Datasets
Standard Dataset
ASVME
- Citation Author(s):
- Submitted by:
- zhen yang
- Last updated:
- Mon, 10/21/2024 - 05:57
- DOI:
- 10.21227/xy72-tp37
- License:
99 Views- Categories:
- Keywords:
Abstract
The breath rate (BR), heart rate (HR), breathing-breathing interval (BBI) and heart rate variability (HRV) are the critical vital sign parameters. In this article, a novel method named adaptive separation variational mode extraction algorithm (ASVME) is proposed to accurately monitor multi-variable vital signs (MVVS) at the same time with a frequency-modulated continuous wave (FMCW) radar system in practical scenarios. Firstly, a minimum variance distortionless response (MVDR) spectrum estimation algorithm is proposed to accurately locate respiratory and heartbeat components, which can effectively restrain the influence of respiratory harmonics on HR and R-R intervals (RRI) measurements. Subsequently, an adaptive variational mode extraction (AVME) algorithm is proposed to accurately extract respiratory waves and heartbeat waves after accurate frequency location.
The proposed AVME is an adaptive parameter optimization technique of the variational mode extraction (VME) algorithm based on the high correlation between the extracted signal and the modeled signal as well as the minimum energy loss rate of the extracted signal. Furthermore, an additional sliding window extraction method is proposed in the ASVME method, which further improves the accuracy and stability of the detection of MVVS. The method is verified in a variety of experimental scenarios. The experimental results indicate that the ASVME can accurately detect multi-variable vital signs, with the RMSE of the BR and HR demonstrating a high accuracy of 0.46 bpm and 0.80 bpm, respectively, and the RMSE of BBI and RRI being 0.059 s and 0.0046 s, respectively.
