Using of regression analysis methods for monitoring and managing telecommunication systems

Ruslan Abduramanov, Yurii Khlaponin, Anton Khaddad


Telecommunication systems are a crucial part of modern human life. Absolutely all aspects of the activities of modern people have become dependent on the effective operation of telecommunications networks. The development of telecommunications systems and computer networks necessitates the creation and reliable operation of a large set of info communication services that ensure the effective operation of the user with heterogeneous information in the telecommunications network. The historically formed heterogeneity of both telecommunication systems, computer networks, network information resources, and the audience of users to whom this information is addressed complicates the objective analysis and monitoring of telecommunication architectures and resources. Therefore, it is certainly true that when operating telecommunication systems and computer networks, a fairly wide range of modern and scientifically sound technical and technological solutions for their analysis and monitoring should be used. Due to this fact, solving problems of the monitoring and managing telecommunications systems is of utmost importance. Recently, intelligent decision-making systems, based on data processing systems with the use of machine learning technologies, have become popular. In this paper, one such technology based on logistic regression is considered. Using real data of telecommunication network functioning, the model describing network functioning has been constructed. In particular, the possibility of using logistic regression to predict the probability of inefficient operation of networks based on the processes that occur in them has been proved. Based on the model, an intelligent decision-making system has been built, which is used to monitor and control the state of the network in real time.


Telecommunication system; intellectual decision-making system; logistic regression; computer network; SNMP protocol.


Y.I. Khlaponin, H.B. Zhyrov, and O.M. Nikitchin, “Application of Neural Networks in the Statistical System for Analysis and Monitoring of Telecommunication Networks”, Technological Audit and Production Reserves, vol. 5, no. 2, pp. 35-41, 2016.

I.S. Eniukov, I.V. Retinskaia, and A.K. Skuratov, Statistical analysis and monitoring of scientific and educational Internet networks. Moscow, Russia: Finance and Statistics, 2004.

Y.P. Nedaibida, Y.V. Kotova, and Y.I. Khlaponin, “Modern problems of creating complex real-time information management systems”, Ukrainian Information Security Research Journal, vol. 14, no. 4 (57), pp. 50-55, 2012.

doi: 10.18372/2410-7840.14.3486.

S.A. Aivazian, V.M. Bukhtshtaber, I.S. Eniukov, and L.D. Meshalkin, Applied statistics: classification and reduction of dimension. Moscow, Russia: Finance and Statistics, 1989.

D. Hosmer, and S. Lemeshow, Applied Logistic Regression, 2nd ed. New York, USA:

Wiley, 2002.

doi: 10.1002/0471722146.

R. Koenker, and G. Bassett, “Regression Quantiles”, Econometrica. vol. 46, no.1,

pp. 33-50, 1978.

doi: 10.2307/1913643.

R. Koenker, K.F. Hallock, “Quantile Regression” , Journal of Economic Perspectives, vol. 15, no. 4, pp. 143-156, 2001.

doi: 10.1257/jep.15.4.143.

ISSN 2411-1031 (Print), ISSN 2518-1033 (Online)