Crowdsourcing-Based Indoor Wi-Fi Fingerprint Map Generation Method

With the rapid deployment of WLAN and the widespread popularity of mobile intelligent terminals,people's demand for Location Based Services(LBS)is increasing.In the field of indoor positioning,Wi-Fi fingerprint positioning technology based on Received Signal Strength(RSS)has the advantages of low cost,low power consumption,and easy implementation,which has become the research hotspots of indoor positioning.However,building a Wi-Fi fingerprint map will consume a lot of time and labor costs,thus limiting the further development of Wi-Fi fingerprint positioning technology.The crowdsourcing model proposed in recent years has made the construction of Wi-Fi fingerprint maps simple and easy,saving considerable time and labor costs.This paper makes an in-depth study on the selection of Access Point(AP)in Wi-Fi fingerprint positioning technology based on crowdsourcing mode and how to construct a lightweight Wi-Fi fingerprint map.The main work as follows:(1)For a large number of APs in the positioning area,this paper proposes Stable AP Based On Regional(SAP-R)selection algorithm.This algorithm comprehensively considers the regionality,stability and frequency of the AP signal,so as to select a reliable AP subset in the positioning area.(2)In the construction of Wi Fi fingerprint map,aiming at the matching problem of Wi Fi fingerprint from Wi Fi RSS sequence,this paper proposes a Coefficient Weighting(CW)algorithm based on APs rank and improved Euclidean distance method,which can judge whether Wi-Fi RSS fingerprints in different sequences come from the same or similar positions more accurately.(3)In the construction of Wi-Fi fingerprint maps,in order to solve the problem of Wi-Fi RSS sequence merging,this paper first constructs a scoring matrix based on the Wi-Fi RSS sequence,and uses the Coefficient Weighting algorithm to judge whether the Wi-Fi fingerprints in different sequences match.The element values in the matrix are calculated according to the matching results;then the scoring matrix is backtracked,and the subsequences with the same averaging position are combined according to the backtracking results.The Wi-Fi fingerprint map constructed by this method can not only improve the positioning accuracy,but also realize the map update while reducing data redundancy.Finally,this paper deploys three different measured scenarios to complete the experimental testing of the entire indoor positioning process.The experimental results show that: on the basis of selecting AP subsets,the method proposed in this paper can reduce the redundant data caused by crowdsourcing,support map update?...