Efficient and secure federated dynamic spectrum access for LEO satellite Internet-of-Things

• Main Authors: Shen, Bowen, Goh, Si Qi, Lam, Kwok-Yan, Li, Feng
• Other Authors: College of Computing and Data Science
• Format: Conference or Workshop Item
• Language: English
• Published: 2025
• Subjects: Engineering; Satellite communications; Spectrum allocation; Federated learning; Deep reinforcement learning
• Online Access: https://hdl.handle.net/10356/183695
• Institution: Nanyang Technological University

The explosive growth in demand for Internet of Things (IoT) and significant satellite-to-ground latency have anticipated the transition from 5G to 6G communication. Thus, LEO satellites became a crucial element to the terrestrial network due to their ability to achieve seamless and energy-efficient services even in remote areas. However, the integration of LEO satellites into the broader IoT ecosystem presents significant challenges due to their rapid movement which causes frequent changes in spectrum availability leading to interference. In this paper, we propose a federated dynamic spectrum access (FDSA) scheme. First, bi-level KMeans clustering (Bi-KMeans) is utilized to cluster the terrestrial users (TUs) to improve the training efficiency in federated learning (FL). To address the interference caused by frequent changes in location of TUs, we then design a federated PPO-based algorithm to obtain an optimal strategy to work even under an interactive environment for efficient resource allocation. FL contributes to efficient aggregation while preserving the privacy of users’ data during the training. The simulation results demonstrate the effectiveness and improved performance of the proposed FDSA scheme.