Zastosowanie metod uczenia maszynowego w sieciach teleinformatycznych

• Autorzy: Walkowiak, Krzysztof , Knapińska, Aleksandra , Lechowicz, Piotr

Warianty tytułu

EN

Application of machine learning methods in communication networks

• Konferencja: Konferencja Radiokomunikacji i Teleinformatyki (11-13.09.2024 ; Poznań, Polska)
• Języki publikacji: PL

Abstrakty

PL

W ostatnich latach ważnym czynnikiem wpływającym na rozwój sieci teleinformatycznych są metody uczenia maszynowego. Wynika to głównie z dużej konkurencji na rynku usług sieciowych, co pociąga za sobą bezustanną potrzebę jednoczesnego usprawnienia działania sieci komputerowych oraz obniżania kosztów działania sieci komputerowych. W artykule zostaną omówione przykładowe zastosowania metod uczenia maszynowego w sieciach teleinformatycznych. Zostaną przedstawione najnowsze technologie sieci teleinformatycznych, w których stosowane są metody uczenia maszynowego, w tym: automatyzacja sieci oraz koncepcja cyfrowego bliźniaka. Zostaną również zaprezentowane najważniejsze wyzwania związane ze stosowaniem metod uczenia maszynowego w sieciach teleinformatycznych, takie jak: dostęp do danych, potrzeba ciągłej aktualizacji modeli w związku ze zmieniającymi się wzorcami w sieciach, wyjaśnialną sztuczną inteligencję (ang. Explainable Artificial Intelligence).

EN

In recent years, machine learning (ML) methods have been an important factor influencing the development of communications networks. This is mainly due to the high competition in the ICT sector, which entails a relentless need to simultaneously improve the operation of communication computer and reduce the OPEX and CAPEX cost of networks. The paper will discuss examples of applications of machine learning methods in communication networks. The latest networking technologies that use machine learning methods will be presented, including network automation and digital twin. The most important challenges of applying machine learning methods in communication networks will also be described, including: datasets, updating ML models to changing patterns in networks, Explainable Artificial Intelligence. Keywords: communication network, machine learning, optimization, network automation, digital twin. The most important challenges of applying machine learning methods in communication networks will also be describes, including: datasets, updating ML models to changing patterns in networks, Explainable Artificial Intelligence.

Słowa kluczowe

PL

sieci teleinformatyczne; uczenie maszynowe; optymalizacja; automatyzacja sieci; cyfrowy bliźniak

EN

microstrip antennas; antenna simulation; antenna gain; reflection coefficient; FDTD method

• Czasopismo: Przegląd Telekomunikacyjny + Wiadomości Telekomunikacyjne
• Rocznik: 2024
• Tom: nr 4
• Strony: 18--24
• Opis fizyczny: Bibliogr. 32 poz., rys.

Twórcy

autor

Walkowiak, Krzysztof

• Politechnika Wrocławska, Wrocław,

autor

Knapińska, Aleksandra

• Politechnika Wrocławska, Wrocław

autor

Lechowicz, Piotr

• Politechnika Wrocławska, Wrocław
• Chalmers University of Technology, Gothenburg, Sweden

