Multi-Target Respiration Sensing Based on Cell-Free Millimeter-Wave System

In recent years, respiration sensing has flourished. However, most existing solutions are based on Wi-Fi signals, suitable for single-target scenarios, and are limited in sensing distance due to noise interference. Compared to Wi-Fi systems, cell-free millimeter-wave systems have advantages such as large bandwidth, better security, and stronger stability, enabling finer-grained respiration sensing. Therefore, we proposed a multi-target respiration sensing method based on cell-free millimeter-wave system. First, we use the Extensive Cancellation Algorithm (ECA) to extract static signals. By leveraging the time-invariant property of static signals, we compensate for the asynchronous clocks of the transceivers without compromising the linear property of the received signal. For multi-target scenarios, we decompose the received signal in the sub carrier domain using Independent Component Analysis (ICA) and demonstrate the feasibility of this approach. Simulation and experimental results show that the proposed method achieves a wider sensing coverage when only one pair of transceiver nodes is used, up to a maximum sensing distance of 20 meters with a relative error limit of 10%. When limited to the same absolute error constraint (0.5 bpm), the proposed scheme achieves about 40 % improvement in sensing range compared to other sensing schemes.