Tracking Human Motion Direction With Commodity Wireless Networks

Detecting when a person leaves room, or house is essential to create safe living environment for people suffering from dementia other mental disorders. The approaches based on wearable devices, e.g. GPS bracelets may detect such events require periodic maintenance recharge replace batteries, and therefore not be suitable certain types of users. On the hand, camera-based systems illumination raise potential privacy concerns. In this paper, we propose device-free walking direction detection approach RF-sensing, which does wear any equipment. proposed monitors signal strength fluctuations caused by human body ambient wireless links analyses its spatial patterns using convolutional neural network identify direction. has been evaluated experimentally achieve up 98% classification accuracy depending environment.