Extracting Gait Velocity and Stride Length from Surrounding Radio Signals
summary
A wall-mounted radio sensor uses reflections to track location and isolate walking periods, then extracts gait velocity and stride length by finding stable periodic phases; lab validation with VICON on 18 subjects shows 98% gait-velocity accuracy and 96% stride-length accuracy, walking detection reports 100% precision and 96% recall, and a multi-month deployment in 14 homes demonstrates continuous monitoring without wearables or cameras and high user acceptance.
AI-written summary · about
chapters (11)
- 0:00Introduction to gait monitoring
- 4:00WiGait system overview
- 7:35Challenges in motion detection
- 10:46Walking period identification
- 12:13Extracting gait velocity
- 13:36Calculating stride length
- 13:57Lab experiment results
- 15:41User acceptance study
- 16:48Real-world deployment insights
- 18:33Conclusion and future work
- 18:53Q&A session
related talks
01h 13mHealth Monitoring with Wireless Sensors and Machine Learning
06mV2iFi: in-Vehicle Vital Sign Monitoring via Compact RF Sensing
18mVibrosight: Long-Range Vibrometry for Smart Environment Sensing
06mBodyCompass: Monitoring Sleep Posture with Wireless Signals
06mExploring LoRa for Long-range Through-wall Sensing
06mER-Rhythm: Coupling Exercise and Respiration Rhythm Using Lightweight COTS RFID
06mReal-time Arm Gesture Recognition in Smart Home Scenarios via Millimeter Wave Sensing
21mViBand: High-Fidelity Bio-Acoustic Sensing Using Commodity Smartwatch Accelerometers