A Novel Scheme for High-Accuracy Frequency Estimation in Non-Contact Heart Rate Detection Based on Multi-Dimensional...

Area of Science:

• Biomedical Engineering
• Radar Signal Processing
• Remote Sensing

Background:

• Millimeter-wave (mmWave) radar offers non-contact vital sign monitoring.
• Challenges include weak signals, interference, and achieving high-precision, super-resolution frequency estimation.
• Existing methods like Fast Fourier Transform (FFT) have limitations in accuracy and computational load.

Purpose of the Study:

• To develop a novel heart rate detection scheme for mmWave radar vital sign monitoring.
• To address challenges of weak signals, interference, and computational constraints.
• To achieve high-precision and super-resolution heart rate estimation.

Main Methods:

• Proposed a multi-dimensional coherent accumulation (MDCA) method to enhance signal-to-noise ratio (SNR) using spatial and temporal information.
• Applied the fast iterative interpolated beamforming (FIIB) algorithm for super-resolution frequency estimation with low computational complexity.
• Introduced a reordering strategy to mitigate FIIB misjudgments and analyzed key FIIB parameters (L, Q).

Main Results:

• The FIIB algorithm improved heart rate estimation accuracy by 1.08 beats per minute (bpm) compared to FFT.
• Achieved a root mean square error (RMSE) of less than 1.12 bpm on average at 1 meter distance.
• Demonstrated high accuracy and robust performance in extensive subject experiments.

Conclusions:

• The proposed MDCA and FIIB framework effectively enhances mmWave radar-based heart rate detection.
• The approach overcomes critical challenges, offering a precise and computationally efficient solution.
• Validated accuracy and robustness for practical vital sign monitoring applications.