A real-time heart rate analysis for a remote millimeter wave I-Q sensor

Area of Science:

• Biomedical Engineering
• Remote Sensing
• Signal Processing

Background:

• Remote vital sign monitoring using millimeter wave (mmW) sensors offers non-contact physiological assessment.
• Millimeter wave (mmW) radar signals are susceptible to noise and artifacts, necessitating robust signal processing for accurate vital sign extraction.
• Traditional methods for extracting heart rate (HR) from mmW signals can be complex and may struggle with low signal-to-noise ratios (SNR).

Purpose of the Study:

• To develop and evaluate a novel parameter optimization method for analyzing heart rate (HR) from millimeter wave (mmW) I-Q sensor data.
• To improve the accuracy and robustness of remote vital sign monitoring, particularly in challenging low SNR conditions.
• To enable the identification of beat-to-beat HR and individual heartbeat magnitude for potential medical diagnostic applications.

Main Methods:

• Utilized a 94 GHz millimeter wave (mmW) I-Q sensor to collect physiological tracings.
• Applied a parameter optimization method based on the nonlinear Levenberg-Marquardt algorithm.
• Directly fitted the real (I) and imaginary (Q) parts of the mmW signal, avoiding phase unwrapping, and compared results with the discrete Fourier transform (DFT).

Main Results:

• The developed method successfully processed mmW radar signals to obtain true HR, overcoming challenges posed by respiration, movement, and noise.
• Direct fitting of I-Q signal components proved effective, especially in low signal-to-noise ratio (SNR) scenarios, circumventing phase unwrapping.
• The method demonstrated the capability to identify both beat-to-beat HR and individual heartbeat magnitude, showing comparable mean HR to DFT.

Conclusions:

• The proposed method offers a robust and effective approach for extracting heart rate (HR) from millimeter wave (mmW) sensor data.
• This technique enhances the feasibility of remote biometric monitoring and has potential applications in medical diagnostics.
• The ability to handle low SNR conditions and extract detailed heartbeat information represents a significant advancement in non-contact vital sign monitoring.