ANALYSIS OF AUTHENTICATION MODELS AND ALGORITHMS BASED ON BIOMETRIC DATA

DOI: 10.31673/2409-7292.2025.022701

Abstract

User authentication is one of the key aspects of information security, which provides access control to resources and
protection of confidential data. Traditional authentication methods, such as passwords and PIN codes, have a number of
shortcomings, in particular, vulnerability to attacks such as brute force, phishing, and interception. In this regard, there is
growing interest in biometric authentication methods that provide a higher level of security and ease of use. The article considers
modern models and algorithms of biometric authentication, their advantages and disadvantages. The features of the use of
unimodal and multimodal systems are analyzed. Special attention is paid to promising methods for increasing the accuracy of
authentication and the security of biometric data storage. An analysis of modern research in this area is presented. The main
algorithms used in biometric authentication systems are also considered, including image processing methods, neural networks,
machine learning, and cryptographic technologies. The possibilities of using multi-factor authentication, which combines
biometric parameters with other methods of identity verification, which significantly increases the level of security, are
analyzed. The prospects for the development of biometric authentication systems are considered, in particular, the introduction of new technologies, such as artificial intelligence, blockchain and quantum cryptography. An analysis of possible risks
associated with biometric authentication is carried out. It is concluded that the use of biometric methods allows to significantly
increase the efficiency of authentication, reduce the risks of data compromise and ensure convenience for users, however, their
implementation requires taking into account issues of confidentiality, reliability and legal regulation. An authentication system
based on artificial intelligence, blockchain, quantum cryptography is proposed and its effectiveness is analyzed.
Keywords: biometric authentication, unimodal systems, multimodal systems, artificial intelligence, cryptography,
steganography.

References
1. Ганін, І. В., & Ковальчук, О. П. (2021). Сучасні методи біометричної ідентифікації. Вісник
Національного технічного університету України "КПІ". Серія: Інформаційна безпека, (3), 26-32. Отримано з
https://ela.kpi.ua/bitstream/123456789/9839/1/26.pdf.
2. Коваленко, Р. С., & Ігнатенко, Т. В. (2022). Вибір переважного методу біометричної автентифікації.
Інформаційна безпека та кіберзахист, 5(12), 44–57. Отримано з https://isg-journal.com/isjea/article/download
/444/246/457.
3. Руда, Х., Сабодашко, Д., Микитин, Г., Швед, М., Бордуляк, С., & Коршун, Н. (2024). Порівняння
методів цифрової обробки сигналів та моделей глибинного навчання у голосовій аутентифікації. Кібербезпека:
освіта, наука, техніка, 1(25), 140–160. https://doi.org/10.28925/2663-4023.2024.25.140160.
4. Kaur, P., Sharma, M., & Sharma, N. (2020). A robust multimodal biometric authentication system using deep
learning. Expert Systems with Applications, 157. https://doi.org/10.1016/j.eswa.2020.113486.
5. Liu, X., Yin, F., Wang, L., & Xu, W. (2023). Deep Learning in Biometrics: A Review. Pattern Recognition
Letters, 45(3), 128–140. https://doi.org/10.1016/j.patrec.2022.10.015.
6. Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep Learning. MIT Press.
7. He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep Residual Learning for Image Recognition. Proceedings of
the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 770–778. https://doi.org/10.1109
/CVPR.2016.90.
8. Smith, P., & Jones, R. (2021). Spoofing Attacks in Biometric Systems: Detection and Prevention. Journal of
Cybersecurity, 8(3), 210–225. https://doi.org/10.1093/cybsec/tyab012.
9. Скорик, Ю., & Безрук, В. (2023). Вибір переважного методу біометричної автентифікації. International
Science Journal of Engineering & Agriculture, 2(4), 28–34.
10. Іосіфов, Є., & Соколов, В. (2024). Порівняльний аналіз методів, технологій, сервісів та платформ для
розпізнавання голосової інформації в системах забезпечення інформаційної безпеки. Кібербезпека: освіта, наука,
техніка, 1(25), 468-486.
11. Ledig, C., et al. (2017). Photo-realistic single image super-resolution using a generative adversarial network.
Proceedings of the IEEE conference on computer vision and pattern recognition (CVPR), 4681-4690.
https://doi.org/10.1109/CVPR.2017.19.