Investigating Network Traffic and Selecting a Matching Mathematical Model

Authors: Chernigovskiy A.V., Krivov M.V., Istomin A.L. Published: 10.09.2020
Published in issue: #3(132)/2020  
DOI: 10.18698/0236-3933-2020-3-84-99

Category: Informatics, Computer Engineering and Control | Chapter: Mathematical Support and Software for Computers, Computer Complexes and Networks  
Keywords: network traffic, self-similarity, Hurst parameter, fractal Brownian motion, ARMA model, ARIMA model

The investigation aimed to study various network traffic types so as to derive a mathematical description not only for a specific type of traffic, but also for the aggregated network traffic. We characterized the main types of data transmitted during network operation and compared the results with the most common mathematical models, that is, Poisson, Pareto, Weibull, exponential and lognormal distributions. We established that regardless of traffic type the volume distribution of data packets transmitted has a "long tail" and is well described by the lognormal distribution model. We evaluated the autocorrelation function, which showed that a long-range dependence characterises virtually all data, which indicates their self-similarity. We also confirmed this conclusion by calculating the Hurst exponent. At the same time, we determined that the degree of self-similarity depends not only on the type of data transmitted, but also on the data ratio in the aggregated network traffic. We selected the following models so as to compare the mathematical descriptions of traffic: classical and fractal Brownian motion, and the AR, MA, ARMA and ARIMA models. The results showed that the fractal Brownian motion model provides the most accurate mathematical description of network traffic


[1] Chernigovskiy A.V., Krivov M.V. Methods of network traffic distribution between users. Bulletin of the Angarsk State Technical University, 2018, vol. 1, iss. 12, pp. 168--173 (in Russ.). DOI: https://doi.org/10.36629/2686-777X-2018-1-12-168-173

[2] Chen Th.M. Network traffic modeling. In: The Handbook of Computer Networks. Vol. 3. Wiley, 2007, pp. 326--339.

[3] Ihler A., Hutchins J., Smyth P. Learning to detect events with Markov-modulated Poisson processes. ACM Trans. Knowl. Discov. Data, 2007, vol. 1, no. 3, art. 13. DOI: https://doi.org/10.1145/1297332.1297337

[4] Paxson V., Floyd S. Wide area traffic: the failure of Poisson modeling. IEEE/ACM Trans. Netw., 1995, vol. 3, iss. 3, pp. 226--244. DOI: https://doi.org/10.1109/90.392383

[5] Sidorova O.I. Matematicheskie modeli trafika v sovremennykh telekommunikatsionnykh sistemakh. Avtoref. dis. kand. fiz.-mat. nauk [Mathematical traffic models in modern telecommunication systems. Abs. Cand. Sc. Phys.-Math. Diss.]. Tver, TSU Publ., 2009.

[6] Sidorova O.I. ON/OFF-model’ trafika s neodnorodnymi istochnikami v rezhime "bystrogo rosta chisla soedineniy" [ON/OFF model with heterogenous sources under fast growth condition]. Vestnik TvGU. Seriya: Prikladnaya matematika [Herald of Tver State University. Series: Applied Mathematics], 2015, no. 2, pp. 75--94 (in Russ.).

[7] Heffes H., Lucantoni D. A Markov modulated characterization of packetized voice and data traffic and related statistical multiplexer performance. IEEE J. Sel. Areas Commun., 1986, vol. 4, iss. 6, pp. 856--868. DOI: https://doi.org/10.1109/JSAC.1986.1146393

[8] Singh L.N., Dattatreya G.R. A novel approach to parameter estimation in Markov-modulated Poisson processes. Proc. IEEE (ETC), 2004, pp. 1--6.

[9] Leland W.E., Taqqu M.S., Wilinger W., et al. On the self-similar nature of Ethernet traffic (extended version). IEEE/ACM Trans. Netw., 1994, vol. 2, iss. 1, pp. 1--15. DOI: https://doi.org/10.1109/90.282603

[10] Adas A. Traffic models in broadband networks. IEEE Commun. Mag., 1997, vol. 35, iss. 7, pp. 82--89. DOI: https://doi.org/10.1109/35.601746

[11] Petrov V.V. To, chto vy khoteli znat’ o samopodobnom teletrafike, no stesnyalis’ sprosit’ [What you wanted to know about self-similar network traffic, but were embarrassed to ask]. pi.314159.ru: website (in Russ.). Available at: http://pi.314159.ru/petroff1.pdf (accessed: 03.01.2020).

[12] Edemskaya E.N., Belkov D.V. Network traffic research by the Hurst function. Informatika i kibernetika [Informatics and Cybernetics], 2015, no. 2, pp. 39--46 (in Russ.).

[13] Fedorova M.L., Ledeneva T.M. About research of property of self-similarity of the traffic of a multiservice network. Vestnik VGU. Seriya: Sistemnyy analiz i informatsionnye tekhnologii [Proceedings of Voronezh State University. Series: Systems Analysis and Information Technologies], 2010, no. 1, pp. 46--54 (in Russ.).

[14] Petrov V.V. Statisticheskiy analiz setevogo trafika [Statistical analysis of network traffic]. pi.314159.ru: website (in Russ.). Available at: http://pi.314159.ru/petroff2.pdf (accessed: 03.01.2020).

[15] Vorobyev O.V., Mozhaev A.A., Oskolkov A.P. Analysis of traffic description models. Nauka і tekhnіka Povіtryanikh Sil Zbroynikh Sil Ukraїni [Science and Technology of the Air Force of Ukraine], 2014, no. 2 (15), pp. 170--172 (in Russ.).