Personal cryptography using artificial neural networks

Безобразов, Сергей Валерьевич; Blackledge, J. M.

Search

View/Open document files

Научный доклад (1.031Mb)

Author

Безобразов, Сергей Валерьевич
Blackledge, J. M.

Date

2016

Publisher

Национальный исследовательский ядерный университет МИФИ

Show full item record

Citation

Bezobrazov, S. V. Personal cryptography using artificial neural networks / S. V. Bezobrazov, J. M. Blackledge // XVIII международная конференция "Нейроинформатика – 2016" : сборник научных трудов, Москва, 25–29 апреля 2016 г. / Российская академия наук [и др.] ; отв. ред. А. Г. Трофимов. – Москва : МИФИ, 2016. – Ч. 3. – Р. 152–161. – Bibliogr.: p. 161 (7 titles).

Abstract

This paper presents a method of generating encryption algorithms using neural networks. Based on the application of natural noise sources we ‘teach’ a system to approximate the input noise with the aim of generating an output nonlinear sequence. This output is then treated as an iterator which is subjected to a range of tests to check for potential cryptographic strength in terms of metric. This approach provides the potential for generating an unlimited number of unique Pseudo Random Number Generator (PRNG) that can be used on a ‘one-to-one’ basis.