Efficient scheduler selection is critical in 5G networks to ensure optimal allocation of shared communication resources among users. Several previous studies have demonstrated that machine learning (ML) algorithms, such as k-nearest neighbor (KNN), Random Forest, and reinforcement learning, can enhance scheduling by enabling predictive decision-making based on traffic patterns and user mobility, dynamic resource allocation, and traffic-aware prioritization, surpassing the limitations of traditional fixed-algorithm schedulers. However, existing research has not fully exploited the potential of ML-based approaches that can adaptively select the most suitable scheduler. This paper explores two ML-based scheduling strategies in an uplink 5G system. The first one employs ML algorithms such as KNN, Naïve Bayes, Decision Tree, Random Forest, and multilayer perceptron (MLP) to directly predict the optimal user equipment (UE) to schedule. The second one proposes an adaptive scheduler selection mechanism using the same ML models. Both strategies incorporate Bayesian optimization for performance enhancement. Simulation results indicate that the first strategy yields better performance overall, with the KNN-optimized scheme outperforming the conventional best channel quality indicator (CQI) algorithm and the Decision Tree-optimized scheme by 21.7% and 22.1% in average throughput, respectively.
ImtiazSGhauchHKoudouridisGP, et al. Random forest resource allocation for 5G systems: performance and robustness study. In: 2018 IEEE wireless communications and networking conference workshops (WCNCW), Barcelona, 15–18 April.
8.
LopesVHLNahumCVDreifuerstRM, et al. Deep reinforcement learning-based scheduling for multiband massive MIMO. IEEE Access2022; 10: 130204–130214.
9.
ZhuYSunLWangJ, et al. Deep reinforcement learning-based joint scheduling of 5G and TSN in industrial networks. Electronics2023; 12: 2686.
10.
AnQSegarraSDickC, et al. A deep reinforcement learning-based resource scheduler for massive MIMO networks. IEEE Trans Mach Learn Commun Netw2023; 1: 242–257.
11.
ComşaISZhangSAydinME, et al. Towards 5G: a reinforcement learning-based scheduling solution for data traffic management. IEEE Trans Netw Serv Manag2018; 15: 1661–1675.
12.
Al-TamFCorreiaNRodriguezJ.Learn to schedule (LEASCH): a deep reinforcement learning approach for radio resource scheduling in the 5G MAC layer. IEEE Access2020; 8: 108088–108101.
13.
ZhangTShenSMaoS, et al. Delay-aware cellular traffic scheduling with deep reinforcement learning. In: Proceedings of GLOBECOM 2020-2020 IEEE global communications conference, Taipei, 7–11 December.
14.
YuanMCaoQPunMO, et al. Towards user scheduling for 6G: a fairness-oriented scheduler using multi-agent reinforcement learning. arXiv preprint, 2020. DOI:10.48550/arXiv.2012.15081.
15.
MathurSChabaYNoliyaA.Performance analysis of support vector machine learning based carrier aggregation resource scheduling in 5G mobile communication. Procedia Comput Sci2023; 218: 2776–2785.
16.
HabibyAAMMehaboobA. Application of reinforcement learning for 5G scheduling parameter optimization. arXiv:1911.07608, 2019.
17.
GuZSheCHardjawanaW, et al. Knowledge-assisted deep reinforcement learning in 5G scheduler design: from theoretical framework to implementation. IEEE J Sel Areas Commun2021; 39: 2014–2028.
18.
YanMFengGZhouJ, et al. Intelligent resource scheduling for 5G radio access network slicing. IEEE Trans Veh Technol2019; 68: 7691–7703.
19.
YuXGuoJLiX, et al. Deep learning based user scheduling for massive MIMO downlink system. Sci China Inf Sci2021; 64: 182304.
20.
JiangJQiuYSuY, et al. Low-latency resource elements scheduling based on deep reinforcement learning model for UAV video in 5G network. J Phys Conf Ser2021; 1827: 012071.
21.
ComşaISZhangSAydinM, et al. A comparison of reinforcement learning algorithms in fairness-oriented OFDMA schedulers. Information2021; 10: 315.
22.
VoraAKangK.Effective 5G wireless downlink scheduling and resource allocation in cyber-physical systems. In: IEEE 5G world forum (5GWF), Santa Clara, CA, USA, 9–11 July 2018.
23.
MilovanovicDABojkovicZSFowdurTP.Driving 5G mobile communications with artificial intelligence towards 6G. New York: CRC Press, 2023.
24.
MohammedMKhanMBBashierEBM. Machine learning: algorithms and applications. 1st ed. New York: CRC Press, 2016.
LeeVELiuLJinR.Decision Trees: theory and algorithms. In: AggarwalCC (ed.) Data Classification: Algorithms and Applications. New York: Chapman and Hall/CRC, 2014, pp. 87–120.
PerdanaDSanyotoANBisonoYG.Performance evaluation and comparison of scheduling algorithms on 5G networks using network simulator. Int J Comput Commun Control2019; 14: 530–539. doi:10.15837/ijccc.2019.4.3570.
46.
MamaneAFattahMEl GhaziM, et al. Proportional fair buffer scheduling algorithm for 5G enhanced mobile broadband. Int J Electr Comput Eng2021; 11: 4165–4173.
HuangYLiSLiC, et al. A deep reinforcement learning-based approach to dynamic eMBB/URLLC multiplexing in 5G NR. IEEE Internet Things J2020; 7: 6439–6456.
