Cyclostationary modelling of amplitude and frequency modulated signals in heart and respiration monitoring Doppler radar systems

Abstract

Unobtrusive and continuous monitoring of cardiopulmonary activity at a distance provides a potential tool in making health care and emergency delivery more efficient. Doppler radar remote sensing of vital signs has shown promise to this end, with proof of concept demonstrated for various applications such as more comfortable monitoring for many people. In this study, the authors present a second-order cyclostationary model for the returned amplitude and frequency modulated radar signal, which is contaminated with various noises, body motion artefacts and dc offset, to extract vital signs. They validate their model with confidence interval extraction from 500 simulation runs and also with 50 experiment runs on a female student using a miniaturised non-contact heartbeat and respiration monitoring radar system based on the quadrature Doppler effect.