Abstract: ‍ ‍Non-contact vital sign monitoring can be used in health care， ambient assisted living， automobile transportation， search and rescue， and public safety. Radar-based non-contact vital-sign sensing has been proven to be effective and suitable for continuous monitoring. It does not require the users to be exposed to any camera， which protects their privacy. In this work， we developed a reconfigurable intelligent surface four-dimensional （RIS-4D） bioradar， and proposed a corresponding method to simultaneously localize multiple human targets and monitor their corresponding respiratory and heartbeat micromotion signals. It has the advantages of low cost， low energy consumption， programmability， and easy deployment. The RIS-4D bioradar employs an RIS to dynamically manipulate electromagnetic waves for beamforming and three-dimensional radar imaging. Thus， it can localize multiple human targets and extract respiratory and heart signals from the continuous echoes in the one-dimensional time dimension. The RIS-4D bioradar utilizes the RIS as a surface array antenna to resolve the azimuth and height， and transmits a broadband signal to achieve a high resolution range. Thus， it is capable of 4D sensing during continuous-time detection. The 4D space-time high-resolution information from three spatial dimensions and one temporal dimension allows the RIS-4D bioradar to perceive the target and environment in multiple dimensions. Spatial beamforming focuses the electromagnetic wave energy on the chest area of the human body， which helps to reduce the interference and distortion caused by clutter （such as motion artifacts induced by the limbs and environmental multipath clutter）. The beamforming significantly improves the detection performance of the bioradar and enables the bioradar to distinguish multiple targets spatially. The higher-order cumulant is applied to the radar image sequence to detect multiple subjects. The variational mode separation algorithm is employed to adaptively separate the respiratory and heartbeat waveforms from the micro-motion signal extracted through the phase of the radar image. These signal processing steps for the RIS-4D bioradar effectively realize the localization of multiple subjects， as well as the monitoring of vital signs from each subject. The proposed system was verified by obtaining data in a real indoor office environment， and the experimental results showed that the system could accurately localize and monitor the respiratory and heartbeat signals of multiple subjects. The monitoring results for the instantaneous heart rate and heart rate variability further verified the system's effectiveness. The proposed RIS-4D bioradar provides a multi-target， multi-functional， and high-precision method for non-contact physiological signal acquisition.