АРХИТЕКТУРА СИСТЕМЫ КЛИЕНТ-СЕРВЕРНОЙ БИОМЕТРИЧЕСКОЙ ИДЕНТИФИКАЦИИ ЛИЧНОСТИ В РАСПРЕДЕЛЕННЫХ ИНФОРМАЦИОННЫХ СИСТЕМАХ

Вера Аркадьевна Частикова; Сергей Анатольевич Жерлицын; Александр Самвелович Макарян

doi:10.17308/sait/1995-5499/2026/1/167-178

Authors

Vera A. Chastikova Kuban State Technological University
Sergey A. Zherlitsyn Kuban State Technological University
Alexander S. Makaryan Kuban State Technological University

DOI:

https://doi.org/10.17308/sait/1995-5499/2026/1/167-178

Keywords:

biometric identification, information security, neural networks, artificial intelligence, client‑server architecture, distributed systems

Abstract

Purpose. This study develops a secure client-server architecture for biometric personal identification in geographically distributed information systems. Relevance stems from the absence of a unified design methodology and the threat of biometric template interception in open networks. Materials. The work considers regulatory requirements, related research, threat modelling, and sequence diagrams for registration, authentication and record‑deletion procedures. Methods. The secret‑separation principle is applied: a biometric client captures a sample, performs neural‑network feature extraction and encrypts the resulting Biometric Feature Vector with a one‑time symmetric key; server‑side processing is distributed among an analysing service, an account service and a coordinating centre that communicate over protected channels. Results. Excluding raw biometric storage and physically separating biometric and personal data prevents their simultaneous compromise; interception of an encrypted feature vector does not expose personal data, while the record‑deletion workflow guarantees user control over information. Additional design features support horizontal scalability and hot‑swapping of neural models, facilitating integration with external trusted identity providers and heterogeneous infrastructures. Conclusions. The proposed architecture offers a comprehensive, scalable and regulation‑compliant platform suitable for governmental and corporate systems that demand legally significant and reproducible biometric identification; it closes gaps highlighted in prior work and establishes a foundation for further research on secure biometrics.

Author Biographies

Vera A. Chastikova, Kuban State Technological University

PhD in Technical Sciences, Associate Professor
Sergey A. Zherlitsyn, Kuban State Technological University

Postgraduate Student
Alexander S. Makaryan, Kuban State Technological University

PhD in Technical Sciences, Associate Professor, Head of Department of Cybersecurity and Information Protection

References

1. Aftab A., Khan F. A., Khan M. K. [et al.] Hand-based multibiometric systems: state-ofthe-art and future challenges // PeerJ Computer Science. – 2021. – 7:e707. DOI: 10.7717/peerjcs.707.

2. Федеральный закон от 27.07.2006 г. № 152-ФЗ «О персональных данных».

3. Abdullahi S. M., Sun S., Wang B. [et al.] Biometric template attacks and recent protection mechanisms: a survey // Information Fusion. – 2024. – Vol. 103. – 102144. DOI: 10.1016/j.inffus.2023.102144.

4. Arjona R., López-González P., Román R., Baturone I. Post-Quantum Biometric Authentication Based on Homomorphic Encryption and Classic McEliece // Applied Sciences. – 2023. – Vol. 13, No 2. – P. 757. DOI: 10.3390/ app13020757.

5. Saračevič M., Elhoseny M., Selimi A., Lončeravič Z. Possibilities of Applying the Triangulation Method in the Biometric Identification Process // В кн.: Biometric Identification Technologies Based on Modern Data Mining Methods; под ред. S. Bilan, M. Elhoseny, D. J. Hemanth. Cham: Springer, 2021. – P. 1–17. DOI: 10.1007/978-3- 030-48378-4_1.

6. Ghaeminia M. H., Shokouhi S. B., Amir_ khani A. Biometric Gait Identification Systems: From Spatio-Temporal Filtering to Local PatchBased Techniques // В кн.: Biometric Identification Technologies Based on Modern Data Mining Methods. Cham: Springer, 2021. – P. 19–37. – DOI: 10.1007/978-3-030-48378-4_2.

7. Chater A., Lasfar A. Automated Calculation of Fundamental Matrix from Stereo Imagesfrom a Different Point of View // В кн.: Biometric Identification Technologies Based on Modern Data Mining Methods. Cham: Springer, 2021. – P. 105–118. DOI: 10.1007/978-3-030-48378-4_7.

8. Akhtar Z., Buriro A. Multitrait Selfie: LowCost Multimodal Smartphone User Authentication // В кн.: Biometric Identification Technologies Based on Modern Data Mining Methods. Cham: Springer, 2021. С. 159–175. DOI: 10.1007/978-3-030-48378-4_11.

9. Motornyuk R., Bilan A., Bilan S. Research of Biometric Characteristics of the Shape of the Ears Based on Multi-Coordinate Methods // В кн.: Biometric Identification Technologies Based on Modern Data Mining Methods. Cham: Springer, 2021. – P. 177–194. DOI: 10.1007/978- 3-030-48378-4_12.

10. Venkatachalam K., Vinoth Kumar M. [et al.] Secure biometric authentication with de-duplication on distributed cloud storage // PeerJ Computer Science. – 2021. – 7:e569. DOI: 10.7717/peerj-cs.569.

11. Kim J., Jung Y. G., Teoh A. B. J. Multimodal Biometric Template Protection Based on a Cancelable SoftmaxOut Fusion Network // Applied Sciences. – 2022. – Vol. 12, No 4. – P. 2023. DOI: 10.3390/app12042023.

12. Arjona R., Baturone I. A Post-Quantum Biometric Template Protection Scheme Based on Learning Parity with Noise Commitments // IEEE Access. – 2020. – Vol. 8. – P. 182355– 182365. DOI: 10.1109/ACCESS.2020.3028703.

13. Федеральный закон от 29.12.2020 г. № 479-ФЗ «О внесении изменений в отдельные законодательные акты Российской Федерации».

14. Федеральный закон от 29.12.2022 г. № 572-ФЗ «Об осуществлении идентификации и (или) аутентификации физических лиц с использованием биометрических персональных данных…».

15. Acquah M. A., Chen N., Pan J.-S., Yang H.-M., Yan B. Securing Fingerprint Template Using Blockchain and Distributed Storage System // Symmetry. – 2020. – Vol. 12, No 6. – P. 951. DOI: 10.3390/sym12060951.