Software simulation model of computer network with multifractal traffic simulation based on Markov chain
Abstract
based on the Markov chain for testing routing algorithms is developed. A method based on the theory of complex networks has been developed to generate the structure of a computer network. To simulate network traffic, a method of generating a multifractal binary sequence using the Markov chain has been developed. A computer network in a developed model is represented by a fully connected undirected weighted graph, in which nodes are routers, and edges are network connections between them. The weight of the edges is the inverse of the bandwidth of the communication channel. Nodes contain queues in which received packets are placed before determining the route of their dispatch and sending them to the next node. Time in the model is represented by discrete iterations. Routing is carried out on the basis of those algorithms, which must be tested on the model. To simulate network traffic, in the developed software simulation model, a method of generating a binary multifractal sequence based on Markov chains with a stochastic automaton, which makes possible to control the fractal dimension of the binary series on different scales, is proposed. As a result of a numerical experiment, the fact of the possibility of adjusting the Hurst index on a given time scale has been established. It is shown that the obtained time series using the cascading generator of the binary sequence have multifractal properties. That is, the cascading generator has more possibilities for adaptation to real examples of binary traffic. The scientific novelty of the conducted research is as follows: 1. The method of network traffic generation based on the Markov chain has been improved, which differs from the known ones in that it uses a cascade model of the generator of a binary numerical sequence at the "packet present" - "packet absent" level and allows to generate a traffic with multifractal properties with the possibility of their adjustment. 2. A method of software simulation modeling of a computer network based on the theory of complex networks and an improved method of generating network multifractal traffic, which allows to test routing algorithms and protocols, has been developed
Keywords
computer networks; software simulation model; network traffic; fractal dimension; Hurst exponent; multifractality
References
[1] A.-L. Barabási, Network Science. Cambridge University Press, 2018. [Online]. Available: http://networksciencebook.com/
[2] V. V. Pasichnyk, and N. M. Ivanushchak, "Research and modeling of complex networks", Skhidno-Yevropeiskyi zhurnal peredovykh tekhnolohii, iss. 2, no. 3 (44), pp. 43-48, 2010 [in Ukrainian].
[3] V. А. Traag, Algorithms and Dynamical Models for Communities and Reputation in Social Networks. Springer International Publishing, 2014, p. 229. [Online]. Available: https://doi.org/10.1007/978-3-31906391-1
[4] D. J. Watts, and S. H. Strogatz, "Collective dynamics of "small-world" networks", Nature, vol. 393 (6684), pp. 440-442, 1998. [Online]. Available: https://www.nature.com/ articles/30918
[5] S. Robert, and J. Y. Le Boudec, "New models for pseudo self-similar traffic", Performance Evaluation, vol. 30 (1-2). pp. 57-68, 1997.
[6] G. Horn, A. Kvalbein, J. Blomskøld, and E. Nilsen, "An empirical comparison of generators for self-similar simulated traffic", Performance Evaluation, vol. 64 (2), pp. 162-190, 2007.
[7] T. Sobh, K. Elleithy, and A. Mahmood, Eds., Novel Algorithms and Techniques in Telecommunications and Networking. Springer, 2010, pp. 41-46.
[8] H. M. Drieieva, O. A. Smirnov, O. M. Drieiev, and T. V. Smirnova, "A fractal analysis of a Markov chain based self-similar traffic generator", Central Ukrainian Scientific Bulletin, Engineering sciences, vol. 2 (33), pp. 161-172, 2019.
[9] H. Drieieva, O. Drieiev, Ye. Meleshko, M. Yakymenko, and V. Mikhav, "A method of determining the fractal dimension of network traffic by its probabilistic properties and experimental research of the quality of this method", CEUR-WS, vol. 3171, pp. 1694-1707, Gliwice, Poland, 2022. [Online]. Available: http://ceur-ws.org/Vol-3171/paper120.pdf
[10] Ye. Meleshko, H. Drieieva, ta V. Mikhav, "Software simulation model of a computer network for testing traffic routing algorithms", in Int. Sci. and Tech. Conf. Automation, Computer-Integrated Technologies and Problems of Energy Efficiency in Industry and Agriculture (AKIT-2022). Kropyvnytskyi: EkskliuzyvSystem, 2022, pp. 26-27 [in Ukrainian].
[11] A.-L. Barabási, and R. Albert, "Emergence of scaling in random networks", Science, vol. 286, no. 5439, pp. 509-512, 1999. [Online]. Available: https://doi.org/10.1126/ science.286.5439.509
[12] G. Millán, and G. Lefranc, "A fast multifractal model for self-similar traffic flows in high-speed computer networks", Information Technology and Quantitative Management (ITQM2013). Procedia Computer Science, 17, pp. 420-425, 2013.
[13] E. Areström, and N. Carlsson, "Early online classification of encrypted traffic streams using multi-fractal features", in IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS), 2019, pp. 84-89. doi: 10.1109/INFCOMW.2019.8845127.
[14] V. J. Ribeiro, Z.-L Zhang, S. Moon, and C. Diot, "Small-time scaling behavior of Internet backbone traffic", Computer Networks, vol. 48, pp. 315-334, 2005. doi: 10.1016/j.comnet.2004.11.012.
[15] "Hurst exponent evaluation and R/S-analysis doi: 10.1109/INFCOMW.2019.8845127. in Python", GitHub - Mottl/hurst. [Online]. Available: https://github.com/Mottl/hurst