METHOD OF AGENT INTERACTION IN A MULTI-AGENT CYBERSECURITY MANAGEMENT SYSTEM OF A TRANSPORT TELECOMMUNICATION NETWORK DURING CYBERATTACK DIAGNOSIS
DOI: 10.31673/2412-4338.2026.019005
Abstract
The paper addresses the problem of organizing a multi-agent cybersecurity management system for transport telecommunication networks (TTN) under conditions of increasing complexity and stealthiness of cyberattacks. A hierarchical system model is proposed that integrates local and global control loops and is aligned with the structural organization of the TTN. The basic types of agents are defined, including monitoring, threat detection, risk assessment, decision-making, response, and coordination agents, as well as service- and policy-level agents. Their functional roles and information interactions within a unified control cycle are described. A classification of cyberattacks based on diagnostic determinability is proposed. Diagnostically detectable attacks, which cause network parameters to exceed permissible limits and can be identified by basic diagnostic mechanisms, are distinguished from diagnostically undetermined (stealth) attacks that do not violate controlled parameters and therefore require additional intelligent or correlation-based analysis. It is substantiated that TTNs are particularly susceptible to stealth attacks, where contradictory diagnostic information may be generated by self-diagnostic agents, complicating reliable attack identification. To improve diagnostic efficiency, a two-level stealth attack detection system is proposed. The first level implements a fast primary analysis algorithm based on monitoring and threat detection data. If the obtained results are insufficiently reliable, a second level employing enhanced risk assessment and decision-making procedures is activated. Simulation results confirm that diagnostic reliability increases with network scale and decreases acceptably as the number of compromised nodes grows. The diagnostic time increases almost linearly and remains predictable, demonstrating the scalability and practical applicability of the proposed approach.
Keywords: cybersecurity, transport telecommunication network, intelligent agent, multi-agent system, cyberattack diagnosis, stealth attack, hierarchical control model.
References
- Gol, V. D., & Irkha, M. S. (2021). Telecommunications and Information Networks. Kyiv, Igor Sikorsky Kyiv Polytechnic Institute, 250 p. https://ela.kpi.ua/server/api/core/bitstreams/35d4a2d2-53ed-453f-9bcd-fa883a982f53/content
- Panovyk, U. P. (2024). Cybersecurity in Telecommunication Networks and Systems. Naukovi Zapisky, 1(68), 122–135. https://nz.uad.lviv.ua/media/1-68/13.pdf
- Khoroshko, V., Khokhlachova, Y., & Vyshnevska, N. (2023). Decomposition of Computer Network Technology In Their Design. Ukrainian Scientific Journal of Information Security, 29(3), 130–137. https://doi.org/10.18372/2225-5036.29.18072
- Khavina, I. P., Hnusov, Yu. V., & Mozhaiev, O. O. (2022). Development of multi-agent information security management system. Law and Safety, 87(4), 171–183. https://doi.org/10.32631/pb.2022.4.14
- Kitura, O. V. (2023). Methodology for forming a transport network control system. Dissertation Doctor of Philosophy in speciality 172 “Telecommunications and Radio Engineering”. Kyiv, DUT, 133 p. https://duikt.edu.ua/uploads/p_2625_85571738.pdf
- Bougueroua, N., et al. (2021). A Survey on Multi-Agent Based Collaborative Intrusion Detection Systems. Journal of Artificial Intelligence and Soft Computing Research, 11(2), 111–142. https://doi.org/10.2478/jaiscr-2021-0008
- Torres, M. (2025). Enhancing Distributed Intrusion Detection Systems Using Multi-Agent AI Models. International Annals of Intelligent Learning Systems Research (IAILSR), 9, 22–35. https://iailsr.org/index.php/iailsr/article/view/13
- Sen, J. (2011). A Distributed Intrusion Detection System Using Cooperating Agents. arXiv:1111.0382. https://doi.org/10.48550/arXiv.1111.0382
- Aydın, H., Aydın, G. Z. G., Sertbaş, A., & Aydın, M. A. (2023). Internet of things security: A multi-agent-based defense system design. Computers and Electrical Engineering, 111(B), 108961, https://doi.org/10.1016/j.compeleceng.2023.108961
- Landolt, C. R., Würsch, C., Meier, R., Mermoud, A., & Jang-Jaccard, J. (2025). Multi-Agent Reinforcement Learning in Cybersecurity: From Fundamentals to Applications. arXiv:2505.19837. https://doi.org/10.48550/arXiv.2505.19837
- Kozlovsky, O. V., & Zharikova, M. V. (2025): Development of a security model for a multi-agent network in a cyber-physical system. Bulletin of the Kherson National Technical University, 2, 1(92), 76–83. https://doi.org/10.35546/kntu2078-4481.2025.1.2.11
- Shamshirband, S., Anuar, N. B., Kiah, M. L. M., & Patel, A. (2013). An appraisal and design of a multi-agent system based cooperative wireless intrusion detection computational intelligence technique, Engineering Applications of Artificial Intelligence, 26(9), 2105–2127. https://doi.org/10.1016/j.engappai.2013.04.010
- Gallo, A. J., Barboni, A., & Parisini, T. (2020). On detectability of cyber-attacks for large-scale interconnected systems. Preprints of the 21st IFAC World Congress (Virtual), Berlin, Germany, July 12–17. https://ifatwww.et.uni-magdeburg.de/ifac2020/media/pdfs/1984.pdf
- Jakobsson, M., Wetzel, S., & Yener, B. (2003). Stealth attacks on ad-hoc wireless networks. IEEE Vehicular Technology Conference, 58(3), 2103–2111. https://doi.org/10.1109/vetecf.2003.1285396
- Chmut, O. V., Kalinichenko, O. G., & Bodashevsky, E. M. (2023). Technology for creating a fault-tolerant multi-module software complex based on the procedure of mutual internal checks. Modern Information Security, 4(56), 52–61. https://doi.org/10.31673/2409-7292.2023.030606
- Musienko, A. P. Methodological foundations of ensuring the functional stability of wireless sensor networks based on multi-criteria optimization: Dissertation of Doctor of Technical Sciences: special. 05.13.06 - Information Technologies. Kyiv, DUT, 2019. – 328 p.
- Hnatyuk, Ya. (2016). Logic: a modern perspective on traditional theory. Ivano-Frankivsk, “Symphony forte”, 2016, 356 p.