Published Paper Details:

The fifth generation (5G) network delivers ultra-high data rates, massive connectivity, and low-latency services essential for applications such as autonomous systems, smart cities, and industrial automation. However, the rapidly fluctuating traffic demands and heterogeneous service requirements create significant challenges for efficient allocation of limited network resources. Static allocation strategies lead to congestion, uneven performance, and inefficient spectrum usage. To address these limitations, the study focuses on an AI-driven adaptive orchestration approach that dynamically optimizes resource distribution in 5G environments. A nested neural decision model combined with a Markov decision process enables intelligent prediction of traffic patterns and proactive re-allocation of resources in real time. The system supports QoS-aware decision-making across network slices, ensuring fairness, reduced delay, and enhanced throughput. Simulation-based analysis demonstrates improved resource utilization efficiency, minimized latency, and better user experience when compared to conventional allocation techniques. This study highlights the significance of autonomous learning-based resource management for advancing next-generation mobile communication networks.