 |
Present and Future Challenges of HBMMyeongjae Park
Biography
Myeong-Jae Park received the B.S., M.S., and Ph.D. degrees from Seoul National University, Seoul, South Korea, in 2003, 2006, and 2014, respectively. He joined SK hynix, Icheon, Korea in 2014 where he led the design of multiple HBM products including HBM2E, HBM3, HBM3E and HBM4. He is currently in charge of the HBM Design division as the vice president. Prior to joining SK hynix, Dr. Park was with Anapass, Inc., Seoul, Korea as Principal Engineer from 2016 to 2020 where he invented AiPi(Advanced intra-panel interface). His research interests include 3D-DRAM, low-power mixed signal systems and their design methodologies.
Abstract
HBM(High-Bandwidth Memory) is the best-performing memory product that is used in high-end computing systems such as supercomputers and AI accelerators. The recent boom in machine learning is due in part to the development of computing technologies where HBM, the fastest DRAM played a key role in overcoming the performance limitations. In order to achieve high bandwidth of HBM, various technologies and sophisticated design techniques are required. Especially, the complex structure of stacking a single logic die with 4 to 16 DRAM dies also makes its development more challenging. Since SK hynix developed its first HBM product in 2015, bandwidth has increased by 12 times and power efficiency has improved by 3-fold. These performance improvements have been made possible due to various design and process innovations, and given the rapid progress in AI technology, the continuous improvement of high-end memory products like HBM is essential. However, HBM is now facing various technical challenges. Increasing bandwidth, power, and capacity on a small interposer is reaching its technical limits, and issues such as thermal and reliability are becoming more serious. This keynote will present a broad overview of the latest developments in HBM including the current technical challenges from design to devices and packaging technologies, and present future directions for HBM’s developments. |
 |
Towards 6G Communication for a Hyper-Connected WorldJungwon Lee
Biography
Jungwon Lee (M’05–SM’12–F’18) is a Corporate EVP in charge of modem development at Samsung Electronics in South Korea. He received his BS degree from Seoul National University and MS and PhD degrees from Stanford University, all in Electrical Engineering. From 2003 to 2010, he was with Marvell Semiconductor, Inc., where he developed LTE, WiMAX, Wi-Fi, Bluetooth, and HD Radio chips. From 2010 to 2020, he was with Samsung USA, developing cellular modems, connectivity chips, NPUs, and ISPs. From 2013 to 2019, he was also with University of California, San Diego, as an Associate Adjunct Professor. Since 2021, he has been with Samsung Electronics in Korea. He is currently conducting research on 6G and AI. He has co-authored over 150 papers and holds approximately 400 patents in the US and many more worldwide. Dr. Lee became an IEEE Fellow in 2018.
Abstract
As global momentum builds toward 6G, the vision for next-generation connectivity is expanding beyond incremental improvements to a bold reimagination of the communication landscape. This keynote will explore three foundational pillars shaping the 6G era: Frequency Range 3 (FR3), Non-Terrestrial Networks (NTN), and Artificial Intelligence (AI). FR3, which includes the upper mid-band spectrum between 7-15 GHz, is expected to improve spectrum utilization and provide better coverage through advanced modulation and coding schemes that support massive MIMO. This will enable innovative applications such as immersive extended reality, widespread smart city deployments with extensive coverage and capacity, and dense urban broadband. Complementing terrestrial infrastructure, NTNs—leveraging satellites across a range of altitudes from low earth orbit (LEO) to geostationary orbit (GEO) and aerial relays—will provide global coverage, bridging the digital divide and ensuring resilient connectivity in underserved or disrupted areas. AI will play a pivotal role in 6G through AI-native network design across all layers, end-to-end network automation, adaptive resource allocation, and dynamic traffic management. By integrating these advanced paradigms, 6G will transcend the limits of conventional mobile networks and catalyze a hyper-connected, intelligent, and sustainable digital society. This keynote will present a strategic outlook on how these technologies converge, the challenges ahead, and the collaborative innovation required across industry, academia, and policy to shape the 6G future. |
 |
AI Technology Innovation for Digital HealthcareProf. Myung Hoon Sunwoo
Biography
Myung Hoon Sunwoo. (SM’01–F’11) He received his BS degree from Sogang University in 1980, his MS from KAIST in 1982, and his Ph.D. degree from the University of Texas at Austin in the United States in 1990. He worked at ETRI in Daejeon from 1982 to 1985, and worked at Motorola in Austin, Texas, from 1990 to 1992 in Digital Signal Processor Operations. Since 1992, he has worked at the Department of Electrical and Computer Engineering in Ajou University, Suwon, and is currently a professor. In 2012, he served as General Co-chair of the International Symposium on Circuits and Systems (ISCAS) in Seoul and ISCAS 2021 in Daegu, Korea. He will serve as General Cochair of ISCAS 2028 in Twin Cities, Minnesota, USA, and ISCAS 2029 in Songdo, Korea. He served as the Honorary Co-chair of APCCAS 2025, ISICAS 2024, APCCAS 2024, AICAS 2022, and APCCAS 2016. From 2007 to 2012, he served as the honorary ambassador of the Korea Tourism Organization. He served as a distinguished lecturer at the IEEE Association of Circuits and Systems (CASS) (2009-2010), and served as the CASS Board of Governors (BoG) (2011-2016) and the Vice President of the IEEE CASS Conference (2018-2021). He served as the IEEE CAS Presidentelect (2022-2023). As the Vice President of the IEEE CASS Conference, he founded the IEEE International Conference on Artificial Intelligence Circuits and Systems (AICAS) in 2019. He is the Chair of the Steering Committee of AICAS. He served as the President of the Institute of Electronics and Information Engineers (IEIE) Semiconductor Society (2012-2013). He has written numerous papers, held over 130 patents, and received over 60 awards. He has served as a co-editor for several books, including “Selected Topics of Biomedical Circuits and Systems” and “A Short History of Circuits and Systems” (River Publishers, 2021 and 2024). He is the director of the Medical Image-Based Intelligent Diagnosis Solutions (MIIDS) Research Center. His research interests include AI/DL circuits and systems, low-power algorithms and architectures, medical image diagnosis, and deep learning-based channel coding. He is currently the President of IEEE CASS (2024-2025) and is an IEEE Fellow.
