Original title: Direct Satellite-to-Device Communications: From Cooperative Task Offloading to Non-Cooperative Access Monitoring
Sai Huang, Wanli Ni, Ke Lv, Pengcheng Zhang +4 · September 2026
Connecting phones straight to satellites helps bring network service to remote places. However, satellites move quickly, their signals fade, and their computing power is tight. At the same time, unauthorized devices can jump on these networks and threaten spectrum security. To fix this, researchers built a two-part satellite system. For approved tasks, the system estimates channel shifts and uses smart decision software to decide when to offload computing work. For unapproved signals, it uses AI pattern-spotting models to catch and identify unauthorized broadcasts. They found this approach reduces average delay by up to 225% compared to static rules. It spots satellite signals with an average 90.5% detection probability, and its signal-identifying tool boosts accuracy by 9.4% in low-signal conditions over older methods.
Imagine a mobile library truck that serves remote villages. The librarian must decide which books to hand out immediately and which to borrow from a central warehouse, but the road is bumpy and the truck's engine is weak. Worse, strangers sometimes impersonate villagers and steal books. Here, the truck is the satellite, the books are computing tasks, and the central warehouse is the ground network. The system decides the best time to borrow resources, and also learns to spot fake villagers by the way they ask for books.
This system brings edge computing to places without cell towers, so remote areas get faster data processing. It also catches unauthorized satellite users who steal bandwidth, protecting the spectrum for legitimate access. By doing both with the same technology, it makes 6G satellite networks more practical and secure.
The abstract doesn't specify limitations, but it reports simulation results rather than real-world deployment. The 90.5% detection probability means nearly one in ten unauthorized signals might be missed. The latency reduction of up to 225% compares only to static association policies; performance relative to other dynamic methods is not shown.
Ask anything about this paper. Answers come from its abstract, and say so when the abstract doesn't cover your question.
Direct Satellite-to-Device Communications: From Cooperative Task Offloading to Non-Cooperative Access Monitoring — Sai Huang, Wanli Ni, Ke Lv et al.
Abstract licence: CC0
Explanation generated by Synthesis from the abstract — it is not the original paper. This is a preprint, so it hasn't necessarily been peer reviewed.