Algorithm for the operation of a software-hardware subsystem for biometric identification based on keystroke dynamics analysis

This paper presents an algorithm for the operation of a software-hardware subsystem for biometric identification based on keystroke dynamics analysis. The system supports identification mode (1:N) and verification mode (1:1) in accordance with GOST R 54412–2019. A review of current scientific literature indicates that biometric systems can utilize various features, such as typing speed or digraph entry times. It was found that hardware solutions improve the accuracy of capturing time intervals between consecutive key presses (digraphs); however, no existing solutions comply with GOST R 54412–2019. The proposed algorithm ensures a complete processing cycle within a distributed architecture comprising a client, a server, and an Arduino-based hardware module. The model was evaluated for compliance with standard requirements, demonstrating robust performance on the ATmega32U4 platform. Integration efficiency into biometric systems is shown through support for both online and offline modes. Comparison time is ≤ 190 ms, with memory consumption of approximately 1.9 KB. The applicability of the model in signal processing and decision-making subsystems using distribution similarity metrics is described. These results can be employed in developing GOST-compliant biometric identification systems that secure access without modifying client operating systems.

1. Shklyar E.V., Shulzhenko A.D. The mathematical model of keystroke dynamics biometric reference. Modeling, Optimization and Information Technology. 2026;14(1). (In Russ.). https://doi.org/10.26102/2310-6018/2026.52.1.001

2. Dovgal V.A. Features of obtaining information on parameters of keystroke dynamics. Bulletin of the Adygea State University. Series: Natural-Mathematical and Technical Sciences. 2017;(2):102–108. (In Russ.).

3. Ivanov D., Nikitin A. Counteraction against keyboard handwriting analysis. Bulletin of the RSUH. Series: Documentation and Archival Science. Computer Science. Information Protection and Information Security. 2014;(11):178–183. (In Russ.).

4. Sharipov R.R., Katasev A.S., Kirpichnikov A.P. Methods of analyzing users' keyboard handwriting using reference Gaussian signals. Herald of Technological University. 2016;19(13):157–160. (In Russ.).

5. Ulinskas M., Woźniak M., Damaševičius R. Analysis of Keystroke Dynamics for Fatigue Recognition. In: Computational Science and Its Applications – ICCSA 2017: Proceedings: Part V: 17th International Conference, 03–06 July 2017, Trieste, Italy. Cham: Springer; 2017. P. 235–247. https://doi.org/10.1007/978-3-319-62404-4_18

6. Monaco J.V., Tappert Ch.C. The partially observable hidden Markov model and its application to keystroke dynamics. Pattern Recognition. 2018;76:449–462. https://doi.org/10.1016/j.patcog.2017.11.021

7. Shadman R., Wahab A.A., Manno M., et al. Keystroke Dynamics: Concepts, Techniques, and Applications. ACM Computing Surveys. 2025;57(11). https://doi.org/10.1145/3733103

8. Senerath D., Tharinda S., Vishvajith M., et al. BehaveFormer: A Framework with Spatio-Temporal Dual Attention Transformers for IMU-Enhanced Keystroke Dynamics. In: 2023 IEEE International Joint Conference on Biometrics (IJCB), 25–28 September 2023, Ljubljana, Slovenia. IEEE; 2023. P. 1–9. https://doi.org/10.1109/IJCB57857.2023.10448997

9. Shklyar E.V. An Algorithm for Generating Word Lists with a Specified Bigram Distribution for Keystroke Dynamics Biometric Template Registration. IT Security (Russia). 2025;32(3):74–89. (In Russ.).

10. Roy S., Pradhan J., Kumar A., et al. A Systematic Literature Review on Latest Keystroke Dynamics Based Models. IEEE Access. 2022;10:92192–92236. https://doi.org/10.1109/ACCESS.2022.3197756