Rhythm-adaptive signal processing for effective ECG and PPG-based authentication under dynamic physiological...

Area of Science:

• Biometrics and Signal Processing
• Wearable Technology and Health Monitoring

Background:

• Biometric authentication using electrocardiogram (ECG) and photoplethysmogram (PPG) signals is challenged by heart rate variability and morphological changes during physical activity.
• Traditional methods struggle with the reduced reliability of stationary cardiac representations under dynamic conditions.

Purpose of the Study:

• To develop and evaluate a rhythm-adaptive framework for ECG- and PPG-based biometric authentication robust to mild physical activity.
• To enhance the reliability of cardiac signal representations by modeling them as cyclic random processes (CRPs) with variable rhythm.

Main Methods:

• Modeled ECG and PPG signals as cyclic random processes (CRPs) with variable rhythm.
• Utilized the estimated rhythm function to align cardiac cycles before extracting compact statistical features.
• Evaluated the framework on the PhysioNet Pulse Transit Time PPG Dataset (PTTPD) using walking records.

Main Results:

• For ECG, Linear SVM achieved perfect scores: Accuracy, Balanced Accuracy, and F1-score of 1.000, with 0.000 for False Rejection Rate (FRR), False Acceptance Rate (FAR), and HTER.
• For PPG, Nearest Neighbors achieved high performance: Accuracy = 0.962, Balanced Accuracy = 0.961, F1-score = 0.960, with FRR = 0.057, FAR = 0.020, and HTER = 0.039.
• The rhythm-adaptive processing effectively stabilized cardiac-cycle morphology during movement.

Conclusions:

• The proposed CRP-based rhythm-adaptive framework provides a stable and interpretable method for ECG- and PPG-based authentication during physical activity.
• This approach offers a viable alternative to purely data-driven representations for wearable and telemedicine authentication systems.