通訊系非常榮幸邀請到瑞典克里斯蒂安斯塔德大學（Kristianstad University）的 Dr. Ali Hassan Sodhro (IEEE Senior Member) 蒞校進行學術交流。
Dr. Sodhro 將於 8 月上旬帶來兩場前沿技術專題演講，並特別加開一場「歐盟聯合獎學金與台歐交換生機會」國際交流座談會。本系列活動精彩難得，誠摯歡迎校內外學者、研究人員及同學們踴躍報名參加！
(註：本校車輛進出採車牌辨識收費，前 30 分鐘免費，每小時 30 元，每日收費上限為 100 元，校外開車蒞臨者請於出校前至繳費機或掃碼繳費。)
本活動無須事先繳費，現場座位有限，敬請提早入場。期待您的參與！
聯絡窗口：國立中正大學通訊系 潘仁義教授 <jypan@comm.ccu.edu.tw>
Dr. Ali Hassan Sodhro is an internationally recognized researcher in 6G wireless communications, Artificial Intelligence (AI), Edge Intelligence, Federated Learning, TinyML, Internet of Things (IoT), and Cyber-Physical Systems. He received his Ph.D. from the University of Chinese Academy of Sciences (UCAS), China, and has held research positions at Linköping University, Luleå University of Technology, University Lumière Lyon 2 (France), and the Chinese Academy of Sciences.
His research focuses on AI-native 6G networks, energy-efficient communications, trustworthy edge intelligence, distributed machine learning, and smart healthcare systems. He has authored 100+ scientific publications, received over 5,900 citations (h-index 44), and has been recognized among Stanford University's World's Top 2% Scientists. He serves as an IEEE Senior Member, Associate Editor for several IEEE and IET journals, and actively contributes to European COST Actions on 6G security, cybersecurity, and science diplomacy.
Title: Adaptive Federated Learning for 6G: A Multi-Agent Architecture for 6G Edge Intelligence
The lecture presents an AI-native framework for adaptive federated learning in 6G networks. It introduces the MAAFL-6G device-edge-cloud architecture that coordinates distributed intelligence through hierarchical multi-agent reinforcement learning. The framework jointly optimizes learning accuracy, energy efficiency, latency, trust and security while supporting heterogeneous devices, non-IID data and dynamic wireless environments. Trust-aware aggregation, energy-aware client participation and power-signature based anomaly detection improve resilience against unreliable and malicious clients. Experimental results demonstrate nearly 99.8% learning reliability, reduced latency and lower energy consumption compared with existing approaches. The lecture concludes with future directions including semantic communications, digital twins, intent-driven networking and trustworthy AI for 6G edge intelligence.
Title: Towards Adaptive IoT-5G Authentication for Healthcare Applications: From Developments to Implementation Recommendations
The rapid integration of the Internet of Things (IoT) with 5G networks is transforming smart healthcare by enabling real-time patient monitoring, telemedicine, remote surgery, and intelligent medical services. However, the heterogeneous nature of IoT devices and the dynamic characteristics of 5G networks introduce significant security challenges, particularly in authentication.
This lecture presents a novel Adaptive, Continuous, and Reliable (ACR) authentication framework that leverages physical-layer (PHY) attributes and AI-driven adaptive learning to provide lightweight, secure, and continuous authentication for IoT-5G healthcare environments. The proposed framework improves trust, detects suspicious device behavior at the network edge, and enhances resilience against evolving cyber threats while maintaining low latency and high reliability. The lecture also outlines practical implementation recommendations and discusses future research directions toward trustworthy, intelligent, and secure 6G-enabled healthcare systems.