Wang, T, Fang, K, Tian, J, Feng, H, Al Dabel, MM, Bashir, AK ORCID: https://orcid.org/0000-0001-7595-2522 and Wang, W
(2025)
AI-Backed Network Security for Connecting Air, Space, and Ground.
IEEE Wireless Communications, 32 (3).
pp. 80-87.
ISSN 1536-1284
|
Accepted Version
Available under License Creative Commons Attribution. Download (1MB) | Preview |
Abstract
With the development of air-space-ground integrated communication networks, security concerns are increasingly prominent. This article explores the application of artificial intelligence (AI) technologies in enhancing the security of these networks from three key perspectives: signal localization optimization, federated learning (FL) frameworks, and attack detection methods. First, AI optimizes signal propagation paths through deep learning and machine learning algorithms, enhancing localization accuracy, communication quality, and system reliability. Second, FL, as a distributed machine learning method, protects data privacy while achieving collaborative optimization of distributed nodes through model parameter sharing and updates. Finally, AI-driven attack detection methods use anomaly detection and behavior analysis to identify and respond to network threats in real time, improving system protection capabilities. Additionally, this article designs a robust FL framework, called mobile optimization-based federated learning (MOBFL), to simulate attack detection in high-speed mobile scenarios in non-terrestrial networks. Experimental results demonstrate MOB-FL's excellent attack detection capabilities.
Impact and Reach
Statistics
Additional statistics for this dataset are available via IRStats2.