PPG-based biometric authentication: A review on architectures, datasets, attacks and security challenges

Recent Photoplethysmography (PPG-based) authentication approaches are emerging as promising biometric modalities, driven by the widespread adoption of wearable devices capable of noninvasive and continuous identity verification. These advances have also introduced new challenges and threats to security. This review presents a taxonomy of PPG-based authentication and identification systems, emphasizing a comparative analysis of different authentication models using PPG signals, in terms of performance, preprocessing algorithms, sensors and hardware platforms, as well as security. The lack of standardized datasets, which limits reproducibility and generalization across demographics and real-world conditions, is discussed. Unlike other studies, this work focuses on the impact and mitigation strategies of the main attacks, such as spoofing, replay, and presentation, in PPG-based authentication systems. Additionally, based on the gaps and challenges identified, a security analysis is presented, emphasizing standardized metrics such as False Acceptance Rate (FAR), False Rejection Rate (FRR), and Equal Error Rate (EER), and a dataset standardization for scalability. Finally, this research highlights new challenges and future work, such as the integration of robust ML architectures with standardized datasets and security evaluations to enable reliable PPG-based authentication systems in real-world applications.