QoE Optimization for AR/VR Services over 5G Networks Using Deep RL
Keywords:
5G, augmented reality, virtual reality, quality of experience, deep reinforcement learning, network slicing, resource allocation, latency, fairness, governanceAbstract
The proliferation of augmented reality and virtual reality (AR/VR) applications imposes stringent quality of experience (QoE) requirements that fifth-generation (5G) networks must simultaneously satisfy. These requirements span ultra-low latency, high throughput, consistent reliability, and seamless mobility, yet they are often in tension due to shared radio resources, dynamic channel conditions, and heterogeneous service demands. Traditional optimization approaches, such as static resource allocation or reactive heuristics, struggle to adapt to the highly stochastic and multi-objective nature of AR/VR traffic. This paper presents a comprehensive system-level investigation into the use of deep reinforcement learning (Deep RL) for QoE optimization of AR/VR services over 5G networks. We examine the architectural trade-offs involved in integrating Deep RL agents within the 5G core and radio access network, focusing on state representation, action spaces, reward design, and training stability. We extend the discussion to deployment considerations, including computational overhead, latency constraints, and the need for online versus offline learning. Robustness and fairness are analyzed in the context of service-level agreements and network slicing. Governance and policy implications are explored, particularly regarding data privacy, model transparency, and accountability in autonomous network decision-making. Through a synthesis of recent advances in deep reinforcement learning, network slicing, and QoE modeling, we argue that Deep RL offers a compelling path toward adaptive, multi-objective resource control, yet significant challenges remain in real-world deployment. This paper contributes a structured framework for researchers and practitioners aiming to build intelligent, self-optimizing 5G infrastructures capable of delivering high-quality immersive experiences.
References
1. Elbamby, M. S., Perfecto, C., Bennis, M., & Doppler, K. (2018). Toward low-latency and ultra-reliable virtual reality. IEEE Communications Magazine, 56(3), 78–84. https://doi.org/10.1109/MCOM.2018.1700882
2. Yang, J., He, S., Lin, Y., & Lv, Z. (2019). Multimedia quality of experience in mobile edge computing. IEEE Access, 7, 152468–152486. https://doi.org/10.1109/ACCESS.2019.2946978
3. Bastug, E., Bennis, M., & Debbah, M. (2014). Living on the edge: The role of proactive caching in 5G wireless networks. IEEE Communications Magazine, 52(8), 82–89. https://doi.org/10.1109/MCOM.2014.6871674
4. He, Q., Shi, W., Chen, Q., & Yao, S. (2020). A survey on QoE of immersive video services over 5G networks. IEEE Access, 8, 158882–158902. https://doi.org/10.1109/ACCESS.2020.3010495
5. Mao, H., Alizadeh, M., Menache, I., & Kandula, S. (2016). Resource management with deep reinforcement learning. In Proceedings of the 15th ACM Workshop on Hot Topics in Networks (pp. 50–56). https://doi.org/10.1145/3005745.3005755
6. Li, Q. (2026). QoS Assurance Mechanism for 5G Network Slicing Based on the Deep Reinforcement Learning PPO Algorithm. arXiv preprint arXiv:2605.03345.
