Respiratory rate is considered one of the important indicators to reveal human health and sleep quality. Conventional respiratory rate detection requires the user to equipped with wearable or handheld devices which are cumbersome and infeasible for long-term usages. Radar has been developed as a device-free solution for respiratory detection in recent years; however, high price issue prevents it from being generally adopted. With abundant channel state information (CSI) from Wi-Fi signals, we can detect the chest fluctuations generated by a human body by proposing the self-adaptive respiratory rate estimation (SARE) for device-free CSI-based Wi-Fi systems. The main features of proposed SARE system include self-adaptive anti-interference technique, slight turbulence and frequency detection, and multi-antenna spatial detection. Without the pre-trained data and models, our proposed SARE system can provide precise respiratory rate detection such as to predict the occurrence of sleep apnea.

Without requiring to equip with wearable devices, we proposed a self-adaptive respiratory rate estimation (SARE) for Wi-Fi systems. The key features of SARE system include self-adaptive anti-interference technique, slight turbulence and frequency detection, and multi-antenna spatial detection. By adopting signal processing techniques, it is not required for the SARE system to collect pre-trained data for the construction of feature models. The self-adoption feature can rapidly adjust the signal pattern in a short time period to calculate the human＇s respiratory rate. We process the channel state information (CSI) of Wi-Fi signals to offer centimeter-level estimation accuracy; whereas multiple antennas of Wi-Fi devices provide the spatial diversity for turn over detection of human body.

We proposed a self-adaptive respiratory rate estimation (SARE) for device-free CSI-based Wi-Fi systems. Through the device-free and adaptive features, the SARE system can be easily and painlessly implemented in a household, helping individuals to monitor their own sleep breathing quality via the commercially available Wi-Fi routers and devices.  

It has been reported that 50% of people aged 20-70 have sleep apnea problems. The proposed SARE system can help users detect the respiratory rate, the number of human body＇s turn overs during sleep, and eventually detect whether there are symptoms related to sleep apnea. Furthermore, the SARE system can precisely detect the “Shortness of Breath”, which has been claimed by WHO that 18.6% of the COVID-19 infected people will possess this symptom.