WiPasLoc:A Novel Passive Indoor Human Localization Method Based on WiFi

Abstract

Passive indoor human localization is the basis for implementing ubiquitous wireless sensing systems.However,commercial WiFi signals are easily affected by the surrounding environment in our life,which makes it difficult for existing WiFi-based indoor localization works to accurately separate the dynamic human components from the complex received signals.To address this problem,this paper proposes the WiPasLoc,a passive indoor human localization system,which achieves high accuracy indoor localization by using the channel state information(CSI) extracted from commercial WiFi devices.Firstly,the Doppler frequency shift(DFS) estimation is carried out in combination with the signal quality of the CSI subcarriers.Then,the target person signal component is precisely separated from the channel state information by using a double-window-based angle of arrival(AoA) estimation method.Finally,combined with the initial position information of the personnel,accurate passive indoor human localization is achieved by the proposed trajectory fitting algorithm.Experimental results show that the median error of WiPasLoc for indoor personnel trajectory positioning is 80cm,which is 25.9% higher than the existing typical Widar2.0 positioning accuracy.

Keywords

• wifi|indoor localization|channel state information(csi)|doppler frequency shift(dfs)