Large Language Models (LLMs) in Space‑Air‑Ground Integrated Networks (SAGINs) by Prof Arumugam Nallanathan

Abstract

Space‑Air‑Ground Integrated Networks (SAGINs) combine satellites, aerial platforms (such as UAVs and high‑altitude platforms), and terrestrial networks to provide ubiquitous, high‑speed global coverage. UAVs have been widely deployed for reliable and energy‑efficient data collection in Internet of Things (IoT) applications. Recently, Large Language Models (LLMs) have been used in SAGINs to enable more intelligent, adaptive, and autonomous networks, particularly for 6G and beyond.

In this talk, inspired by the strong generalization and reasoning capabilities of LLMs, an LLM‑based channel prediction framework (CPLLM) is presented to forecast future channel state information (CSI) for LEO satellites using historical CSI data. An LLM‑empowered critic‑regularized decision transformer (LLM‑CRDT) framework for learning effective UAV control policies is also introduced.

Biography

Arumugam Nallanathan is Professor of Wireless Communications and the founding head of the Communication Systems Research (CSR) group at the School of Electronic Engineering and Computer Science, Queen Mary University of London, since September 2017. He previously served as Professor of Wireless Communications at King’s College London (2013–2017) and as Assistant Professor at the National University of Singapore (2000–2007).

His research interests include 6G wireless networks and the Internet of Things (IoT). He has published nearly 900 technical papers, received close to 40,000 citations with an H‑index of 100, and won multiple Best Paper Awards, including the IEEE Communications Society Leonard G. Abraham Prize (2022). He is an IEEE Fellow and IEEE Distinguished Lecturer.