A Joint Hybrid Resource Allocation Framework for Enhanced Performance in Cloud Radio Access Networks

Abstract

The fifth-generation (5G) mobile network aims to support high data rates and massive connectivity, driven in part by small cell technology. The Cloud Radio Access Network (C-RAN) has emerged as a promising architecture to address the increasing resource demands of a growing user base by separating base station functions into centralized baseband units (BBUs) and distributed remote radio heads (RRHs). Despite its advantages, C-RAN introduces challenges in efficiently allocating resources to dynamic and heterogeneous users. This paper proposes a hybrid resource allocation framework that enhances system efficiency while meeting varying user demands in C-RAN environments. The approach integrates centralized control with multi-agent-based decision-making, where a centralized controller within the BBU pool collaborates with virtual base stations (VBSs) acting as agents in a multi-agent system (MAS). Resource allocation decisions are made by jointly considering real-time resource requests from agents and historical demand predictions generated by the centralized controller. Simulation results demonstrate that the proposed method outperforms conventional random and fixed allocation schemes in terms of resource utilization, fairness, and reduction of unmet user demands. By effectively combining realtime and historical information, the proposed strategy ensures improved long-term resource planning and adaptability, making it a robust solution for dynamic C-RAN systems.