A Review of Open-RAN Intelligence: Opportunities, Challenges, Real-life Applications and Impacts

What is it about?

This article presents one of the first systematic and comprehensive reviews of intelligence in Open Radio Access Networks (Open-RAN). As mobile networks evolve toward 5G, Beyond-5G, and 6G, operators are increasingly adopting Open-RAN for its flexibility, interoperability, and ability to integrate AI-driven optimisation. However, the field lacks a unified understanding of how intelligence; spanning AI, ML, DL, RL, FL, probabilistic models, and Generative AI, that can be designed, deployed, and evaluated within Open-RAN systems. Our work fills this gap by analysing 21 rigorously selected publications, mapping how intelligence is implemented across the 5G Core, RAN, and air interface, and identifying the advantages, limitations, and open challenges. We review real-world industrial deployments, the role of the RAN Intelligent Controller (RIC), and practical use-cases including resource allocation, anomaly detection, network slicing, mobility optimisation, and Massive MIMO enhancements. We also examine the societal, environmental, and economic impacts of AI-enabled Open-RAN, highlighting how intelligent automation can reduce operational cost, improve energy efficiency, and support emerging domains such as industrial IoT, smart cities, and autonomous systems. Finally, the article proposes a roadmap of intelligent algorithms tailored for Open-RAN and identifies future directions including standardisation, trustworthy AI, digital twins, and AI-native 6G architectures. This review serves as a foundational reference for researchers, engineers, and policymakers aiming to advance intelligent, interoperable, and sustainable mobile networks.

Featured Image

Why is it important?

This work is important because modern mobile networks face dynamic, unpredictable, and complex environments that traditional architectures cannot handle efficiently. Open-RAN introduces intelligent, adaptable, and learning-based capabilities that enhance resource efficiency, optimise communication networks, reduce operational costs, and support autonomous adaptation across diverse scenarios. The review addresses the absence of a systematic investigation into the integration of intelligence within Open-RAN and consolidates fragmented studies into a unified understanding of the RIC, intelligent algorithms, potential use-cases, real-life industrial applications, and the societal, environmental, and economic impact of deploying AI and ML in the network. It provides clarity on how open interfaces, virtualisation, and intelligent decision-making improve performance, enable flexible deployments, support multi-vendor interoperability, and allow continuous data-driven optimisation across the core, RAN, and air interface. By systematically analysing advantages, challenges, limitations, and opportunities, the work fills a critical gap in the literature and supports the evolution of intelligent and flexible architectures for 5G, B5G, and 6G.

Perspectives