Abstract
As the world enters a super-aged society, both medical expenses and the number of patients per capita are rising rapidly. Faced with a shortage of medical personnel, artificial intelligence (AI) must fill this gap. The advent of AI technology has opened a new era of digital healthcare. By leveraging big data, AI can assist in disease diagnosis, and innovations are now spreading across the healthcare field. Leading companies—including Microsoft, Google, Samsung, and OpenAI—are actively involved. In this presentation, we will highlight the latest innovations in AI-based digital healthcare. Next, through joint research with universities, hospitals, and companies, we will introduce the Medical Image-based Intelligence Diagnostic Solution (MIIDS) research center, which brings together new interdisciplinary technologies. The MIIDS center has developed several advanced systems for disease diagnosis, which we will showcase in this talk. |
 |
A New Millennium: Design for AI and AI for DesignAlbert Zeng
Biography
Dr. Albert Zeng received his Ph.D. from Texas A&M University in 2011. He joined Cadence in 2012 as a software engineer and advanced to the position of Sr. Software Engineering Group Director. During his tenure at Cadence, he led multiple RD teams, including Voltus, Celsius, and Reality. Dr. Zeng is dedicated to pioneering groundbreaking innovations, current-edge technology and transformative product developments. Over the year, his teams have delivered game-changing solutions like Voltus XP, Voltus XM, Voltus InsightAI, Celsius Studio, Tenacity Stress Solver, and the Reality Digital Twin Platform. Currently, he leads advanced technology development at HSA in Cadence, focusing on GPU acceleration and AI. Dr. Zeng has been recognized with the EE Awards Asia – Innovation R&D Award in 2024 and Cadence Achievement Award in 2016.
Abstract
A New Millennium: Design for AI and AI for Design The AI revolution demands a radical reimagining of semiconductor design. As AI training compute explodes 7x annually, we’ve reached the physical limits of traditional 2D scaling. Enter 3D Integrated Circuits – the next frontier that promises to unlock unprecedented performance density, but at what cost? The 3DIC Frontier: Imagine stacking multiple silicon dies vertically, creating computing towers with breathtaking performance. Yet this vertical paradise becomes a thermal nightmare – power densities that would melt conventional designs, heat trapped in silicon labyrinths with no escape, and mechanical stresses that can literally tear chips apart. Traditional design flows crumble under this complexity, requiring months to analyze what must be decided in hours. The Multi-Physics Multi-Scale Storm: 3DIC design isn’t just electrical anymore. It’s a perfect storm where thermal physics meets mechanical engineering meets electromagnetic field theory – all interacting simultaneously across nanometer to system scales. Hotspots migrate through vertical structures. Warpage cascades across multiple die levels. Signal integrity battles thermal noise in ways never before encountered. 3DIC design also unleashes a complexity tsunami spanning seven orders of magnitude – from nanometer transistor interactions to millimeter package dynamics, which requires a brand new multi-scale analysis framework. The AI-Powered Breakthrough: But what if AI could solve what AI created? Revolutionary breakthroughs in GPU-accelerated simulation and AI-driven design optimization are transforming impossible problems into real-time solutions. Thermal management becomes predictive. Stress analysis becomes proactive. Multi-physics optimization becomes automatic. The New Reality: We’re witnessing the birth of a new design paradigm where AI doesn’t just drive demand for better chips – it fundamentally transforms how we create them. Welcome to the millennium where design challenges become design opportunities, and the impossible becomes inevitable. Join us for a journey into the future of semiconductor design, where physics meets AI and complexity becomes creativity. |