High efficiency wifi fingerprint localization based on distance costraints

Ying Liu, Zhiyong Tao, Xiaobing Li

Abstract: In complex indoor environment, the measurement values of Wi-Fi fingerprint are severely limited to multipath and unpredictable environmental dynamics, thereby reducing indoor location accuracy. In this paper, we propose a high efficiency WiFi fingerprint localization based on distance constraint. First, we train the raw fingerprint using the clustering method and to mint out stable and effective fingerprint features. Then on the base of these, the sampling point strategy of the annular coverage is further constructed. Futhermore, completes the distance measurement and fingerprint match, and give the location of the target. Unlike previous works, the resulting system is established to combine range-based scheme and fingerprint-based, in order to make up for the disadvantage of the ranging scheme, and reduce the matching cost of the fingerprints in the online stage. Meanwhile, it can disambiguate of the fingerprint space by use this distance. The simulation results show that the algorithm has obvious advantages in a complex environment, and the positioning model has better performance.