Article 3416

Title of the article

THE FRACTAL-CORRELATION FUNCTIONAL USED WHEN SEARCHING FOR PAIRS OF WEAKLY
DEPENDENT BIOMETRIC DATA IN SMALL SAMPLES 

Authors

Volchikhin Vladimir Ivanovich, Doctor of engineering sciences, professor, President of Penza State University (40 Krasnaya street, Penza, Russia), president@pnzgu.ru
Akhmetov Berik Bakhytzhanovich, Doctor of engineering sciences, professor, vice-president, Hodja Ahmet Yassawi International Kazakh-Turkish University (29 B. Sattarkhanov, Turkestan, Kazakhstan), berik.akhmetov@ayu.edu.kz
Ivanov Aleksandr Ivanovich, Doctor of engineering sciences, associate professor, head of the laboratory of biometric and neural network technologies, Penza Research Institute of Electrical Engineering (9 Sovetskaya street, Penza, Russia), ivan@pniei.penza.ru
Serikova Yuliya Igorevna, Student, Penza State University (40 Krasnaya street, Penza, Russia), ivan@pniei.penza.ru

Index UDK

519.7; 519.66; 57.087.1, 612.087.1

DOI

10.21685/2072-3059–2016-4-3

Abstract

Background. The aim of the work is to partially compensate random methodological errors in calculation of almost zero values of correlation coefficients. The cause of the random component of methodological errors is insufficient biometric data. The article solves the problem of increasing the accuracy of calculations (regularization) for pair correlation coefficients calculated by the classical formula in the range of –0.1 to +0.1.
Materials and methods. The authors have proposed to use a variant of the fractal correlation functional, built on the determination of the ratio of the length of step lines derived by ordering the data for each of the studied variables. The used correlation functional is an analog of measurement of coastlines’ length on maps of different scales, described in the classical sources on fractal calculations. The article presents a table of scaling factors of the new computational formula for weakly correlated data at different volumes of the test sample.
Results and conclusions. It is shown that the proposed fractal-correlation functional in combination with the classical correlation functional on a sample of 16 examples reduces the random component of methodological errors by about 20 %, which is equivalent to an increase of the sample size to 24 experiments. The work describes the table of weight coefficients of the linear combination of a pair of proposed and classical correlation functionals.

Key words

biometric identification, simple correlation coefficients of pair correlation, functionals of pair correlation, test sample volumes, fractal calculations

Download PDF
References

1. Dodis Y., Reyzin L., Smith A. Proc. EUROCRYPT. 2004, April 13, pp. 523–540. 2. Ramírez-Ruiz J., Pfeiffer C., Nolazco-Flores J. Advances in Artificial Intelligence – IBERAMIA-SBIA 2006 (LNCS 4140). 2006, pp. 178–187.
3. Hao F., Anderson R., Daugman J. IEEE TRANSACTIONS ON COMPUTERS. 2006, vol. 55, no. 9, pp. 1073–1074.
4. 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.
5. Akhmetov B. S., Ivanov A. I., Funtikov V. A., Bezyaev A. V., Malygina E. A. Tekhnologiya ispol'zovaniya bol'shikh neyronnykh setey dlya preobrazovaniya nechetkikh biometricheskikh dannykh v kod klyucha dostupa: monogr. [A technology of large neural network application for fuzzy biometric data transformation into access key  code: monograph]. Kazakhstan, Almaty: Izd-vo LEM, 2014, 144 p. Available at:  http://portal.kazntu.kz/files/publicate/2014-06-27-11940.pdf
6. GOST R 52633.5–2011. Zashchita informatsii. Tekhnika zashchity informatsii. Avtomaticheskoe obuchenie neyrosetevykh preobrazovateley biometriya-kod dostupa [Information protection. Data protection technology. Automatic learning of “biometricsaccess code” converters]. Moscow, 2011.
7. Kachaykin E. I., Ivanov A. I. Voprosy kiberbezopasnosti [Cybersafety problems]. 2015, no. 4 (12), pp. 42–47.
8. Ivanov A. I., Lozhnikov P. S., Kachaykin E. I. Voprosy zashchity informatsii [Data protection problems]. 2015, no. 2, pp. 48–54.
9. Ivanov A. I., Lozhnikov P. S., Kachaykin E. I., Sulavko A. E. Voprosy zashchity informatsii [Data protection problems]. 2015, no. 3, p. 48–54.
10. Akhmetov B. S., Volchikhin V. I., Ivanov A. I., Malygin A. Yu. Algoritmy testirovaniya biometriko-neyrosetevykh mekhanizmov zashchity informatsii [Testing algorithms for biometric neural network mechanisms of data protection]. Kazakhstan, Almaty: KazNTU im. Satpaeva, 2013, 152 p. Available at: http://portal.kazntu.kz/ files/publicate/2014-01-04-11940.pdf
11. Peters E. Khaos i poryadok na rynkakh kapitala. Novyy analiticheskiy vzglyad na tsikly, tseny i izmenchivost' rynka [Chaos and order in capital markets. New analytical views of cycles, prices and market changeability]. Moscow: Mir, 2000, 333 p.
12. Gil'mutdinov A. Kh. Fraktaly i drobnye operatory [Fractals and fractional operators]. Kazan: Izd-vo Fen Akademii nauk RT, 2010, 488 p.
13. Shreder M. Fraktaly, khaos, stepennye zakony. Miniatyury iz beskonechnogo raya [Fractals, chaos, power laws. Miniatures from endless heaven]. Izhevsk: Regulyarnaya i khaoticheskaya dinamika, 2005, 528 p.

 

Дата создания: 02.08.2017 14:45
Дата обновления: 03.08.2017 11:50