Leveraging Polarization of WiFi Signals to Simultaneously Track Multiple People
summary
WiPolar is a system that uses commodity Wi‑Fi with dual‑polarized antennas to separate multipaths from different people by their horizontal/vertical radar cross sections and polarization angle, then pairs angles of arrival across two receivers and tracks people with a GM‑PHD filter. Implemented with off‑the‑shelf Wi‑Fi and HTC Vive ground truth, it tracks up to five people (including stationary targets) with a 56 cm median error versus 97 cm without polarization, and 74 cm median for stationary or briefly paused subjects.
AI-written summary · about
related talks
42mYour Wi-Fi Is the Eavesdropper’s Radar: How to Counter Privacy Threats...
24mPinPoint: Localizing Interfering Radios
36mFinding Responsibility
12mBFMSense: WiFi Sensing Using Beamforming Feedback Matrix
09mWIP: Hijacking Attacks on UAV Follow-Me Systems in Realistic Scenarios
40mRogue One: A Wifi Story
06mRF-Identity: Non-Intrusive Person Identification Based on Commodity RFID Devices
23mWiFi Pineapple Active Detection via a Raspberry Pi Zero