[Paper Review] Artificial Intelligence for Satellite Communication and Non-Terrestrial Networks: A Survey

G. Fontanesi, Flor G. Ortiz+19 more|arXiv (Cornell University)|Apr 25, 2023

Satellite Communication Systems25 citations

TL;DR

This survey maps AI/ML applications in Satellite Communication (SATCOM) and Non-Terrestrial Networks (NTN), detailing use cases, architectures, hardware, challenges, and future directions for onboard and on-ground AI deployments.

ABSTRACT

This paper surveys the application and development of Artificial Intelligence (AI) in Satellite Communication (SatCom) and Non-Terrestrial Networks (NTN). We first present a comprehensive list of use cases, the relative challenges and the main AI tools capable of addressing those challenges. For each use case, we present the main motivation, a system description, the available non-AI solutions and the potential benefits and available works using AI. We also discuss the pros and cons of an on-board and on-ground AI-based architecture, and we revise the current commercial and research activities relevant to this topic. Next, we describe the state-of-the-art hardware solutions for developing ML in real satellite systems. Finally, we discuss the long-term developments of AI in the SatCom and NTN sectors and potential research directions. This paper provides a comprehensive and up-to-date overview of the opportunities and challenges offered by AI to improve the performance and efficiency of NTNs.

Motivation & Objective

• Identify and classify AI/ML use cases in SATCOM and NTN across space, ground, and user segments.
• Analyze challenges and benefits of AI-enabled solutions for SATCOM system optimization and automation.
• Review current hardware platforms and architectures supporting AI/ML in satellite systems.
• Evaluate on-board vs. on-ground AI deployment trade-offs and their impact on OPEX and latency.
• Highlight future research directions and potential standards for AI in SATCOM/NTN.

Proposed method

• Systematic literature review of journal and conference papers from IEEE, MDPI, Elsevier, and Wiley.
• OSI-oriented classification of SATCOM/NTN use cases and alignment with traditional network layers.
• Description of conventional tools, ML techniques, and training/inference paradigms for each use case.
• Assessment of AI hardware ecosystems and comparison of commercial AI chipsets for SATCOM applicability.
• Discussion of future deployment challenges, trends, and research directions.

Experimental results

Research questions

• RQ1 What are the major SATCOM/NTN use cases where AI/ML can add value?
• RQ2 What are the challenges and trade-offs of onboard versus on-ground AI deployments in SATCOM/NTN?
• RQ3 Which ML techniques (supervised, unsupervised, RL, DL) are most suitable for different SATCOM use cases?
• RQ4 What are the current hardware options for implementing AI/ML in SATCOM, and how do they compare?
• RQ5 What future directions and research gaps exist for AI in SATCOM and NTN?

Key findings

• The paper provides an extensive classification of SATCOM/NTN use cases and associates ML techniques with each case.
• It discusses onboard and on-ground AI architectures and their respective advantages and drawbacks.
• It reviews current hardware solutions and commercial AI chipsets relevant to SATCOM applications.
• It highlights how digital payloads and centralized/Cloud-like architectures enable data-driven optimization and automation in SATCOM/NTN.
• The survey identifies challenges and proposes future directions, including integration with NGSO, HAP/LAP, and space–air–ground networks.
• The methodology relies on a systematic literature review and OSI-oriented organization of use cases and solutions.