Accurate Measurement in Doppler Radar Vital Sign Detection Based on Parameterized Demodulation

• Published in: IEEE transactions on microwave theory and techniques Vol. 65; no. 11; pp. 4483 - 4492
• Main Authors: Xiong, Yuyong, Chen, Shiqian, Dong, Xingjian, Peng, Zhike, Zhang, Wenming
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.11.2017; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Baseband; Computer simulation; Computing time; Demodulation; Digital signal processors; Doppler radar; Fast Fourier transformations; Heart beat; Kernel functions; nonlinear phase modulation; Optimization; parameter estimation; Parameterization; parameterized demodulation (PD); Parameters; Phase demodulation; Radar detection; Respiration; Signal processing; Signal processing algorithms; Time-frequency analysis; Trigonometric functions; vital sign detection
• DOI: 10.1109/TMTT.2017.2684138
• ISSN: 0018-9480

Utilizing Doppler radar to conduct noncontact vital sign detection has attracted growing interest in recent years. Aiming to extract the vital sign information from the baseband signal effectively and accurately, a novel signal processing method based on parameterized demodulation (PD) is proposed. To effectively characterize the baseband signal whose phase consists of two oscillating components (i.e., the respiration and heartbeat components), the proposed algorithm defines a demodulation operator with sine kernel functions and formulates the phase demodulation as a parameter optimization problem. To increase the computational efficiency of the algorithm, the parameters corresponding to the respiration and heartbeat components are estimated sequentially. Specifically, the respiration component is first estimated and removed from the phase of the baseband signal, and then, the heartbeat component is extracted from the residual signal. Compared with the existing methods, the proposed algorithm is free of the resolution problem in fast Fourier transform-based methods with a limited data length and can obtain accurate rate tracking in a noisy environment. Both simulated and experimental results are provided to demonstrate the advantages and effectiveness of the proposed method for accurate noncontact vital sign detection.