Space-O-RAN: Enabling Intelligent, Open, and Interoperable Non Terrestrial Networks in 6G

2025-02-21

AI Summary

Current non-terrestrial networks (NTNs) are deployed in isolation, suffering from poor scalability, high operational costs, and challenges in cross-domain coordination. To address these limitations, this paper proposes a 6G-oriented integrated space-air-ground intelligent network architecture. We introduce a novel hierarchical closed-loop control framework and a dynamic link-interface mapping mechanism, and— for the first time—extend the O-RAN paradigm to NTN by designing distributed Space-RICs, satellite-native applications (sApps), device-native applications (dApps), and a digital-twin-driven closed-loop management and control system. Leveraging AI-enabled real-time satellite resource scheduling and cross-domain task offloading, simulations demonstrate that intra-cluster coordination latency meets signaling constraints, while offloading non-real-time tasks to terrestrial infrastructure significantly enhances system scalability. This work establishes a new interoperable paradigm for 6G NTN, characterized by low latency, high elasticity, and cost efficiency.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Application Domains: Internet of Things, Sensor Networks & Smart Cities
• Intelligent Robots: Multi-Robot Systems

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Web applications in cross-disciplinary domains and verticals such as mixed reality, smart cities, and digital health
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Non-terrestrial networks (NTNs) are essential for ubiquitous connectivity, providing coverage in remote and underserved areas. However, since NTNs are currently operated independently, they face challenges such as isolation, limited scalability, and high operational costs. Integrating satellite constellations with terrestrial networks offers a way to address these limitations while enabling adaptive and cost-efficient connectivity through the application of Artificial Intelligence (AI) models. This paper introduces Space-O-RAN, a framework that extends Open Radio Access Network (RAN) principles to NTNs. It employs hierarchical closed-loop control with distributed Space RAN Intelligent Controllers (Space-RICs) to dynamically manage and optimize operations across both domains. To enable adaptive resource allocation and network orchestration, the proposed architecture integrates real-time satellite optimization and control with AI-driven management and digital twin (DT) modeling. It incorporates distributed Space Applications (sApps) and dApps to ensure robust performance in in highly dynamic orbital environments. A core feature is dynamic link-interface mapping, which allows network functions to adapt to specific application requirements and changing link conditions using all physical links on the satellite. Simulation results evaluate its feasibility by analyzing latency constraints across different NTN link types, demonstrating that intra-cluster coordination operates within viable signaling delay bounds, while offloading non-real-time tasks to ground infrastructure enhances scalability toward sixth-generation (6G) networks.

Problem

• Integrating satellite and terrestrial networks for 6G.
• Overcoming isolation and high costs in NTNs.
• AI-driven adaptive resource allocation in NTNs.

Innovation

• Open RAN principles extension
• AI-driven network orchestration
• Dynamic link-interface mapping