Intelligent resource allocation for 5G cloud-radio access networks

The growing demand for mobile data services has driven efforts to improve the quality of service (QoS) in beyond 5G (B5G) networks. Moving beyond conventional static methods, 5G enables flexible programming using software-defined networking, slicing, and network capability exposure to optimise QoS. A key advancement in 5G is the cloud-radio access network (C-RAN), which separates baseband processing units from remote radio heads and centralises them in the cloud. This paper introduces an adaptive, QoS-aware resource allocation model for multi-tenant, multi-service C-RAN scenarios. A deep reinforcement learning (D-RL) approach is used to achieve optimal resource distribution. Experimental validation with a 5G prototype based on open air interface (OAI) confirms the superiority of this method in terms of network throughput, prediction accuracy, and system utilisation compared to other resource allocation techniques.