The rapid evolution of radio access networks(RANs)highlights the pressing need for sustainable and programmable resource management strategies.This paper introduces the energy saving RAN application(ES-rApp),an AI-dri...The rapid evolution of radio access networks(RANs)highlights the pressing need for sustainable and programmable resource management strategies.This paper introduces the energy saving RAN application(ES-rApp),an AI-driven solution designed for intelligent resource orchestration within the open RAN(O-RAN)architecture.ES-rApp integrates traffic-aware prediction models and closed-loop automation to dynamically optimize network operations,enabling adaptive cell deactivation,intelligent transmit-power adjustment,and proactive resource allocation based on predicted traffic demands while preserving service quality.Unlike prior studies that rely primarily on simulations or small-scale prototypes,we present a deployable and O-RAN-compliant implementation validated on a real testbed.Experimental evaluations under diverse traffic profiles demonstrate that ES-rApp achieves up to 19.5%energy savings without degrading quality of service(QoS).These results provide a real-world evidence of AI-native energy optimization in live RAN environments,establishing ES-rApp as a scalable and practical solution for green RAN management.This work contributes a concrete pathway toward transforming conventional RANs into sustainable,intelligent infrastructures that advance both operational efficiency and environmental responsibility in next-generation wireless networks.展开更多
基金funded in part by Start-up Funding for the 2025 National High-End Talent Frontier Project(01)with Grant 70012501A01in part by the National Natural Science Foundation of China under Grant 62301328+1 种基金by the National Research Foundation,Singapore,and Infocomm Media Development Authority under its Future Communications Research&Development Programmein part by the SNS JU project 6G-GOALS under the EU's Horizon program Grant Agreement 101139232.
文摘The rapid evolution of radio access networks(RANs)highlights the pressing need for sustainable and programmable resource management strategies.This paper introduces the energy saving RAN application(ES-rApp),an AI-driven solution designed for intelligent resource orchestration within the open RAN(O-RAN)architecture.ES-rApp integrates traffic-aware prediction models and closed-loop automation to dynamically optimize network operations,enabling adaptive cell deactivation,intelligent transmit-power adjustment,and proactive resource allocation based on predicted traffic demands while preserving service quality.Unlike prior studies that rely primarily on simulations or small-scale prototypes,we present a deployable and O-RAN-compliant implementation validated on a real testbed.Experimental evaluations under diverse traffic profiles demonstrate that ES-rApp achieves up to 19.5%energy savings without degrading quality of service(QoS).These results provide a real-world evidence of AI-native energy optimization in live RAN environments,establishing ES-rApp as a scalable and practical solution for green RAN management.This work contributes a concrete pathway toward transforming conventional RANs into sustainable,intelligent infrastructures that advance both operational efficiency and environmental responsibility in next-generation wireless networks.
