A PRISMA-based systematic review on advances in identity recognition and authentication using human biometric signals...

Area of Science:

• Biometrics
• Signal Processing
• Machine Learning

Background:

• Physiological biometric signals are increasingly used in authentication systems.
• Existing research shows varied performance across different biometric modalities and methods.
• A systematic review is needed to consolidate findings from recent literature.

Purpose of the Study:

• To systematically review the effectiveness of physiological biometric signals in authentication and recognition systems.
• To evaluate different biometric modalities, datasets, signal processing, and classification approaches.
• To compare unimodal and multimodal biometric system performance.

Main Methods:

• Systematic literature review following PRISMA 2020 guidelines.
• Searched major databases (EBSCO, PubMed, IEEE Xplore, Scopus, Web of Science) for studies from 2018-2023.
• Included 80 articles after screening 2,064 records.

Main Results:

• ECG-based systems report high average accuracy (98.6%).
• Multimodal biometric systems achieve accuracy levels exceeding 99%.
• Deep learning approaches generally outperform traditional machine learning methods.

Conclusions:

• Biometric signal-based authentication systems show strong performance.
• Multimodal fusion and deep learning are promising but require standardized evaluations.
• Future research should focus on diverse datasets and optimized signal processing for real-world applicability.