NEURAL NETWORK CONVERSION OF BIOMETRY INTO AUTHENTICATION
CODE: ADDITION OF HAMMING ENTROPY WITH ENTROPY OF CORRELATION RELATIONS BETWEEN DISCHARGES 

Aleksandr I. Ivanov, Doctor of technical sciences, associate professor, leading researcher, Penza Research Electrotechnical Institute (9 Sovetskaya street, Penza, Russia), ivan@pniei.penza.ru 
Aleksey P. Ivanov, Candidate of technical sciences, associate professor, head of the sub-department of technical means of information security, Penza State University (40 Krasnaya street, Penza, Russia), ap_ivanov@pnzgu.ru
Kirill A. Gorbunov, Postgraduate student, Penza State University (40 Krasnaya street, Penza, Russia), kirill.gobunov@gmail.com

Background. The problem of calculating the entropy of 256-bit codes with dependent bits on small test samples consisting of 20 examples is considered. Materials and methods. It is proposed to estimate the entropy of the output codes of the neural network transformer by calculating the mutual correlation coefficients of the code sequences 256 bits long, obtained for examples of one “Alien” image. Results. It is shown that the proposed method is much more accurate than the previously used estimation method by calculating the mathematical expectation and standard deviation of Hamming distances for the same “Alien” image. Conclusions. The results obtained make it possible to raise the question of adjusting the national standard GOST R 52633.3 in the near future through the introduction of an additional section into it concerning the calculation of the correlation entropy.

small sample testing, artificial neurons, biometrics-to-code conversion

Ivanov A.I., Ivanov A.P., Gorbunov K.A. Neural network conversion of biometry into authentication code: addition of hamming entropy with entropy of correlation relations between discharges. Nadezhnost' i kachestvo slozhnykh sistem = Reliability and quality of complex systems. 2023;(1):91–98. (In Russ.). doi:10.21685/2307-4205-2023-1-11