7. Kety, A., Vuppala, S., & Ratnam, V. (2021). QoE prediction for adaptive video streaming using machine learning. IEEE Transactions on Broadcasting, 67(2), 458–471. https://doi.org/10.1109/TBC.2020.3044583
8. Zhang, H., Liu, N., Chu, X., Long, K., & Nallanathan, A. (2020). Network slicing based 5G and future mobile networks: Mobility, resource management, and challenges. IEEE Communications Magazine, 55(8), 138–145. https://doi.org/10.1109/MCOM.2017.1600945
9. Raviv, Y., & Yerushalmi, D. (2020). Dynamic network slicing for 5G using machine learning: A survey. IEEE Communications Surveys & Tutorials, 22(4), 2487–2521. https://doi.org/10.1109/COMST.2020.3011432
10. Wang, L., Chen, M., Zhang, Y., & Poor, H. V. (2019). Deep reinforcement learning for resource allocation in 5G ultra-reliable low-latency communications. IEEE Transactions on Communications, 67(10), 7136–7149. https://doi.org/10.1109/TCOMM.2019.2927079
11. Huang, Y., & Ye, Q. (2021). Hierarchical multi-agent deep reinforcement learning for network slicing in 5G. IEEE Transactions on Mobile Computing, 20(11), 3252–3265. https://doi.org/10.1109/TMC.2020.2994321
12. Li, J., Shi, W., & Ji, Y. (2020). Multi-objective deep reinforcement learning for QoE-aware resource allocation in wireless networks. IEEE Transactions on Wireless Communications, 19(12), 8173–8187. https://doi.org/10.1109/TWC.2020.3016954
13. Mnih, V., Kavukcuoglu, K., Silver, D., et al. (2015). Human-level control through deep reinforcement learning. Nature, 518(7540), 529–533. https://doi.org/10.1038/nature14236
14. Schulman, J., Wolski, F., Dhariwal, P., Radford, A., & Klimov, A. (2017). Proximal policy optimization algorithms. arXiv preprint arXiv:1707.06347.
15. Zhou, Y., Jia, Q., & Zhao, Y. (2021). Adaptive bitrate streaming for virtual reality using deep reinforcement learning. IEEE Transactions on Circuits and Systems for Video Technology, 31(8), 3223–3235. https://doi.org/10.1109/TCSVT.2020.3038081
16. Lu, S., Li, Y., & Chen, J. (2022). Federated deep reinforcement learning for network slicing in multi-operator environments. IEEE Transactions on Network and Service Management, 19(1), 56–70. https://doi.org/10.1109/TNSM.2021.3126584
17. Lowe, R., Wu, Y., Tamar, A., Harb, J., Abbeel, P., & Mordatch, I. (2017). Multi-agent actor-critic for mixed cooperative-competitive environments. In Advances in Neural Information Processing Systems (pp. 6379–6390).
18. Peng, M., Liu, J., & Wang, Y. (2020). Resource allocation in 5G networks: A reinforcement learning approach with safety constraints. IEEE Transactions on Cognitive Communications and Networking, 6(2), 499–511. https://doi.org/10.1109/TCCN.2020.2981537
19. Chen, T., & Liu, X. (2021). Safe reinforcement learning for real-time network control. Journal of Network and Computer Applications, 186, 103080. https://doi.org/10.1016/j.jnca.2021.103080
20. Suryam, A., & Niyato, D. (2022). Digital twins for 6G network optimization using deep reinforcement learning. IEEE Wireless Communications, 29(4), 120–127. https://doi.org/10.1109/MWC.001.2100554
21. Hou, Y., & Cho, H. (2021). Deep learning for QoE estimation in VR video streaming. In 2021 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting (pp. 1–5). https://doi.org/10.1109/BMSB53066.2021.9547943
22. Gupta, A., & Moharir, S. (2020). Fairness-aware resource allocation using multi-objective reinforcement learning. IEEE Communications Letters, 24(9), 1942–1946. https://doi.org/10.1109/LCOMM.2020.2997010
23. Altman, E., & Kamoun, J. (2020). Constrained Markov decision processes for network resource management. Queueing Systems, 95(1–2), 1–33. https://doi.org/10.1007/s11134-020-09637-8
24. Floridi, L., & Cowls, J. (2019). A unified framework of five principles for AI in society. Harvard Data Science Review, 1(1). https://doi.org/10.1162/99608f92.8cd550d1
25. Tene, O., & Polonetsky, J. (2012). Big data for all: Privacy and user control in the age of analytics. Northwestern Journal of Technology and Intellectual Property, 11(5), 239–273.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Journal of Data Intelligence and AI Systems
This work is licensed under a Creative Commons Attribution 4.0 International License.
This article is published under the Creative Commons Attribution 4.0 International License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
