Yang Yang
Shanghai HKU Education Centre, The University of Hong Kong, China
Professor Yang Yang is currently the Executive Director of Shanghai HKU Education Centre, The University of Hong Kong, China. He is also an adjunct professor with the IoT Thrust at HKUST (Guangzhou) and the Chief Scientist of IoT at Terminus Group, China. Before joining HKU, he has held faculty positions at the Chinese University of Hong Kong, China; Brunel University, U.K.; University College London (UCL), U.K.; CAS-SIMIT, China; ShanghaiTech University, China; and HKUST (Guangzhou), China. His research interests include multi-tier computing networks, 5G/6G systems, AIoT technologies and applications, and advanced wireless testbeds. He has published more than 380 papers and filed more than 120 technical patents in these research areas. He is a fellow of the IEEE.
In wireless networks, the inherent conflict between limited computational resources and high-intensity computing workloads makes it extremely challenging to deploy complex tasks based on large AI models in edge devices. Traditional methods rely on aggregating data and transmitting it to cloud platforms for centralized processing; however, this approach is unsustainable as it strains long-distance backhaul transmission and energy-intensive data centers, while also failing to guarantee data security and personal privacy. To address these challenges, we propose the Network Agentic Service Customization Architecture (NASCA) for supporting personalized services. By leveraging the collaborative functions among distributed edge devices, we achieve fast, low-cost, and sustainable edge computing for large AI models. We analyze the engineering issues involved in the coordinated scheduling of multi-element resources—such as communication, sensing, computing, and storage—among edge devices, and verify the feasibility and service efficiency of different parallel processing schemes for large AI models.