EExApp: GNN-Based Reinforcement Learning for Radio Unit Energy Optimization in 5G O-RAN

2026-02-09

AI Summary

This work proposes EExApp, a deep reinforcement learning-based xApp designed for the O-RAN architecture to reduce energy consumption in 5G radio units (RUs) while ensuring quality of service (QoS). EExApp jointly optimizes RU sleep scheduling and distributed unit (DU) resource slicing through a novel dual-actor dual-critic proximal policy optimization (PPO) framework. It leverages a graph attention network (GAT) to model inter-RU coordination and integrates a Transformer encoder to handle dynamic user loads. By incorporating dual-objective feedback—balancing energy efficiency and QoS—into the actor updates, the method adaptively trades off between these competing goals. Experimental results on a real-world 5G O-RAN testbed demonstrate that EExApp significantly reduces RU energy consumption under strict QoS constraints, outperforming existing approaches.

Technology Category

• Planning, Routing, and Scheduling: Applications
• Search and Optimization: Applications
• Machine Learning: Mixture of Experts (MoE)

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Energy management for devices in mobile Web and WoT environments
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

With over 3.5 million 5G base stations deployed globally, their collective energy consumption (projected to exceed 131 TWh annually) raises significant concerns over both operational costs and environmental impacts. In this paper, we present EExAPP, a deep reinforcement learning (DRL)-based xApp for 5G Open Radio Access Network (O-RAN) that jointly optimizes radio unit (RU) sleep scheduling and distributed unit (DU) resource slicing. EExAPP uses a dual-actor-dual-critic Proximal Policy Optimization (PPO) architecture, with dedicated actor-critic pairs targeting energy efficiency and quality-of-service (QoS) compliance. A transformer-based encoder enables scalable handling of variable user equipment (UE) populations by encoding all-UE observations into fixed-dimensional representations. To coordinate the two optimization objectives, a bipartite Graph Attention Network (GAT) is used to modulate actor updates based on both critic outputs, enabling adaptive tradeoffs between power savings and QoS. We have implemented EExAPP and deployed it on a real-world 5G O-RAN testbed with live traffic, commercial RU and smartphones. Extensive over-the-air experiments and ablation studies confirm that EExAPP significantly outperforms existing methods in reducing the energy consumption of RU while maintaining QoS.

Problem

• energy optimization
• 5G O-RAN
• radio unit
• quality-of-service
• base station energy consumption

Innovation

• Graph Attention Network
• Proximal Policy Optimization
• Transformer Encoder
• O-RAN xApp
• Energy-QoS Tradeoff