2026 International Conference on Networks Computers and Communications (ICNCCom 2026)

Modern 5G and 6G mobile networks face significant challenges in radio resource management (RRM) due to dynamic traffic patterns, geographic variability, and limited spectrum utilization. Traditional centralized resource allocation schemes rely primarily on user equipment (UE) channel quality feedback and reactive scheduling, which fail to capture area-level intelligence and collaborative edge learning opportunities.This paper proposes a Federated Learning (FL)-based AI driven resource scheduling framework that enables distributed, privacy-preserving machine learning across base stations while maintaining predictive accuracy for congestion-aware resource allocation. The framework leverages geographic clustering, traffic prediction, and federated aggregation to optimize spectrum utilization and reduce handover failures. Simulation-based evaluation demonstrates improvements in spectral efficiency, QoS performance, and network reliability compared to conventional scheduling techniques.

Federated Learning,5G,6G,Resource Scheduling,Radio Resource Management,Edge Intelligence,Machine Learning