METHODS OF MANAGING A FUNCTIONALLY RESILIENT HETEROGENEOUS TELECOMMUNICATION NETWORK UNDER THE INFLUENCE OF DESTABILIZING FACTORS

DOI 10.31673/2412-4338.2025.017512

Abstract

The article examines methods for managing a functionally resilient heterogeneous telecommunication network under the influence of destabilizing factors. An analysis of existing approaches to ensuring network infrastructure resilience has been conducted, including adaptive routing algorithms, traffic load prediction methods, probabilistic reliability assessment models, resource reservation strategies, and blockchain-oriented security mechanisms. It is shown that each of these methods has its own advantages and limitations that must be considered when developing effective management mechanisms. The study of adaptive routing algorithms, particularly Bellman-Ford and Dijkstra, demonstrates their efficiency in static conditions; however, they exhibit insufficient performance in highly dynamic environments due to high computational complexity and the necessity of precomputing routes. Load forecasting methods, such as ARIMA and LSTM, enhance network management efficiency by accurately predicting traffic changes. However, they have significant computational complexity and require high-quality input data. Probabilistic approaches to network resilience modeling have been considered, including the use of Markov chains for assessing failure risks and ensuring infrastructure reliability. However, it has been found that these methods are often based on assumptions that do not always correspond to real operational conditions. Resource reservation strategies, particularly dynamic channel reservation and M/M/1 models, contribute to improving network resilience. Nevertheless, they require precise load forecasting and optimal allocation of computational resources. Special attention is given to the use of blockchain technologies in authorization processes, data transmission security, and decentralized resource distribution. The use of smart contracts for network management automation is proposed, which reduces the influence of human factors and enhances the security of telecommunication systems. Prospects for further research include the development of hybrid management methods that combine artificial intelligence, machine learning, and blockchain to ensure the adaptability of telecommunication networks. In particular, the use of deep neural networks for traffic optimization and the integration of quantum computing to accelerate route data processing have been proposed. Additionally, the development of methods for early cyberattack detection based on network traffic anomaly analysis remains a relevant direction. Thus, the conducted research demonstrates the importance of a comprehensive approach to managing heterogeneous telecommunication networks and emphasizes the need for further integration of modern technologies to enhance the functional resilience of systems under destabilizing factors.

Keywords: functional resilience, telecommunication networks, adaptive management, load forecasting, routing, probabilistic models, blockchain, artificial intelligence, machine learning, resource reservation.