Radar Signal Decomposition of Multi-Human Respiration Signs with Reduced Antenna Array

In this paper, we propose a beamforming method that can effectively decompose multi-human vital signs. Each human target at the same distance from the radar is localized using radar processing, which includes static clutter removal, signal compression and angle estimation. Subsequently, a signal from a known angle is decomposed by the beamforming technique using receiving antenna array. To ensure efficient decomposition without interference from unwanted targets, the proposed method designs beamforming weights to nullify the beamforming gain toward unwanted targets. In addition, it maximizes the beamforming gain in a desired direction where a human target locates. As a result, we verified that the proposed method shows a higher success probability of respiration detection compared to conventional methods. Furthermore, the proposed method was effective although a reduced antenna array is used. Thus, we expect that it will contribute to building a low-cost Internet of things service for vital signs extraction based on the radar.