| Global Navigation Satellite System(GNSS),in the outdoor open area,using other positioning technologies to assist positionging,can achieve sub-meter-level accuracy.However,in GNSS-free environments,especially inside of buildings in which people generally live and work,satellite signals are attenuated,and even can not receive satellite signals.With the increasing of people's demand of location-based services,meanwhile,smart phones and Wi-Fi routers have become a promising platform for mass-market.Smart phones not only have the built-in Micro-electro Mechanical System(MEMS)sensors but also support Bluetooth and Wi-Fi signal transmission.Currently,indoor and outdoor positioning is one of hot research topics worldwide.Taking smart phones as the research carrier,this paper discusses the background and significance,the current research in the field of indoor positioning,as well as the principles and methods.At present,most of researchs focus on the long and narrow corridor environment,but there is less research on office scenes.This paper takes the office scene as the indoor positioning environment,and researchs on the indoor positioning system for map-matching aided Wi-Fi,iBeacon and PDR Algorithm.The main research results are as follows:In view of the large amount of work to construct fingerprint database and the jump of received signal strength index(RSSI)of Wi-Fi positioning,an interpolation method is proposed to expand the fingerprint database,which reduces the time of establishing the fingerprint database.This paper is divided the indoor location area into 6 regions according to the vector structure of indoor map,used Support Vector Machines(SVM)location fingerprint matching method to identify the region of the pedestrian and the time of walking direction change;the RSSI value is used to construct the ranging model(Shadowing model)in Bluetooth positioning,and a location method is proposed using maximum likelihood estimation(MLE)and performance index.Compared with the traditional MLE,the positioning accuracy is effectively improved.Due to the problem of cumulative error in PDR positioning technology,a correction method using Wi-Fi location region constraint and indoor map matching is proposed to effectively constrain the pedestrian trajectory,which improves the positioning accuracy compared with the traditional PDR algorithm.In order to solve the problem that Wi-Fi,Bluetooth,and PDR positiong used different sensors and time is not synchronized.Under the same experimental environment of the trajectory,this paper proposed two algorithms for fusion positioning of Wi-Fi,PDR and iBeacon respectively.One is the map matching aided moving average filtering algorithm.The location result depends on the size of the sliding window.When the target deviation of the location point is large,the location result depends on the size of the sliding window.The moving average filter can effectively constrain the positioning point to the indoor positioning map area,which can well describe the walking track of pedestrians.The second is the map matching aided Extended Kalman Filter(EKF)algorithm,which fuses the positioning results of Wi-Fi,PDR and Wi-Fi,PDR and iBeacon by moving average filtering.The experimental results show that the map matching aided EKF algorithm effectively improves the trajectory drift and positioning bounce,and the positioning accuracy of about 90%of the positioning points is better than 2.5m.Figure[60]table[13]reference[81]... |
PDF Full Text Request