M2-Fi: Multi-person Respiration Monitoring via Handheld WiFi Devices

Wi-Fi signals are commonly used for conventional communication, yet they can also realize low-cost and non-invasive human sensing. However, Wi-Fi sensing in Multi-person scenarios is still a challenging problem. In this paper, we propose M²-Fi to achieve multi-person respiration monitoring using a handheld device. M²-Fi leverages Wi-Fi BFI (beamforming feedback information) performs respiration monitoring. As a compressed version of the uplink CSI (channel state information), BFI transmission is unencrypted, easily obtained using frame capture, and does not require specific firmware to obtain. M²-Fi is based on an interesting experiment phenomenon that when a Wi-Fi device is very close to a subject, near-field channel changes caused by the subject significantly cancel out changes from other subjects. We employed VMD (Variational Mode Decomposition) to eliminate the interference caused by hand movement in the BFI time series. Subsequently, we devised a deep learning architecture based on GAN (Generative Adversarial Networks) to recover fine-grained respiration waveforms from the respiration patterns extracted from the BFI time series. Our experiments on collected 50-hour data from 8 subjects show that M²-Fi can accurately recover the respiration waveforms of multiple persons with handheld devices.

I. Introduction

In recent years, contact-free sensing technology using wireless signals (e.g., Wi-Fi [7], [12], [37], RFID [6], [11], [17], [18], and acoustic [4], [5], [13], [40]) has attracted extensive attention from both the academic and industrial sectors. Wireless sensing technology enables low-cost, sensor-free human sensing applications such as activity tracking, gesture recognition, respiration monitoring, and motion tracking. Wi-Fi-based wireless sensing has a considerable advantage in deployment cost due to the widely deployed Wi-Fi communication infrastructure. Although significant progress has been made in Wi-Fi sensing systems, it is worth noting that existing Wi-Fi sensing typically uses static transmitters and receivers. However, smartphones, smartwatches, and other sensing devices close to the human body will move with the hand. Besides, limited Wi-Fi bandwidth hinders obtaining sufficient range resolution to distinguish multiple sensing objects during realistic multi-person sensing. Therefore, a critical aspect missing in Wi-Fi sensing is the ability to perform sensing in multi-person scenarios using a handheld device in a moving state.