modeled to minimize the square sum of the difference between the pseudo ranging distance and computed distance between unknown nodes and their neighbors.Finally,a gradient descent based search method is proposed to solve the above minimization problem.In practice,GDLR can be divided into GDLR1(use the information of 1-hop neighbors) and GDLR2(use the information of 1-hop and 2-hop neighbors).The simulation results show that,compared with DV-Hop and the two improved DV-Hop algorithms,GDLR1 improves the positioning accuracy by about 40%,33% and 29%,and GDLR2 improves the positioning accuracy by about 53%,49%and 47%,on average.Moreover,from the perspective of positioning accuracy and energy consumption,an energy efficient GDLR(EGDLR)is proposed.Simulation results show that the positioning accuracy of EGDLR increases by about 22% than that of GDLR1,and the energy consumption decreases by about 65% than that of GDLR2.2)Aiming at the smart phone-based E2E indoor navigation in mobile Internet applications,an E2E indoor navigation algorithm(HMagPDR) based on pedestrian dead reckoning(PDR) and geomagnetic information is proposed for the first time.HMagPDR is composed of a path matching algorithm based on PDR and a geomagnetic matching algorithm based on DTW(dynamic time wraping).The former is used for navigation between adjacent key points(such as turning or going up and down stairs),and the latter is used to detect path deviation.In addition,in order to improve the accuracy of step detection,HMagPDR algorithm proposes a step detection algorithm(ASD) based on a combination filter(sliding mean filter and kalman filter) and adaptive threshold.The results show that compared with the two improved PDR algorithms,the accuracy of ASD steps detection is improved by 18.9% and 10.3%,respectively.Furthermore,HMagPDR can detect adjacent paths with a minimum distance of 10 m through the geomagnetic matching algorithm based on DTW,which meets the design requirements of indoor path deviation detection.3)The E2E indoor navigation system(E2Enav) based on HMagPDR is realized.E2Enav consists of APP and servers.The APP adopts cordova frameworks based on Javascript,including the reference path of leaders and the navigation path of followers.The server uses the MySQL5.6 database.The APP and server are connected via mobile Internet.E2Enav implements the leader-follower navigation pattern by allowing users to self-motivate themselves by recording and sharing path tracking information to navigate followers.E2Enav can providing navigation services in indoor scenarios.The test results show that E2Enav has a matching rate of 100% for key points and can detect 80% of path deviation events within 4 s,which can be used for indoor navigation.