An Adaptive Heart Rate Monitoring System Based on Algebraic Distance Minimization

• Published in: IEEE access Vol. 12; pp. 134366 - 134378
• Main Authors: Li, Mu, Yang, Yangyang, Ke, Xizheng
• Format: Journal Article
• Language: English
• Published: IEEE 2024
• Subjects: adaptive filtering; Biomedical monitoring; Biosensors; FMCW radar sensor; Harmonic analysis; harmonic elimination; Heart beat; Heart rate; heart rate variability; Minimization; minimization of algebraic distance; Power harmonic filters; Radar; Radar antennas
• ISSN: 2169-3536; 2169-3536
• DOI: 10.1109/ACCESS.2024.3462103

A novel adaptive heart rate monitoring and filtering system based on algebraic distance minimization has been designed. This system uses frequency-modulated continuous-wave radar for non-contact monitoring of parameters such as respiration, heart rate, and heart rate variability (HRV). By employing an improved algebraic distance minimization technique, the system suppresses static noise from the environment, enhances the signal-to-noise ratio of the intermediate-frequency signal, and reduces the complexity of subsequent signal processing. To address issues such as low-frequency interference from radar signals reflecting off fixed targets around the subject and the impact of respiratory harmonics on heart rate measurement, this study introduces a respiratory harmonic filter and an adaptive notch filter to eliminate higher-order respiratory harmonic interference. This approach allows for the adaptive decomposition and reconstruction of vital sign signals and final estimation of respiratory heart rate based on the relationship between harmonics. Comparative experiments with intensive care unit medical equipment have validated the superiority and robustness of the proposed algorithm. Experimental results show that the RMSE and MRE of the method are 2.66% and 2.7%, respectively. After multiple measurements across various subjects, the method achieved an average accuracy of 96.5%, demonstrating higher measurement accuracy than other methods in all cases. Additionally, the algorithm also enables the detection of heart rate variability.