Research And Application Of Multi-source Fusion Indoor Positioning Algorithm

With the development of mobile Internet,map surveying and indoor navigation technology,location-based services have received great attention.According to statistics,people spend an average of 80% of their time indoors,and on average 80% of data connection applications are also indoors.The GPS and Beidou positioning systems,which are widely used outdoors,only support outdoor positioning and cannot be accurate in indoor complex environments.Location service.With the advent of the digital age and the rapid development of wireless technology,mobile technology has been widely used,and people have begun to research and explore indoor positioning technology.Because the single positioning technology has problems such as limited accuracy,insufficient stability,and greater environmental impact.This paper proposes a fusion indoor positioning method based on Wi Fi/Bluetooth/PDR.The main content of the research is as follows: A Wi Fi positioning algorithm based on CMF-WKNN is proposed.In the offline phase,the fingerprint database is first established through weighted median filtering and clustering algorithms to reduce the time complexity of matching in the online phase and improve the efficiency of Wi Fi positioning.In the online matching stage,the selection of k value is optimized on the basis of the WKNN algorithm to improve the accuracy of the algorithm.Finally,the final location information is obtained through the weighted fusion method.A Bluetooth positioning algorithm based on JP-WKNN is proposed.Based on the WKNN positioning algorithm,this algorithm uses the weighted centroid method to fuse to obtain the final Bluetooth positioning coordinates.And designed a Wi Fi/Bluetooth fusion positioning algorithm based on self-adaptive weighting to perform multiple positioning of Bluetooth while performing one Wi Fi positioning.After preprocessing,the final positioning result is obtained by adaptive constraint fusion of weights.In this paper,the average error of the fusion positioning algorithm of Wi Fi and Bluetooth decreased from 2.04 m to 1.75 m,the probability of 3m accuracy rose from 82.3% to 86.5%,and the probability of 2m accuracy rose from 50% to67.8%.This algorithm can solve the problem of signal instability in a single positioning,and combine the advantages of the two to improve positioning accuracy and stability.Finally,a Wi Fi/Bluetooth/PDR fusion positioning algorithm based on UKF is designed.First,the initial fusion positioning results of Wi Fi and Bluetooth are obtained through the weight adaptive constraints in the previous step,and the preliminary fusion positioning results and PDR positioning are used to achieve three-source fusion positioning through UKF.This algorithm can solve the problem of accumulated error in PDR and the problem of signal instability in fingerprint positioning.In this paper,the optimal accuracy of Wi Fi/Bluetooth/PDR fusion positioning is improved from 90% to 95% compared with the optimal accuracy of Wi Fi/Bluetooth fusion positioning.