Article 6418

Title of the article

ESTIMATION OF RESISTANCE OF PROTECTED NEURONET
BIOMETRCS-ACCESS CODE CONVERTERS USING LARGE BASES OF SYNTETIC BIOMETRIC IMAGES 

Authors

Mayorov Aleksandr Viktorovich, Specialist, Penza Research Institute of Electrical Engineering (9 Sovetskaya street, Penza, Russia), E-mail: pniei@penza.ru
Somkin Sergey Aleksandrovich, Specialist, Penza Research Institute of Electrical Engineering (9 Sovetskaya street, Penza, Russia), E-mail: pniei@penza.ru
Yunin Aleksey Petrovich, Specialist, Penza Research Institute of Electrical Engineering (9 Sovetskaya street, Penza, Russia), E-mail: pniei@penza.ru
Akmaev Artur Zhigansheevich, Specialist, Penza Research Institute of Electrical Engineering (9 Sovetskaya street, Penza, Russia), E-mail: pniei@penza.ru 

Index UDK

519.2, 004.622, 004.056.531 

DOI

10.21685/2072-3059-2018-4-6 

Abstract

Background. The aim of the work is to test the resistance to attacks of a selection of neural network containers where the neuron link tables and the tables of their weights are protected by gamming.
Materials and methods. The base of “Foe” natural biometric images is used. Next, it is checked how many first bits of the key are picked on the basis of natural biometric images. After that, the size of the base is increased by crossing the parent images and obtaining the descendant images by the algorithm of GOST R 52633.2. The size of the monotonously increasing test base and the number of key bits picked on it are controlled.
Results. In a logarithmic scale, the size of the test base is linearly related to the key length selected in a numerical experiment. Using this property, it is possible to predict the time taken to pick a key of 256 or 512 bits in length according to the real picking of a key of 40 to 64 bits in length.
Conclusions. Protection of neuron link tables by gamming allows for the implementation of an error propagation mechanism. One error of the “Foe” image on one neuron is enough for all subsequent neuron connection tables to recover incorrectly. There occurs the error propagation effect. 

Key words

neural network biometrics-code converter, gamming of neuron connection tables, reproduction of biometric images 

 Download PDF
References

1. GOST R 52633.0–2006. Zashchita informatsii. Tekhnika zashchity informatsii. Trebovaniya k sredstvam vysokonadezhnoy biometricheskoy autentifikatsii [Information protection. Information protection technique. Requirements to the means of highreliability biometric authentication]. Moscow: Standartinform, 2006, 19 p.
2. Yazov Yu. K., Volchikhin V. I., Ivanov A. I., Funtikov V. A., Nazarov I. G. Neyrosetevaya zashchita personal'nykh biometricheskikh dannykh [Neural network protection of personal biometric data]. Moscow: Radiotekhnika, 2012, 157 p.
3. Juels A., Wattenberg M. Proc. ACM Conf. Computer and Communications Security. Singapore, 1999, pp. 28–36.
4. Ramírez-Ruiz J., Pfeiffer C., Nolazco-Flores J. Advances in Artificial Intelligence – IBERAMIA-SBIA 2006 (LNCS 4140). Ribeirão Preto, Brazil, 2006, pp. 178–187.
5. Hao F., Anderson R., Daugman J. IEEE Transactions On Computers. 2006, vol. 55, no. 9.
6. GOST R 52633.5–2011. Zashchita informatsii. Tekhnika zashchity informatsii. Avtomaticheskoe obuchenie neyrosetevykh preobrazovateley biometriya-kod dostupa [Information protection. Information protection technique. Automatic training of neural network biometrics-access code converters]. Moscow: Standartinform, 2011, 20 p.
7. Tekhnicheskaya spetsifikatsiya (proekt, publichnoe obsuzhdenie nachato s 01.02.2017 chlenami TK 26 «Kriptograficheskaya zashchita informatsii») ZAShchITA NEYROSETEVYKh BIOMETRIChESKIKh KONTEYNEROV S ISPOL''ZOVANIEM KRIPTOGRAFIChESKIKh ALGORITMOV [Technical specification (draft, public discussion initiated on 01.02.2017 by members of TC 26 “Cryptographic protection of information”). Protection of neural network biometric containers using cryptographic algorithms]. Moscow, 2017.
8. Pat. 2346397 Russian Federation. Sposob zashchity personal'nykh dannykh biometricheskoy identifikatsii i autentifikatsii [A method of protection of personal data on biometric identification and authentication]. Ivanov A. I., Funtikov V. A., Efimov O. V. appl. 26.06.2007, publ. 10.02.2009, bull. no. 4.
9. GOST P 34.12–2015. Informatsionnaya tekhnologiya. Kriptograficheskaya zashchita informatsii. Rezhimy raboty blochnykh shifrov [Information technology. Cryptographic protection of information. Operating modes of block cyphers]. Moscow: Standartinform, 2015, 21 p.
10. GOST R 34.11–2012. Informatsionnaya tekhnologiya. Kriptograficheskaya zashchita informatsii. Funktsiya kheshirovaniya» [Information technology. Cryptographic protection of information. Hashing function]. Moscow: Standartinform, 2012, 24 p.
11. GOST R 52633.3–2011. Zashchita informatsii. Tekhnika zashchity informatsii. Testirovanie stoykosti sredstv vysokonadezhnoy biometricheskoy zashchity k atakam podbora [Information protection. Information protection technique. Testing of resistance of the means of high-reliability biometric protection to picking attacks]. Moscow: Standartinform, 2011, 23 p.
12. GOST R 52633.1–2009. Zashchita informatsii. Tekhnika zashchity informatsii. Trebovaniya k formirovaniyu baz estestvennykh biometricheskikh obrazov, prednaznachennykh dlya testirovaniya sredstv vysokonadezhnoy biometricheskoy autentifikatsii [Information protection. Information protection technique. Requirements to the formation of bases of natural biometric images intended for testing of the means of highreliability biometric authentication]. Moscow: Standartinform, 2009, 25 p.
13. GOST R 52633.2–2010. Zashchita informatsii. Tekhnika zashchity informatsii. Trebovaniya k formirovaniyu sinteticheskikh biometricheskikh obrazov, prednaznachennykh dlya testirovaniya sredstv vysokonadezhnoy biometricheskoy autentifikatsii. [Information protection. Information protection technique. Requirements to the formation of synthetic biometric images intended for testing of the means of high-reliability biometric authentication]. Moscow: Standartinform, 2011, 22 p.

 

Дата создания: 23.04.2019 14:46
Дата обновления: 23.04.2019 15:20