We Hear Your PACE: Passive Acoustic Localization of Multiple Walking Persons
summary
PACE passively localizes and identifies multiple walking people by pinpointing footsteps, using structure-borne dispersion ranging (high-frequency components travel faster) and airborne AoA + MFCC identification. A prototype circular six-microphone beacon sits on the floor. It uses domain-adversarial adaptation to generalize across floors and spectral weighting for AoA to sharpen low-SNR peaks. Results show a median localization error of about 30 cm, AoA within ~10° with five concurrent walkers, and up to 96% ID accuracy (center loss adds ~12%).
AI-written summary · about
related talks
22mUbicoustics: Plug-and-Play Acoustic Activity Recognition
14mVeCare: Statistical Acoustic Sensing for Automotive In-Cabin Monitoring
07mDecentralized Attention-based Personalized Human Mobility Prediction
13mUltraPoser: Pushing the Limits of IMU-based Full-Body Pose Estimation with Ultrasound Sensing on ...
06mPMF: A Privacy-preserving Human Mobility Prediction Framework via Federated Learning
06mMAIL: Multi-Scale Attention-Guided Indoor Localization Using Geomagnetic Sequences
04mDBTB INT John Whaley r
06mBodyCompass: Monitoring Sleep Posture with Wireless Signals