End-to-End Energy Efficiency Evaluation of 5G Networks in Dense Urban and Rural Scenarios Using Empirical Modeling and Monte Carlo Simulation

Résumé fourni par la source

This paper provides a comprehensive end-to-end evaluation of the energy efficiency of three key fifth-generation mobile (5G) Radio Access Network (RAN) architectures: Distributed RAN (D-RAN), Centralized RAN (C-RAN), and Cloud C-RAN. Two complementary energy efficiency metrics are employed: the data energy efficiency metric EETH (Mbit/J) and the coverage energy efficiency metric EECO (m2/MJ). Analyses are conducted for dense urban and rural environments over a standardized 1 km2 area. Results demonstrate that 5G Cloud C-RAN consistently offers superior energy performance under both metrics. In dense urban conditions, Cloud C-RAN achieves a maximum EETH of 149.96 Mbit/J, representing gains of approximately 186% and 166% over D-RAN and C-RAN, respectively, and a peak EECO of 1,784,599 m2/MJ, outperforming the other architectures by up to 178%. In rural environments, Cloud C-RAN similarly leads with a maximum EETH of 3.98 Mbit/J and an EECO of 40,205,471 m2/MJ. The findings reaffirm the energy efficiency benefits of virtualization-enabled mobile network architectures. A Monte Carlo simulation framework, based on 1000 independent random spatial trials, confirms these results remain accurate to within 1% under realistic base station and user placement, validating a total network power reduction of approximately 65% for Cloud C-RAN over D-RAN in dense urban conditions and 21% in rural conditions. By confirming these gains hold reliably under realistic deployment conditions, this validated evidence base could be a game changer for Mobile Network Operators (MNOs), as they strive to develop unprecedented user experience for 5G use cases whilst adhering to long-term global sustainability goals.