AI-Native Open RAN for Non-Terrestrial Networks: An Overview
Jikang Deng, Fizza Hassan, Hui Zhou, Saad Al-Ahmadi, Mohamed-Slim Alouini, Daniel B. Da Costa
TL;DR
The paper analyzes the integration of AI-native ORAN with Non-Terrestrial Networks (NTN) to enable dynamic configuration, scalability, and intelligent orchestration in 6G. It identifies NTN-specific DevOps challenges and proposes an orchestrated AI-native ORAN-based NTN framework featuring dynamic fronthaul splits, multi-tier RICs for distributed learning, flexible role switching, and cross-domain SMO, all grounded in digital twins. Three use cases—emergency communication, remote-area coverage, and V2X—demonstrate potential benefits, while future directions address infrastructure sharing, hierarchical ML, mobility management, edge AI, cross-domain AI, and digital twins. The framework aims to unlock resilient, on-demand NTN services through AI-native orchestration, enabling efficient cross-domain coordination among RAN, transport, and core networks. Overall, the work offers a holistic blueprint for deploying AI-driven ORAN-enabled NTN solutions in 6G environments.
Abstract
Non-terrestrial network (NTN) is expected to be a critical component of Sixth Generation (6G) networks, providing ubiquitous services and enhancing the system resilience. However, the high-altitude operation and inherent mobility of NTN introduce significant challenges across the development and operations (DevOps) lifecycle. Apart from that, how to achieve artificial intelligence native (AI-Native) capabilities in NTN for intelligent network management and orchestration remains an important challenge. To solve the challenges above, we propose integrating the Open Radio Access Network (ORAN) with NTN as a promising solution, leveraging its principles of disaggregation, openness, virtualization, and embedded intelligence. Despite extensive technical literature on ORAN and NTN, respectively, there is a lack of a holistic view of the integration of ORAN and NTN architectures, particularly in terms of how intelligent ORAN can address the scalability challenge in NTN management. To address this gap, this paper provides a comprehensive and structured overview of an AI-native ORAN-based NTN framework to support dynamic configuration, scalability, and intelligent orchestration. The paper commences with an in-depth review of the existing literature from leading industry and academic institutions, subsequently providing the necessary background knowledge related to ORAN, NTN, and AI-Native for communication. Furthermore, the paper analyzes the unique DevOps challenges for NTN and proposes the orchestrated AI-Native ORAN-based NTN framework, with a detailed discussion on the key technological enablers within the framework. Finally, this paper presents various use cases and outlines the prospective research directions of this study in detail.