WiTraj: Robust Indoor Motion Tracking With WiFi Signals

WiFi-based device-free motion tracking systems track persons without requiring them to carry any device. Existing work has explored signal parameters such as time-of-flight (ToF), angle-of-arrival (AoA), and Doppler-frequency-shift (DFS) extracted from WiFi channel state information (CSI) locate people in a room. However, they are not robust due unreliable estimation of parameters. ToF AoA estimations accurate for current standards-compliant devices that typically have only two antennas limited bandwidth. On the other hand, DFS can be relatively easily on but is susceptible high noise level random phase offset CSI measurement, which results speed-sign-ambiguity problem renders ambiguous walking speeds. This paper proposes WiTraj, indoor system using commodity devices. WiTraj improves robustness three aspects: 1) It significantly quality by ratio each receiver, 2) To better human walking, it leverages multiple receivers placed at different viewing angles capture then intelligently combines best views achieve trajectory reconstruction, and, 3) differentiates in-place activities, interleaved daily life, so non-walking activities do cause errors. Experiments show improve accuracy typical environments compared existing DFS-based systems. Evaluations across 9 participants 3 median error $<2.5\%$ room-sized trajectories.