INTELLIGENT COMPUTER NETWORK TRAFFIC FORECASTING SYSTEM BASED ON SYNPHASE DATA PROCESSING
DOI:
https://doi.org/10.35546/kntu2078-4481.2025.3.2.63Keywords:
computer networks, intelligent system, traffic forecasting, synphase processing, periodically correlated random process, MATLABAbstract
The article considers the problem of increasing the efficiency of computer network management by forecasting their traffic using intelligent data processing methods. The feasibility of using new generation stochastic models, in particular periodically correlated stochastic processes (PCSP), which allow simultaneously taking into account the random nature of traffic formation and its daily cyclicity, is substantiated. Unlike classical models (Poisson, Markov, ARIMA, fractal), which are limited in reproducing complex patterns, the PCSP-based approach provides a more adequate description of real processes in telecommunication systems.The architecture of an intelligent forecasting system is proposed, the core of which is synphase data processing algorithms. The implemented procedures include parametric covariance estimation, spectral analysis of centered signals using the Fourier transform, and averaging of correlation components to reduce the influence of noise components.This ensures high accuracy of reproducing network load variability and forming stable forecasts even under conditions of stochastic deviations.Experimental testing was conducted based on data from the Internet provider UFONet (Ternopil). It was found that the system correctly identifies critical peak intervals (1.8–2.1 TB in the evening) and periods of minimum load (0.5–0.7 TB at night). Additionally, an automatic load level classification module was implemented, which translates numerical forecasts into a categorical form (“minimum”, “average”, “critical”). Such a mechanism allows operators and providers to carry out proactive resource management, reduce the risks of overloads and plan maintenance in the least active time intervals.The practical significance of the results obtained lies in creating a decision support tool for network infrastructure administrators, which increases the stability of the system and the quality of service provision to users. The scientific novelty of the work is determined by the synthesis of PCSP methods and synphase analysis in a single intelligent forecasting system. Prospects for further research are related to the integration of the developed system with machine learning algorithms and the expansion of its application to multi-level computer networks of the next generations.
References
Erlang A. K. The theory of probabilities and telephone conversations. Nyt Tidsskrift for Matematik. 1909. Vol. 20. P. 33–39.
Czachórski T., Grochla K., Jozefiok A., Nycz T., Pekergin F. Performance evaluation of a multiuser interactive networking system: A comparison of modelling methods. Proc. 26th International Symposium on Computer and Information Sciences (ISCIS 2011). London, UK. 2011. P. 215–221.
Domańska J., Domański A., Czachórski T. Internet traffic source based on hidden Markov model. Lecture Notes in Computer Science. Vol. 6869. Springer. 2011. P. 395–404. DOI: 10.1007/978-3-642-22875-9_36
Claypool M. The effect of latency on user performance in real-time strategy games. Computer Networks. 1995. Vol. 49. P. 52–70. DOI: 10.1016/j.comnet.2005.07.003
Daigle J. N. Queueing Theory for Telecommunications. New York: Addison-Wesley. 1990. 456 p.
Box G. E. P., Jenkins G. M. Time Series Analysis: Forecasting and Control. San Francisco: Holden-Day. 1970. 553 p. DOI: 10.1002/9781118619190
Mandelbrot B. B., Van Ness J. W. Fractional Brownian motions, fractional noises and applications. SIAM Review. 1968. Vol. 10. P. 422–437. DOI: 10.1137/1010093
Leland W. E., Taqqu M. S., Willinger W., Wilson D. V. On the self-similar nature of Ethernet traffic (Extended version). /ACM Transactions on Networking. 1994. Vol. 2(1). P. 1–15. DOI: 10.1109/90.282603
Reichl P. A. A generalized TES model for periodical traffic. Proc. IEEE International Conference on Communications. 1998. P. 1–5. DOI: 10.1109/ICC.1998.683236
Chen T. Network traffic modeling. In: Bidgoli H. (ed.) Handbook of Computer Networks. Wiley. 2007. P. 326–339. DOI: 10.1002/9781118256107.ch21
Zhang Y., Roughan M., Duffield N., Greenberg A. Fast accurate computation of large-scale IP traffic matrices from link loads. ACM SIGMETRICS Performance Evaluation Review. 2003. Vol. 31(1). P. 206–217. DOI: 10.1145/885651.781041
Zhang G., Patuwo B.E., Hu M.Y. Forecasting with artificial neural networks: The state of the art. International Journal of Forecasting. 1998. Vol. 14(1). P. 35–62. DOI: 10.1016/S0169-2070(97)00044-7
Хвостівський М. О., Осухівська Г. М., Хвостівська Л. В., Величко Д. В. Розвиток математичного моделювання трафіку комп’ютерних мереж. Матеріали Міжнародної науково-технічної конференції 14–15 травня 2020 року «Фундаментальні та прикладні проблеми сучасних технологій», Тернопіль, Україна. 2021. C. 107–111. DOI: 10.1425/jsdtl
Khvostivska L., Khvostivskyi M., Dediv I., Yatskiv V., Palaniza Y. Method, algorithm and computer tool for synphase detection of radio signals in telecommunication networks with noises. CITI 2023: Proc. of the 1st International Workshop on Computer Information Technologies in Industry 4.0. Тернопіль, Україна. 2023. C. 173–180. ISSN 1613–0073.
