Assessment of Doppler Radar Demodulation Techniques for Accurate Characterization of Cardiorespiratory Signals

Doppler radar systems can demonstrably provide accurate measurement of physiological motion for sedentary and partially sedentary subjects. In this research, quadrature and single-channel demodulation techniques are evaluated and their performance is compared for the extraction of vital-signs-related displacement and rate information from clutter and noise. Computational complexity, degree of displacement waveform correlation, and spectral analysis resolution are taken into account. The demodulation techniques were tested by synthesizing realistic cardiorespiratory motion using a robotic mover, and with human data. Arctangent demodulation (AD) and linear demodulation (LD) were demonstrated to be computationally more efficient than the other methods, and AD provided phase information needed for accurate quantitative tracking of the measured displacement waveform. Divide and cross multiply (DACM) methods were found to have a greater resilience against noise when recovering cardiac rate information, while discrete cosine transform (DCT) methods demonstrated greater frequency resolution over shorter segments of data when making respiratory measurements. These findings provide important guidance when selecting a demodulation method that is best suited for a particular measurement system or application.