A Multiorientation Real-Time Respiration Monitoring System Using Wi-Fi Sensing Based on the Homofocal Hyperbola Model

The respiratory rate monitoring plays an important role in different fields of use, ranging from healthcare to daily work and life. Recently, Wi-Fi-based technology using commercial devices has attracted much attention due to its noninvasive and low cost. Most of the existing Wi-Fi-based respiratory monitoring systems are based on the Fresnel zone (FZ) model. But when the subject breathes with the body oriented parallel to the tangential direction of the FZ, the signal detection becomes difficult, and this orientation dependency hinders the widespread application of Wi-Fi-based respiratory monitoring. To solve this problem, this article proposes and experimentally validates a novel homofocal hyperbola (HH) model for indoor Wi-Fi sensing. By connecting two transmitting antennas to the same signal source and deploying the antenna position appropriately, the sensing effect makes up for the disadvantage of low perception sensitivity when moving along the tangent direction of the FZ. Then, a human respiratory monitoring system is developed based on the HH model. Specifically, the proposed dominant frequency-to-amplitude ratio (DFAR) method is used for subcarrier selection and respiratory signal detection, and the proposed zero-peak (ZP) detection method is used to calculate the respiratory frequency. Compared with the FZ model, the azimuth range of the HH model that can monitor human respiration is nearly doubled with an error of less than 0.7 bpm.