These theses present an advanced methodology for intelligent traffic management and dynamic resource allocation in next-generation network infrastructures. The proposed approach leverages Software-Defined Networking (SDN) combined with Deep Reinforcement Learning (DRL) techniques to enhance network scalability, optimize Quality of Service (QoS), and ensure efficient resource utilization. Experimental results demonstrate significant improvements in network performance, latency reduction, and anomaly detection accuracy. This study emphasizes the practical applicability of intelligent control systems in large-scale environments, such as 5G, IoT, and smart city networks, paving the way for the development of autonomous, adaptive network management solutions.
Onatskyi et al. (Tue,) studied this question.