Deploy AI at the Edge to Bridge the Information Gap: Efficiency and Security
Overview
Abstract
Artificial intelligence is rapidly extending from cloud centers to the physical world, where billions of IoT devices interact through 5G/6G networks. My research explores how to deploy and optimize AI at the edge to enhance both efficiency and security in distributed systems. By integrating intelligent reflecting surfaces (IRS), graph neural networks (GNNs), and on-device large language models (ODLLMs), we design scalable frameworks for energy-efficient communication, secure spectrum sharing, and privacy-preserving inference. Current projects include LLM-enabled IoT attack detection, IRS-assisted secure transmission, and edge-cloud collaboration for real-time wildfire risk analysis. Our work aims to bridge the gap between data-rich edge environments and intelligent cloud processing, creating systems that are adaptive, reliable, and sustainable. Students are invited to join the CIDER Lab to advance next-generation edge intelligence through interdisciplinary research in wireless communication, AI security, and distributed learning.
Speaker Biography
Dr. Qun Wang is an Assistant Professor in the Department of Computer Science at San Francisco State University. His research focuses on edge intelligence, wireless communication, and secure, energy-efficient AI systems. He leads the CIDER Lab, where his team investigates how large language models (LLMs) and distributed learning can be deployed on heterogeneous edge devices to enhance efficiency, privacy, and real-time decision-making in IoT networks. Dr. Wang has received multiple awards, including the NSF CISE Core Small (NeTS), NSF CRII (NeTS), and the OpenAI Cybersecurity Grant. His current projects span IRS-assisted physical layer security, graph-based spectrum sharing networks, and LLM-driven IoT attack detection. Through his interdisciplinary work, he aims to bridge the gap between cloud intelligence and physical-world applications, enabling sustainable and trustworthy AI systems at the network edge.