Bibliografia

• [1] Ericsson Company, Ericsson mobility report, June 2024, https:// www.ericsson.com/en/reports-and-papers/mobility-report/ reports/june-2024, 2024.
• [2] I. Morris, 2023 in review: Job changes and layoffs, https:// www.lightreading.com/ai-machine-learning/2023-in-review-jobchanges- and-layoffs, 2023.
• [3] F. Musumeci, C. Rottondi, A. Nag, I. Macaluso, D. Zibar, M. Ruffini, M. Tornatore, An overview on application of machine learning techniques in optical networks, IEEE Communications Surveys & Tutorials 21 (2), 1383–1408. doi:10.1109/COMST.2018.2880039, 2018.
• [4] T. Panayiotou, M. Michalopoulou, G. Ellinas, Survey on machine learning for traffic-driven service provisioning in optical networks, IEEE Communications Surveys & Tutorials 25 (2) 1412–1443. doi:10.1109/COMST.2023.3247842, 2023.
• [5] E. Coronado, R. Behravesh, T. Subramanya, A. Fernàndez- Fernàndez, M. S. Siddiqui, X. Costa-Pérez, R. Riggio, Zero touch management: A survey of network automation solutions for 5g and 6g networks, IEEE Communications Surveys & Tutorials 24 (4), 2535–2578. doi:10.1109/ COMST.2022.3212586, 2022.
• [6] C. Natalino, A. Panahi, N. Mohammadiha, P. Monti, AI/ML-as-aservice for optical network automation: use cases and challenges [invited], Journal of Optical Communications and Networking 16 (2), A169–A179. doi:10.1364/JOCN.500706, 2024.
• [7] D. Rafique, L. Velasco, Machine learning for network automation: overview, architecture, and applications [invited tutorial], Journal of Optical Communications and Networking 10 (10), D126–D143. doi:10.1364/JOCN.10.00D126, 2018.
• [8] L. Zhang, X. Li, Y. Tang, J. Xin, S. Huang, A survey on QoT prediction using machine learning in optical networks, Optical Fiber Technology 68, 102804. doi:10.1016/j.yofte.2021.102804, 2022.
• [9] Cisco Company, What is network automation?, https://www. cisco.com/c/en/us/solutions/automation/network-automation. html.
• [10] A. Leivadeas, M. Falkner, A survey on intent-based networking, IEEE Communications Surveys Tutorials 25 (1), 625–655. doi:10.1109/COMST.2022.3215919, 2023.
• [11] Cisco Company, What is intent-based networking(IBN)? https:// www.cisco.com/c/en/us/solutions/intent-based-networking.html.
• [12] Clemm, A., Ciavaglia, L., Granville, L., and J. Tantsura, "Intent-Based Networking – Concepts and Definitions", RFC 9315, DOI 10.17487/ RFC9315, https://www.rfc-editor.org/info/rfc9315, 2022.
• [13] F. Tao, H. Zhang, A. Liu, A. Y. C. Nee, Digital twin in industry: State- of-the-art, IEEE Transactions on Industrial Informatics 15 (4), 2405–2415. doi:10.1109/TII.2018.2873186, 2019.
• [14] P. Almasan, M. Ferriol-Galmés, J. Paillisse, J. Suárez-Varela, D. Perino, D. López, A. A. P. Perales, P. Harvey, L. Ciavaglia, L. Wong, V. Ram, S. Xiao, X. Shi, X. Cheng, A. Cabellos-Aparicio, P. Barlet- Ros, Network digital twin: Context, enabling technologies, and opportunities, IEEE Communications Magazine 60 (11), 22–27. doi:10.1109/MCOM.001.2200012, 2022
• [15] D. Wang, Y. Song, Y. Zhang, X. Jiang, J. Dong, F. N. Khan, T. Sasai, S. Huang, A. P. T. Lau, M. Tornatore, M. Zhang, Digital twin of optical networks: A review of recent advances and future trends, Journal of Lightwave Technology, 1–28, doi:10.1109/JLT.2024.3401419, 2024.
• [16] Seattle Internet Exchange Point. https://www.seattleix.net/statistics/.
• [17] P. Lechowicz, R. Goscien, R. Rumipamba-Zambrano, J. Perello, S. Spadaro, K. Walkowiak, Greenfield gradual migration planning toward spectrally-spatially flexible optical networks, IEEE Communications Magazine 57 (10), 14–19. doi:10.1109/ MCOM.001.1900207, 2019
