Research Article
Comparative Analysis of RL-Based Resource Allocation Methods for Optimization in 5G MMWave Network
@INPROCEEDINGS{10.1007/978-3-031-77075-3_32, author={V. Shilpa and Rajeev Ranjan}, title={Comparative Analysis of RL-Based Resource Allocation Methods for Optimization in 5G MMWave Network}, proceedings={Cognitive Computing and Cyber Physical Systems. 5th EAI International Conference, IC4S 2024, Bhimavaram, India, April 5--7, 2024, Proceedings, Part-I}, proceedings_a={IC4S}, year={2025}, month={2}, keywords={Reinforcement Learning (RL) Resource Allocation 5G mmWave State-of-the-art Q-learning Optimization Resource Allocation (RA)}, doi={10.1007/978-3-031-77075-3_32} }- V. Shilpa
Rajeev Ranjan
Year: 2025
Comparative Analysis of RL-Based Resource Allocation Methods for Optimization in 5G MMWave Network
IC4S
Springer
DOI: 10.1007/978-3-031-77075-3_32
Abstract
In this study, resource allocation techniques based on reinforcement learning (RL) for 5G millimeter wave (mmWave) networks are compared and analyzed. The high bandwidth and large available spectrum in mmWave networks offer a great opportunity for high data rate communication but also pose significant challenges in terms of resource allocation. RL, being a powerful tool for decision making in dynamic and uncertain environments, has been widely studied for resource allocation in mmWave networks. In this paper, we survey the state-of-the-art RL-based resource allocation methods for mmWave networks and we also discuss the challenges and open research directions in this field. This study presents an effective method for balancing stable, low-capacity mmWave links with reliable, high-capacity backhaul links designed using stochastic geometry and analytical channel modeling. We also propose a backhaul resource allocation strategy. To increase system utility with a mmWave network’s limited backhaul capacity, a neural network-based backhaul resource allocation has been proposed.
