M.Y. Konyshev1, Y.A. Lezhnina2, V.A. Ivanov3, I.V. Ivanov4, A.V. Markin5

1-4 RTU MIREA (Moscow, Russia) 5 FSUE NTC ORION (Moscow, Russia)

1 konyshev@mirea.ru, 2 lezhnina@mirea.ru, 3 iva.mac@mail.ru, 4 iva.mac@mail.ru, 5 markin.a.v@fgupntcorion.ru

Statement of the problem. Digital str eams circulating in communication networks are non -stationary random processes with respect to the distribution of binary vectors. Monitoring of distributions allows you to control the network state by location, time and mode of operation, however, the quality of such control and the solution of traffic optimization tasks currently does not allow to meet the requirements for quality of service. An algorithm for traffic classification is proposed, which allows to recognize non -stationary traffic in areas of local stationarity for subsequent solution of traffic optimization tasks. Purpose. To improve the quality of flow classification in communication networks during traffic optimization.

Results. An approach to estimating the distribution of binary multidimen sional random variables in a data stream is presented. The features that determine the current traffic portrait in a digital stream segment are identified. The size of the segment is justified, and complex features of the distribution of binary random variables are introduced. The scientific novelty of the work consists in the development of a new evaluation indicator – a modified Minkowski metric, the determination of the optimal connectivity of a Markov chain when approximating the flow for traffic portra itization and the actualization of reference methods of pattern recognition based on committee constructions for its classification.

Practical significance. Consists in the possibility of using the developed algorithm for the classification of digital flo ws for compiling traffic profiles and optimizing traffic under network congestion.

Konyshev M.Y., Lezhnina Y.A., Ivanov V.A., Ivanov I.V., Markin A.V. Classification of digital streams in communication networks based on the dis tribution of multidimensional binary vectors for solving traff ic optimization problems. Dynamics of complex systems. 2026. V. 20. № 2. P. 14−23. DOI: 10.18127/j19997493-202602-02 (in Russian).

1. Kochergin S.V., Artemova S.V., Bakaev A.A. i dr. Povy`sheniebezopasnosti smart-setej: spektral`ny`jifraktal`ny`janaliz kak instrumenty` vy`yavleniya kiberatak. Russian Technological Journal. 2025. T. 13. № 1. S. 7–15.
2. Ivanov V.A., Markin A.V. i dr. Opredelenie stacionarnosti vcifrovy`xpotokaxotnositel`no raspredeleniya veroyatnostej dvoichny`xvektorov. Materialy` idoklady` XIII Vseros. mezhvedomstvennoj nauch. konf. «Aktual`ny`enapravleniya razvitiya sistem obespecheniya bezopasnosti ob``ektov gosudarstvennoj oxrany` ioxranyaemy`x ob``ektov, special`noj svyazidlya nuzhdorganov gosudarstvennoj vlasti ispecial`nogo informacionnogo obespecheniya gosudarstvenny`xorganov». Ch.
3. «Sistemy` special`noj svyazi». Orel: Akademiya FSO Rossii. 2023. S. 77–80.
4. Ivanov V.A., Markin A.V., Akimov E`.M. Programma dlya E`VM. Identifikator uchastkov lokal`noj stacionarnosti v mul`tipleksirovannom cifrovom potoke «EMU_Ident_Loc_Station». Svidetel`stvo ogosudarstvennoj registracii programmy` dlya E`VM № 2023619446 ot 11.05.2023. M.: FGU FIPS.
5. Gluskin V.A., Dement`ev A.N., Gondarenko E.A. i dr. Ocenivanie parametrov istochnikov oshibokvdiskretny`xkanalaxsvyazi sgruppirovaniem oshibok. Dinamika slozhny`xsistem – XXI vek. 2023. T. 17. № 4. S. 56–69.
6. Blejxut R. Teoriya ipraktika kodov, kontroliruyushhix oshibki. M.: Mir. 1986. 576 s.
7. Ivanov V.A., Smirnov D.S. i dr. Metrika dlyasravneniya ryadov raspredeleniya na mnozhestvax dvoichny`xvektorov. Materialy` XI Vseros. mezhvedomstvennoj konf. «Aktual`ny`enapravleniya razvitiya sistemoxrany`, special`noj svyaziiinformacii dlyanuzh dorganov gosudarstvennoj vlasti RF». Ch.
8. Orel: Akademiya FSO Rossii. 2019.
9. Gorelik A.L., Gurevich I.B., Skripkin V.A. Sovremennoe sostoyanie problemy` raspoznavaniya. M.: Radio isvyaz`. 1985. 160 s.
10. Fu K. Strukturny`emetody` vraspoznavanii obrazov. M.: Mir. 1977. 319 s.
11. Sebestian G.S. Processy` prinyatiya reshenij pri raspoznavanii obrazov. Per. sangl. Kiev: Texnika. 1965. 50 s. 1
12. Chernavin F.P. Primenenie metodakomitetov dlyaresheniya zadach klassifikacii. XII Mezhdunar. konf. «Rossijskie regiony` vfokuseperemen» (Ekaterinburg, 16–18 noyabrya 2017 g.). Ekaterinburg: UPI. 2 0 1
13. Ch. 1. S. 437–447. 11. http://www.rts-tender/ru (dataobrashheniya 29.11.2024).