Why Commodity WiFi Sensors Fail at Multi-Person Gait Identification: A Systematic Analysis Using ESP32
A new study published on arXiv analyzes the limitations of using commodity WiFi sensors for multi-person gait identification. Researchers found that current algorithms and ESP32 hardware struggle to accurately distinguish between multiple individuals, with performance typically ranging from 39% to 56% accuracy. The findings suggest that the quality of sensing and spatial diversity, rather than algorithmic choices, are the primary constraints, questioning the practicality of this technology for robust multi-user biometric authentication. AI