Emotion-Aware Multimodal Biometric Identification by Using Biosignals

The advancement of biometric identification systems has become increasingly crucial in modern applications, necessitating secure and user-friendly methods that adapt to evolving security and privacy standards. This study explores the integration of physiological signals and emotional states to enhance the robustness and accuracy of biometric identification mechanisms. Utilizing the largest publicly available multimodal biosignal datasets, we compared the performance of various machine learning algorithms and assessed the contribution of different biosignal modalities to identification success rates. Additionally, the impact of incorporating emotional states into the identification process was examined, revealing that emotional data significantly improves recognition performance. T his work is pioneering in its investigation of emotional states as input for biometric systems, offering a novel approach to multimodal biometric identification that could potentially improve security practices by providing a more personalized and secure user experience.