An emotion recognition method based on frequency-domain features of PPG

• DOI: 10.3389/fphys.2025.1486763
• Publication Date: 2025-02-25T06:54:53Z
• Authors: Zhibin Zhu; Xuanyi Wang; Yifei Xu; Wanlin Chen; Jing Zheng; Shulin Chen; Hang Chen
• Keywords: photoplethysmography (PPG); Physiology; emotion recognition; QP1-981; support vector machine (SVM); dual windkessel model; PPG frequency-domian analysis

This study aims to employ physiological model simulation systematically analyze the frequency-domain components of PPG signals and extract their key features. The efficacy these features in effectively distinguishing emotional states will also be investigated. A dual windkessel was employed signal frequency distinctive Experimental data collection encompassed both (PPG) psychological measurements, with subsequent analysis involving distribution patterns statistical testing (U-tests) examine feature-emotion relationships. implemented support vector machine (SVM) classification evaluate feature effectiveness, complemented by comparative using pulse rate variability (PRV) features, morphological DEAP dataset. results demonstrate significant differentiation responses arousal valence variations, achieving accuracies 87.5% 81.4%, respectively. Validation on dataset yielded consistent 73.5% (arousal) 71.5% (valence). Feature fusion incorporating proposed enhanced performance, surpassing 90% accuracy. uses modeling We effectiveness emotion recognition reveal relationships among parameters, states. These findings advance understanding mechanisms provide a foundation for future research.