• [18] S. Petale, S.-C. Lin, M. Matsuura, H. Hasegawa, S. Subramaniam, PRODIGY: A progressive upgrade approach for elastic optical networks, in: IEEE Global Communications Conference (GLOBECOM), pp. 2129–2134. doi:10.1109/ GLOBECOM54140.2023.10437935, 2023.
• [19] A. Knapińska, P. Lechowicz, K. Walkowiak, Machine-learning based prediction of multiple types of network traffic, in: International Conference on Computational Science (ICCS), Springer, pp. 122–136. doi:10.1007/978-3-030-77961-0_12, 2021
• [20] A. Knapińska, P. Lechowicz, W. Węgier, K. Walkowiak, Long-term prediction of multiple types of timevarying network traffic using chunk-based ensemble learning, Applied Soft Computing 130, 109694. doi:10.1016/j.asoc.2022.109694, 2022
• [21] A. Knapińska, P. Lechowicz, K. Walkowiak, Prediction of multiple types of traffic with a novel evaluation metric related to bandwidth blocking, in: IEEE Global Communications Conference (GLOBECOM), pp. 2927–2932. doi:10.1109/ GLOBECOM48099.2022.10001028, 2022
• [22] A. Knapińska, P. Lechowicz, S. Spadaro, K. Walkowiak, Agnostic prediction of multiple types of timevarying traffic in optical networks, in: IEEE Global Communications Conference (GLOBECOM), pp. 1131–1136. doi:10.1109/GLOBECOM54140. 2023.10436763, 2023.
• [23] Sandvine, The mobile internet phenomena report (May 2021), https://www.sandvine.com/download-mobile-internet-phenomena- report-2021, 2021.
• [24] P. Lechowicz, A. Knapińska, A. Włodarczyk, K.Walkowiak, Traffic Weaver: semi-synthetic time-varying traffic generator based on averaged time series, https://arxiv.org/abs/2403.11388, 2024.
• [25] I. Lohrasbinasab, A. Shahraki, A. Taherkordi, A. Delia Jurcut, From statistical-to machine learningbased network traffic prediction, Transactions on Emerging Telecommunications Technologies 33 (4), e4394. doi:10.1002/ETT.4394, 2022
• [26] G. O. Ferreira, C. Ravazzi, F. Dabbene, G. C. Calafiore, M. Fiore, Forecasting network traffic: a survey and tutorial with open-source comparative evaluation, IEEE Access Vol. 11, 6018–6044. doi: 10.1109/ACCESS.2023.3236261, 2023
• [27] A. Knapińska, R. Goścień, P. Lechowicz, K. Walkowiak, Link load prediction in an optical network with restoration mechanisms, Journal of Optical Communications and Networking 15 (5), B42– B52. doi:10.1364/JOCN.479849, 2023.
• [28] S. M. Lundberg, S.-I. Lee, A unified approach to interpreting model predictions, Advances in neural information processing systems (NIPS) 30, 2017.
• [29] A. Knapińska, O. Ayoub, C. Rottondi, P. Lechowicz, K. Walkowiak, Explainable artificial intelligence-guided optimization of ML-based traffic prediction, in: 28th International Conference on Optical Network Design and Modeling (ONDM), pp. 1–6, 2024.
• [30] A. Knapińska, K. Półtorak, D. Poręba, J. Miszczyk, M. Daniluk, K. Walkowiak, On feature selection in short-term prediction of backbone optical network traffic, in: 26th International Conference on Optical Network Design and Modeling (ONDM), pp. 1–6, 2022.
• [31] A. Knapińska, R. Kanimba, Y. Yesilyurt, K. Walkowiak, Application of ensemble regression methods in elastic optical network optimization, in: 5th Polish Conference on Artificial Intelligence (PP-RAI), pp. 1–6, 2024.
• [32] K. Duszyńska, P. Polski, M. Włosek, A. Knapińska, P. Lechowicz, K. Walkowiak, XAI-guided optimization of a multilayer network regression model, in: 1st International Workshop on Trustworthy and Explainable Artificial Intelligence for Networks (TX4Nets) at the IFIP/IEEE Networking Conference, pp. 170–175, 2024.

Identyfikatory

DOI

10.15199/59.2024.4.2