Leveraging mmWave for Contactless Breath Rate Estimation of Moving Subjects

The breath rate (BR) measurement is fundamental for comprehensive human health monitoring in many scenarios. Accurately estimating BR, particularly in moving subjects, poses several challenges, including the necessity for direct physical contact with the individual or constraints on the individual's movements and mobility. In this paper, a methodology is proposed that employs frequency-modulated continuous wave (FMCW) radar, operating at a frequency of 77 GHz, to estimate the BR of subjects moving freely within an environment. The proposed methodology features a pre-processing pipeline designed to refine the radar's captured signals. The processed signals are used to accurately track the subject, and estimate their corresponding BR. Preliminary outcomes of our investigation indicate that the BR estimates generated by our system exhibit a mean absolute error (MAE) of 2.48 breaths per minute, as averaged across four subjects when compared with the ground truth.