A Robust Super-Resolution Algorithm in a Low SNR Environment for Vital Sign Radar

We propose a robust super-resolution algorithm for vital sign radar in a low signal-to-noise ratio (SNR) environment. Conventional approaches, such as fast Fourier transform and super-resolution based algorithms, suffered to provide reliable results due to the limited data length and high noise level. To overcome these limitations, our proposed algorithm utilizes a low-complexity least mean square (LMS) filter and relaxation (RELAX) techniques to achieve robust performance in low SNR environments. To evaluate the effectiveness of our algorithm, we conducted both simulation and experimental studies. Our results show that the proposed method significantly outperforms conventional methods, with Monte-Carlo simulations of respiration and heartbeat achieving an RMSE approximately 7 and 120 times lower than that of the conventional method, respectively. Overall, our algorithm provides a promising solution for robust vital sign detection in challenging low SNR environments.