EEMD-MICA Based Heart Rate Extraction Algorithm for Radar Signals

Millimeter wave radar-based vital sign monitoring technology can measure the target human heart rate in non-contact scenarios. A heart rate extraction algorithm for radar signals is proposed to address the challenges of noise cancellation and harmonic suppression. In the chest localization phase, the vital sign signal’s periodicity and the noise’s non-periodicity are analyzed. An average abatement method is designed to eliminate environmental multipath effects, static object clutter, and the direct current (DC) components generated after the Fourier transform. In the phase extraction stage, phase unwrapping and interpolation smoothing are used to extract high-quality phase signals. In the signal separation stage, Ensemble Empirical Mode Decomposition (EEMD) and Modified Independent Component Analysis (MICA) are combined. Through suppressing the masking of heartbeat frequencies by respiratory harmonic frequencies, the combined algorithm obtains a more accurate heart rate signal while ensuring the anti-noise performance. The experimental results show that the EEMD-MICA-based algorithm has a higher heart rate extraction accuracy than the band-pass filtering, wavelet decomposition, and modal decomposition algorithms.