Usage of hypercomplex numbers in a cryptographic key agreement protocol based on neural networks

P. P. Urbanovich ; N. P. Shutko

Journal of the Belarusian State University. Mathematics and Informatics, Tome 2 (2024), pp. 81-92.

Résumé

We analyse the features of the structural and functional organisation of two interacting neural networks based on the known architecture in the form of a tree parity machine (TPM) using algebras of real and hypercomplex numbers. Such machines are used as an alternative to the Diffie – Hellman algorithm to generate a shared secret cryptographic key between two parties. The main elements of mathematical models of TPMs, operating on the basis of the listed algebras, are considered. The features of the software implementation of a system simulator based on interacting TРMs are described, and the results of using the developed tool for analysing the dynamics of processes in the system under consideration are presented. Mutual learning and data exchange of two TРMs are realised based on the transmission control and Internet protocols (TCP/IP). The synchronisation state of the networks is determined by the equality of the hashes that each party calculates based on the secure hash algorithm. A hash size of 512 bits are generated by transforming the string representation of the current input vector of neuron weights. The effectiveness of possible attempts by a third party to synchronise with two legitimate TPMs operating on the basis of algebras of hypercomplex numbers is assessed.

Détail

Keywords: neural cryptography; tree parity machines; hypercomplex numbers; networks mutual learning

Bibliographie

[1] W. Diffie, M. Hellman, “New directions in cryptography”, IEEE Transactions on Information Theory, 22(6) (1976), 644–654 | DOI

[2] W. Kinzel, I. Kanter, “Neural cryptography”, Proceedings of the 9th International conference on neural information processing (Singapore), Orchid Country Club, Singapore, 2002, 1351–1354 | DOI

[3] A. Klimov, A. Mityagin, A. Shamir, “Analysis of neural cryptography”, Advances in Cryptology – ASIACRYPT 2002, Springer, Berlin, 2002, 288–298 (Lecture notes in computer science; volume 2501) | DOI

[4] M. Rosen-Zvi, I. Kanter, W. Kinzel, “Cryptography based on neural networks analytical results”, Journal of Physics A. Mathematical and General, 35(47) (2002), 707–713 | DOI

[5] M. Plonkovski, P. P. Urbanovich, “Kriptograficheskoe preobrazovanie informatsii na osnove neirosetevykh tekhnologii”, Trudy Belorusskogo gosudarstvennogo tekhnologicheskogo universiteta. Seriya 6, Fiziko-matematicheskie nauki i informatika, 13 (2005), 161–164

[6] M. Plonkowski, R. Urbanowicz, “Liczby podwojne i ich modyfikacje w neurokryptografii”, Przegląd Elektrotechniczny, 88(11b) (2002), 340–341

[7] Y. Choi, J. Sim, L-S. Kim, “CREMON: cryptography embedded on the convolutional neural network accelerator”, IEEE Transactions on Circuits and Systems II. Express Briefs, 67(12) (2020), 3337–3341 | DOI

[8] S. Jeong, C. Park, D. Hong, C. Seo, N. Jho, “Neural cryptography based on generalized tree parity machine for real-life systems”, Security and Communication Networks, 11 (2021), 1–12 | DOI

[9] A. Sarkar, “Neural cryptography using optimal structure of neural networks”, Applied Intelligence, 51 (2021), 8057–8066 | DOI

[10] S. Dourlens, Neuro-cryptographie appliquée et neuro-cryptanalyse du DES, University of Paris, Paris, 1995, 218 pp. | DOI

[11] A. Ruttor, Neural synchronization and cryptography [dissertation], Julius-Maximilians-Universitat Wurzburg, Wurzburg, 2006, +120 pp. | DOI

[12] M. Plonkovski, P. P. Urbanovich, “Sinkhronizatsiya kriptograficheskikh klyuchei na osnove neironnykh setei i v sistemakh kriptopreobrazovaniya na osnove XML”, Trudy Belorusskogo gosudarstvennogo tekhnologicheskogo universiteta. Seriya 6, Fiziko-matematicheskie nauki i informatika, 14 (2006), 152–155

[13] A. Ruttor, W. Kinzel, I. Kanter, “Dynamics of neural cryptography”, Physical Review E, 75(5) (2007), 056104 | DOI

[14] M. Dolecki, R. Kozera, “Distribution of the tree parity machine synchronization time”, Advances in Science and Technology, 7(18) (2013), 20–27 | DOI

[15] P. P. Urbanovich, K. V. Churikov, “Sravnitelnyi analiz metodov vzaimoobucheniya neironnykh setei v zadachakh obmena konfidentsialnoi informatsiei”, Trudy BGTU. Fiziko-matematicheskie nauki i informatika, 6 (2010), 163–166

[16] L. F. Seoane, A. Ruttor, “Successful attack on permutation-parity-machine-based neural cryptography”, Physical Review E, 85(2) (2012), 025101 | DOI

[17] L. N. Shacham, E. Klein, R. Mislovaty, I. Kanter, W. Kinzel, “Cooperating attackers in neural cryptography”, Physical Review E, 69(6) (2004), 066137 | DOI

[18] I. L. Kantor, A. S. Solodovnikov, Giperkompleksnye chisla, Nauka, Moskva, 1973, +144 pp.

[19] M. Płonkowski, R. Urbanowicz, E. Lisica, “Wykorzystanie kwaternionow w protokole uzgadniania klucza kryptograficznego, opartym na architekturach sieci neuronowych TPQM”, Przegląd Elektrotechniczny, 86(7) (2010), 90–91

[20] T. Dong, T. Huang, “Neural cryptography based on complex-valued neural network”, IEEE Transactions on Neural Networks and Learning Systems, 31(11) (2020), 4999–5004 | DOI

[21] Y. Zhang, W. Wang, H. Zhang, “Neural cryptography based on quaternion-valued neural network”, International Journal of Innovative Computing, Information and Control, 6(22) (2022), 1871–1883

[22] J. Wu, L. Xu, F. Wu, Y. Kong, L. Senhadji, H. Shu, “Deep octonion networks”, Neurocomputing, 397 (2019), 179–191 | DOI

[23] K. Takahashi, M. Fujita, M. Hashimoto, “Remarks on octonion-valued neural networks with application to robot manipulator control”, 2021 IEEE International Conference on Mechatronics (ICM) (Kashiwa, Japan), IEEE, 2021, 1–6 | DOI

[24] A. Cariow, G. Cariowa, “Fast algorithms for deep octonion networks”, IEEE Transactions on Neural Networks and Learning Systems, 34(1) (2023), 543–548 | DOI

[25] C. Ricotta, J. Podani, “On some properties of the Bray – Curtis dissimilarity and their ecological meaning”, Ecological Complexity, 31 (2017), 201–205 | DOI