Contactless Respiration Monitoring During Sleep with a pair of Wi-Fi devices

Abstract

It is essential to detect human breathing signals during sleep because it can help us discover some potential disease risks in time. In recent years, contactless respiration detection based on WiFi signals has aroused extensive research interest. The paper adopts a pair of off-the-shelf WiFi devices and exploits the fine-grained channel state information (CSI) to track the respiration rate during sleep. We present a novel method to combine the amplitude and phase of the CSI ratio to address the “blind spots” issue. Our system utilizes two complementary antenna pairs to perform respiration monitoring based on the Fresnel zone model. Extensive experiment results demonstrate that our system can achieve contactless and sustainable detection of a person's respiration in different sleeping positions.