Reinforcement Learning-Enhanced Fair Resource Management for eMBB and URLLC Traffic in 5G Networks

Authors : Kasım Alpay, Muge Erel-ozcevik, Akın Özçift

Pages : 333-340

Doi:10.18586/msufbd.1749404

View : 28 | Download : 100

Publication Date : 2025-12-24

Article Type : Research Paper

Abstract :In this paper, we propose a novel reinforcement learning (RL)-aided adaptive scheduling mechanism for fair resource scheduling of enhanced Mobile Broadband (eMBB) and Ultra-Reliable Low Latency Communication (URLLC) traffic in 5G networks. To this end, we introduce a comprehensive Simulink simulation tool that visually facilitates intelligent QoS management based on adaptive thresholds and feedback performance loops. The suggested approach dynamically allocates resources by designing RL-like adaptive learning logic using off-the- shelf Simulink blocks. It also adapts in real-time to evolving traffic conditions by providing quality-of-service (QoS) differentiation. Experimental results demonstrate that our method achieves a 67% improvement in system efficiency and a 45% reduction in QoS violations compared to conventional baselines. This reflects effective learning dynamics and improved resource utilization, with O(1) computational complexity per scheduling decision.
Keywords : 5G Ağları, Uyarlanabilir Zamanlama, eMBB, Simulink, URLLC, Kaynak Yönetimi

ORIGINAL ARTICLE URL