A combined method of formation of a cryptographic key with secret modification of the results of synchronization of artificial neural networks
https://doi.org/10.21122/2309-4923-2021-3-51-58
Abstract
This article discusses one of the ways to generate a common cryptographic key using synchronized artificial neural networks. This option is based on a combined method of forming a cryptographic key [1]. The proposed combined formation consists of two stages: the formation of partially coinciding binary sequences using synchronized artificial neural networks and the elimination of mismatched bits by open comparison of the parities of bit pairs. The purpose of this article is to increase the cryptographic strength of this method in relation to a cryptanalyst. In this regard, it is proposed to prematurely interrupt the synchronization process at the first stage of the combined method and make changes to the resulting binary sequence by randomly inverting a certain number of bits. To confirm the quality of this method, possible attacks are considered and the scale of enumeration of possible values is illustrated. The results obtained showed that the combined method of forming a cryptographic key with a secret modification of the synchronization results of artificial neural networks, proposed in this article, provides its high cryptographic strength, commensurate with the cryptographic strength of modern symmetric encryption algorithms, with a relatively simple implementation.
About the Author
M. L. Radziukevich
Research Institute for the Technical Protection of Information
Belarus
Belarus
Radziukevich Maryna Lvovna, M. Sci. (Eng.), Head of the Testing Laboratory for Information Security Requirements
References
1. Radziukevich M. L., Golikov V. F. Combined formаtion of a cryptographic key using synchronized artificial neural networks. Doklady BGUIR. 2021; 19(1): 79–87.
2. Golikov V. F., Radziukevich M. L. [The formation of a common secret using artificial neural networks]. Sistemnyy analiz i prikladnaya informatika [System Analysis and Applied Informatics], 2019, no. 2, рр. 49–56 (in Russian).
3. Pivovarov V. L., Holikau U. F. [Method of generating common cryptographic keys for loosly coincident binary sequences]. Informatics. 2016;(3):31–37. (In Russ.).
4. Kinzel, W. Neural Cryptography / W. Kinzel, / I. Kanter // 9th International Conference on Neural Information Processing, Singapore, 2002.
5. Kanter, I. Secure exchange of information by synchronization of neural networks / I. Kanter, W. Kinzel, E. Kanter//arxiv: cond/0202112v1, [cond-mat.stat-mech], 2002.
6. Kanter, I. The Theory of Neural Networks and Cryptography, Quantum Computers and Computing / I. Kanter, W. Kinzel. –2005. Vol. 5, n.1. – P. 130–140.
7. Ruttor, A. Dynamics of neural cryptography / A. Ruttor, I. Kanter, and W. Kinzel // Phys. Rev. E, 75(5):056104, 2007.
8. Plonkovski, M. Cryptographic transformation of information based on neural network technology / M. Plonkovski, P. P. Urbanovich // Proceedings of BSTU. Series VI. Physics and Mathematics and Informatics; by ed. I. M. Zharsky. – Minsk: BSTU, 2005.
9. Radziukevich M. L., Golikov V. F. [Enhancing the secrecy of a cryptographic key generated using synchronized artificial neural networks]. Informatics. 2020;17(1):102–108. (In Russ.).
10. Golikov V. F., Ksenevich A. Y. [Attack on synchronized artificial neural networks, forming a common secret by deferred search]. Doklady BGUIR. 2017;(8):48–53. (In Russ.).
Review
For citations:
Radziukevich M.L. A combined method of formation of a cryptographic key with secret modification of the results of synchronization of artificial neural networks. «System analysis and applied information science». 2021;(3):51-58. (In Russ.) https://doi.org/10.21122/2309-4923-2021-3-51-58
Views:
550
Email this article 