Passive Indoor Tracking Fusion Algorithm Using Commodity Wi-Fi

Abstract

Recent studies have found the mapping relationship between channel state information used in commercial Wi-Fi devices and environmental changes in the indoor environment, which can be used for sensing purposes. With the advantages of low cost and wide deployment of Wi-Fi facilities, passive indoor tracking systems based on Wi-Fi have huge potential. This article proposes and builds a passive indoor tracking system using commercial Wi-Fi devices, which realizes the function of tracking the human body’s trajectory in indoor environment. The system uses only commercial Wi-Fi devices. It processes the collected channel state information data by sending and receiving two pairs of Wi-Fi devices, and extract the movement information the messy data to obtain the trajectory of the human body. The system conducts a geometric feature analysis in the complex plane to obtain accurate displacement information, and utilize a fusion algorithm, combining the AoA (Angle of Arrival) information obtained by MUSIC algorithm, to obtain accurate human trajectory. In the experiment, the complex plane geometric feature analysis algorithm reaches centimeter-level accuracy in obtaining displacement information, while the system reaches decimeter-level accuracy on in obtaining indoor human trajectory on a simulation dataset.