Threshold-Based Fall Detection System Utilizing Passive Wi-Fi Radar Technology

• Published in: IEEE Symposium on Wireless Technology and Applications, ISWTA
• Main Author: Razali H.; Abd Rashid N.E.; Azman A.B.; Nasarudin M.N.F.; Khan Z.I.; Ab Rahim S.A.C.
• Format: Conference paper
• Language: English
• Published: IEEE Computer Society 2024
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85203804944&doi=10.1109%2fISWTA62130.2024.10651725&partnerID=40&md5=4f2f405f5b9b968862b5c8f6d337804b

This research introduces a Wi-Fi passive radar approach to address key challenges in current fall detection systems, including privacy concerns associated with camera-based systems, and the inconvenience of wearable sensors. By leveraging existing Wi-Fi signals, this research offers a non-intrusive and low-cost solution for continuous fall monitoring in home environments. The primary objective is to enhance safety for individuals at risk of falling by developing a sensor capable of accurately differentiating between fall and non-fall incidents. The methodology involves analyzing a dataset of 10 radar signals, equally distributed between falls and non-falls by using effective thresholding techniques. The system examines features such as voltage drop and signal envelope characteristics to identify fall incidents. The threshold-based classification method demonstrates remarkable accuracy, successfully distinguishing between falls and non-falls with 100% precision in our experimental dataset. This research contributes to the development of non-intrusive, reliable fall detection systems, potentially improving the safety and independence of elderly individuals living alone. © 2024 IEEE.