Machine Learning Techniques for Resource Allocation In Wireless Communication Networks: a Review

Opeyemi A. Ajibola; Samuel B. Ajibowu; Adewole A. Osobukola; Mubaraq O. Asafa; Joseph B. Samson; Surajudeen A. Babatunde

doi:10.5455/EJCSI.20250530124623

Abstract
In wireless communication networks (WCNs), resource allocation has become an important task due to the varying nature of wireless networks. Conventional approaches for resource allocation always find it difficult to adapt to the complex nature and real-time variations of network conditions. Machine Learning (ML) has emerged as a promising technology to improve resource allocation leveraging data-driven insights and intelligent decision-making. This paper reviews various ML algorithms including Graph Neural Network (GNN), Deep Reinforcement Learning (DRL), Federated Learning (FL), Federated Reinforcement Learning (FRL), and GNN-based DRL for allocating resources in wireless communication networks. The paper further discusses the applications of ML for WCNs resource allocation, the challenges in ML based resource allocation, and recommends prospective future directions.

Key words: Machine Learning, Resource Allocation, Wireless Communication, Federated Learning, Deep Reinforcement Learning