提出AI原生开放无线网络框架,解决天基网络的运维难题。
AI-Native Open RAN for Non-Terrestrial Networks: An Overview

- 将AI原生能力融入开放无线网络,实现智能协同控制
- 构建面向天基网络的统一架构,支持高动态环境下的稳定运行
- 适合6G研发人员与空天地一体化网络设计者参考
非地面网络(NTN)被视为第六代移动通信(6G)的关键组成部分,可实现无处不在的服务并增强网络弹性。然而,其固有的高动态移动性与高空作业特性给开发与运维(DevOps)全生命周期带来重大挑战。将NTN与开放无线接入网(ORAN)结合是一种有前景的解决方案,因为ORAN具备解耦、开放性、虚拟化和嵌入式智能等优势。尽管已有大量关于ORAN和NTN的研究,但针对基于ORAN的NTN体系架构的系统性综述仍显不足,尤其缺乏对如何有效应对NTN现有挑战的深入探讨。同时,虽然人工智能原生(AI-Native)能力在提升网络智能控制与优化方面具有潜力,但在NTN中的实际应用尚未充分研究。因此,本文提供了关于AI-Native ORAN用于NTN的全面且结构化的综述,首先回顾相关文献并介绍ORAN、NTN及通信领域中AI-Native的基础背景;随后分析了NTN的DevOps挑战,并提出一个协同的AI-Native ORAN-based NTN框架,讨论其关键技术使能器;最后展示代表性应用场景,并展望未来研究方向。
原文摘要 · Abstract (English)
Non-terrestrial network (NTN) is envisioned as a critical component of Sixth Generation (6G) networks by enabling ubiquitous services and enhancing network resilience. However, the inherent mobility and high-altitude operation of NTN pose significant challenges throughout the development and operations (DevOps) lifecycle. To address these challenges, integrating NTNs with the Open Radio Access Network (ORAN) is a promising approach, since ORAN can offer disaggregation, openness, virtualization, and embedded intelligence. Despite extensive literature on ORAN and NTN, a holistic view of ORAN-based NTN frameworks is still lacking, particularly regarding how ORAN can effectively address the existing challenges of NTN. Furthermore, although artificial intelligence native (AI-Native) capabilities have the potential to enhance intelligence network control and optimization, their practical realization in NTNs has not yet been sufficiently investigated. Therefore, in this paper, we provide a comprehensive and structured overview of AI-Native ORAN for NTN. This paper commences with an in-depth review of the existing literature and subsequently introduces the necessary background about ORAN, NTN, and AI-Native for communication. After analyzing the DevOps challenges for NTN, we propose the orchestrated AI-Native ORAN-based NTN framework and discuss its key technological enablers. Finally, we present the representative use cases and outline the prospective future research directions of this study.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品;中文卡片由大模型生成,请以原文为准。