Discovering the Unseen: Radar-Based Estimation of Heartbeat, Breathing Rate, and Underlying Muscle Expansion Without Probes

Abstract

Effective monitoring of vital signs is a fundamental aspect of healthcare. To measure vital signs, patients often hesitate to wear probes and body-worn sensors for extended periods because these devices can limit their movement and cause discomfort. In this study, we present three radar-based techniques to estimate vital signs and underlying muscle expansion. The first method employs a short-time Fourier transform (STFT), but it has limitations due to its fixed resolution and its performance dependency on the carrier frequency. The second method modifies the Hilbert-Huang transform (HHT) to address the mode-mixing problem. The HHT breaks down the signal into its fundamental components. By subsequently applying Fourier transform and signal filtering, we demonstrate its feasibility of estimating heartbeat and breathing rates. In our latest method, which constitutes the primary contribution of this study, we exploit the repetitive patterns inherent in both heartbeat and breathing signals. This involves representing the spectrum of the received signal as a discrete frequency spectrum and subsequently applying harmonic accumulation. Our simulation results consistently demonstrate that the harmonics accumulation (HA) algorithm outperforms other algorithms in terms of accuracy and effectiveness. To assess the performance of our suggested algorithms, we derive the Cramér-Rao lower bound (CRLB) as a benchmark. Our results show the effectiveness of the proposed methods.