The concept of using explainable artificial intelligence to solve the problem of traffic distribution

Modern infocommunication networks are a set of a significant number of nodes and communication channels between them, which complicates the process of analyzing the characteristics of such networks. Analytical methods contain a large number of assumptions and cannot fully reflect all the nuances of maintenance procedures. Simulation allows you to take into account many components of maintenance procedures, but requires a lot of time and computing costs to assess the characteristics of real networks. Currently, there are machine learning methods that can be used to solve traffic distribution problems by evaluating service quality characteristics. The methods of explainable artificial intelligence allow us to obtain information about the most influential factors in order to use them further in solving the problem of traffic distribution. Therefore, the purpose of this work is to develop the concept of using explainable artificial intelligence to solve the problem of traffic distribution in an infocommunication network

Keywords: traffic distribution, quality of service, explainable artificial intelligence, infocommunications networks, time delay.

References

  1. V. S. Elagin, A. S. Vasin. Analiz modelej upravleniya setevymi resursami v setyah 5G [Analysis of network resource management models in 5G networks]//T-Comm: Telekommunikacii i transport [T-Comm: Telecommunications and transport]. 2023. Vol. 17. № 5. P. 32-41.
  2. N. M. Redrugina. Modeli i metody vychisleniya zaderzhek pri predostavlenii uslug pol'zovatelem na servisnyh platformah seansovyh infokommunikacionnyh uslug [Models and methods for calculating delays in the provision of services by the user on the service platforms of session infocommunication services]//T-Comm: Telekommunikacii I transport [T-Comm: Telecommunications and transport]. 2023. Vol. 17. № 4. P. 32-38.
  3. M. E. Sudareva, V. N. Yashin. Analiz granichnyh zaderzhek v seti TSN Ethernet [Analysis of boundary delays in the TSN Ethernet network]//Infokommunikacionnye tekhnologii [Information and communication technologies]. 2023. Vol. 21. № 2 (82). P. 37-43.
  4. D. Yu. Ponomarev. Ocenka srednego vremeni zaderzhki v infocomminikacionnykh setyakh s ispolzovaniem metodov tenzornogo analiza//Modelirovanie, optimizatciya I informacionnye tekhnologii. Nauchny zhurnal. 2018. T. 6. # 4. S. 155-167.
  5. D. Yu. Ponomarev, A. A. Lacinnik. Ocenka harakteristik processov peredachi informacii v kombinirovannoj sputnikovoj seti [Assessment of the characteristics of information transmission processes in a combined satellite network]//Sistemy svyazi i radionavigacii. Sb. tezisov [Communication and radio navigation systems: collection of abstracts]. Krasnoyarsk: AO «NPP «Radiosvyaz'», 2023. P. 127-130.
  6. A. V. Morozov, D. Yu. Ponomarev. Model' optimizacii raspredeleniya informacionnyh potokov v seti interneta veshchej [A model for optimizing the distribution of information flows in the Internet of Things network]//Izvestiya Rossijskoj akademii raketnyh i artillerijskih nauk [Proceedings of the Russian Academy of Rocket and Artillery Sciences]. 2023. № 3 (128). P. 125-132.
  7. R. M. Shukla, S. Sengupta, A. N. Patra. Software-defined network based resource allocation in distributed servers for unmanned aerial vehicles//IEEE 8th Annual Computing and Communication Workshop and Conference. 2018. P. 796-802.
  8. T. Springer, E. Linstead, P. Zhao, C. Parlett-Pelleriti. Towards QoS-Based Embedded Machine Learning//Electronics. 2022. 11. P. 3204-3221.
  9. P. Larrenie, J. F. Bercher, I. Lahsen-Cherif, O. Venard. Low Complexity Adaptive ML Approaches for End-to-End Latency Prediction//Machine Learning for Networking. MLN 2022. Lecture Notes in Computer Science. Vol. 13767. Cham: Springer, 2023.
  10. I. S. Lebedev, I. A. Sikarev, M. E. Suhoparov, B. T. Rzaev. Povyshenie kachestvennyh pokazatelej metodov mashinnogo obucheniya pri obrabotke setevogo trafika na osnove segmentirovaniya vyborki dannyh [Improving the quality indicators of machine learning methods in processing network traffic based on segmentation of a data sample]//T-Comm: telekommunikacii i transport. [T-Comm: telecommunications and transport]. 2023. Vol. 17. № 9. P. 48-54.
  11. F. Krasniqi, J. Elias, J. Leguay, A. E. C. Redondi. End-to-end Delay Prediction Based on Traffic Matrix Sampling//IEEE Infocom 2020 — IEEE Conference on Computer Communications Workshops. 2020. P. 774-779.
  12. A. Palaios, Ch. Vielhaus, D. F. Kulzer et al. Machine Learning for QoS Prediction in Vehicular Communication: Challenges and Solution Approaches//IEEE Access. 2023. Vol. 11. P. 92459-92477.
  13. M. Farreras, P. Soto, M. Camelo et al. Improving Network Delay Predictions Using GNNs//Journal Network System Management. 2023. 31. P. 65-101.
  14. K.-C. Tsai, Z. Zhuang, R. Lent et al. Tensor-Based Reinforcement Learning for Network Routing//IEEE Journal of Selected Topics in Signal Processing. 2021. 15 (3). P. 617-629.
  15. A. V. Morozov, D. Yu. Ponomarev. Ispol'zovanie metodov Data Science s cel'yu prognozirovaniya srednego vremeni zaderzhki v infokommunikacionnyh setyah [Using DataScience methods to predict the average delay time in infocommunication networks]//Izvestiya Rossijskoj akademii raketnyh i artillerijskih nauk. [Proceedings of the Russian Academy of Rocket and Artillery Sciences]. 2023. № 4 (129). P. 96-103.
  16. M. Mehta, V. Palade, I. Chatterjee. Explainable AI: Foundations, Methodologies and Applications. Intelligent Systems Reference Library. Vol. 232. Cham: Springer, 2023.
  17. G. Wassie, J. Ding, Y. Wondie. Traffic prediction in SDN for explainable QoS using deep learning approach//Scientific Reports. 2023. № 13. P. 20607-20623.

Authors