RADIOELECTRONIC AND COMPUTER SYSTEMS

The processes of design, development, testing, maintenance, and management of modern computer systems (CS) require the solution of optimization tasks. The peculiarity of such processes is to obtain many characteristics in real time. At the same time, one of the priority tasks is the task of selecting these characteristics or indicators. The subject of the article study is method for evaluating the informativeness of the indicators of the functioning of computer systems. The purpose of the article is to develop a methodology for selecting a system of indicators for identifying the state of the CS under the condition of fuzzy output data.  To automate the process of choosing the most influential indicators, it is necessary to involve additional information. Information is needed on the impact of individual indicators and effective mathematical decisions on false data cessation. This information can increase the reliability of the results of the identification of the state of the CS. This determines the relevance of the task of developing a methodology for selecting a system of indicators for identifying the state of the CS. Results. The analysis showed that the identification problem falls into two subtasks. The first is the choice of a set of monitored indicators of an object. The second is the solution for the identification task accordingly. The solution to the problem of choosing a subset of informative parameters from a set of parameters controlled by using the "distances" of Kulbak is considered. This technology requires the following assumption. The densities of system parameters that are monitored are known. They can be statistically evaluated for each state of the system. In real conditions of a small sample of output data, this hypothesis cannot be properly justified. In these circumstances, it is natural to use other possibilities to describe the uncertainty of the output data. At the same time, it is not desirable to use the mathematical apparatus of probability theory. Proceeding from this, the requirements for the informative criterion are formed. It is shown that these requirements are met by a mathematical apparatus of fuzzy mathematics. On the basis of this apparatus, an appropriate method is formed. Conclusions. The method of selection of the system of indicators for the identification of the state of the computer system is developed. This allowed us to expand the range of incoming processed metrics in the state identification system on the condition of fuzzy input data and improve the accuracy of identifying the state of the CS.

computer system; state identification; estimation of informality; "distance" of Kulbak

Semenov, S. G., Davydov, V. V., Gavrilenko, S. Yu. Zashhita dannykh v komp`yuterizirovannykh upravlyayushhikh sistemakh [Data protection in computerized control systems]. Germany, LAP LAMBERT ACADEMIC PUBLISHING Publ., 2014. 236 p.

Shelukhin, O. I., Sakalema, D. Zh., Filinova, A. S. Obnaruzhenie vtorzhenii v komp'yuternye seti [Detection of intrusions into computer networks]. Moscow, Goryachaya liniya-Telekom Publ., 2013. 220 p.

Agrawal, S. Survey on Anomaly Detection using Data Mining Techniques. Procedia Computer Science, 2015, vol. 60, pp. 708-713.

Borovikov, V. P. Iskusstvo analiza dannykh [Art of data analysis]. St. Petersburg, Piter Publ., 2005. 432 p.

Tsypkin, Ya. Z. Informatsionnaya teoriya identifikatsii [Information theory of identification]. Moscow, Nauka, Fizmatlit Publ., 1995. 336 p.

Chandola, V., Banerjee, A., Kumar, V. Anomaly detection for discrete sequences: A survey. IEEE Transactions on Knowledge and Data Engineering, 2012, vol. 24, no. 5, pp. 823-839.

Diligenskaya, A. N. Identifikatsiya ob"ektov upravleniya [Identification of control objects]. Samara, Samar. gos. tekhn. un-t Publ., 2009. 136 p.

Rusakov, K. D., Khil', S. Sh. O zadache vybora priznakov nablyudaemogo sostoyaniya slozhnogo dinamicheskogo ob"ekta v usloviyakh razlichnogo kachestva izmeritel'noi informatsii [On the problem of choosing the signs of the observed state of a complex dynamic object under conditions of different quality measurement information]. Neirokomp'yutery i ikh primenenie: 15-ya Vserossiiskaya nauchn. konf. [Neurocomputers and their applications: the 15th All-Russian Scientific. konf.]. Moscow, FGBOU VO MGPPU, 2017, pp. 246-248.

Kullback, S. Information Theory and Statistics. New York, Wiley Publ., 1959. 395 p.

Gavrylenko, S. Yu. Syntez identyfikatsiinykh vymiriv v kompiuternii systemi krytychnoho pryznachennia [Synthesis of identification measurements in the computer system of critical use]. Suchasnyi stan naukovykh doslidzhen ta tekhnolohii v promyslovosti [Current state of scientific researches and technologies in industry], 2019, no. 2 (8), pp. 36-43. DOI: 10.30837/2522-9818.2019.8.036.

Semenov, S., Sira, O., Gavrylenko, S., Kuchuk, N. Identification of the state of an object under conditions of fuzzy input data. Eastern-European Journal of Enterprise Technologies, 2019, vol. 1, no 4 (97), pp. 22-29. DOI: 10.15587/1729-4061.2019.15708.