Human activity monitoring system with commodity WiFi infrastructure using channel state information

Abstract

Human activity detection is a research field that has been growing rapidly for the last few decades. It opens a wide field of applications in the fields of healthcare, smart homes, robotics, human-media interaction, surveillance and security. WiFi-based solutions have received a lot of attention lately. These are based on the idea that nearby wireless signals are affected by human bodies. Reflections are produced by the presence of static objects like walls and furniture, while additional propagation paths are produced by the presence of dynamic objects such humans. This paper proposes using WiFi for a low-cost, device-free human activity monitoring system, as it is readily available in the office and home these days. It is also ubiquitous in nature as it collects information about the environment while providing Internet. The idea behind this approach is to acquire and model changes in multipath WiFi radio waves due to human movement. Experimental validation in practical situations with varying occupants, different environmental conditions, and interference from WiFi devices is used to demonstrate robustness and scalability.