IoT Day - Monday, June 3rd

Welcome IoT Day : 08:50-09:00

Introductory Remarks

Chenren Xu and Mi Zhang

Keynote 1 : 09:00-10:00

Distinguished Keynote

The Convergence of Computing and Telecommunications Systems

Speaker: Prof. David Farber Keio University Keio University Cyber Civilization Research Center (Co-Director), Japan	Moderator: Prof. Chenren Xu Peking University, China

Abstract: In 18 March 1977 in AAAS SCIENCE magazine, David Farber and Paul Baran wrote about “The Convergence of Computing and Telecommunications Systems”. We believed that the two separate fields would, in the future, merge to revolutionize the future of communications and society. As we have seen, that is indeed what has happened. This convergence is still continuing as the migration of our communications toward the Internet mechanisms as well as the continuing improvements in mobile technology and the continuing improvement in integrated semiconductor has dramatically produced revolutionary changes in how we communicate with each other, how we control our environment, and how we do business. This talk will track these changes and try to predict future direction in both technology and uses, for good and maybe not so good in our civilization.

Bio: Professor Farber is the Guest Professor of Keio University and Co-Director of Keio University Cyber Civilization Research Center. He is also a Trustee Emeritus of the Board of Trustees of Stevens Institute of Technology. Prior to this, he was the Distinguished Career Professor of Computer Science and Public Policy at the School of Computer Science at Carnegie Mellon University. From 2000 to 2001, he served as Chief Technologist for the Federal Communications Commission. Before joining FCC, he served on the U.S. Presidential Advisory Committee on Information Technology. In 2003, he retired from the University of Pennsylvania where he held the Alfred Fitler Moore Chair of Telecommunications with appointments in the Engineering School and the Wharton School. His background includes positions at Bell Labs, the Rand Corporation, Xerox Data Systems, the University of California at Irvine and the University of Delaware. He was selected for 2020 EPIC Lifetime Achievement Award to recognize his longtime effort in defending privacy and Internet security and was elected as the AAAS Fellow by the Council of the American Association for the Advancement of Science in 2018. He was awarded the SIGCOMM Award for life-long contributions to communications and the City of Philadelphia’s John Scott award for Contributions to Humanity as well as an Honorary Doctorate from Stevens Institute of Technology. He has also been recognized as a Pioneer of the Internet Society Hall of Fame. He is a Fellow of ACM and IEEE.

Keynote 2 : 10:30-11:30

Wearable Cognitive Assistance: From Mark Weiser to ChatGPT

Speaker: Prof. Mahadev Satyanarayanan Computer Science Deartment Carnegie Mellon University, USA	Moderator: Prof. Rajesh Krishna BALAN Singapore Management University

Abstract: Real-time augmentation of human cognition has the potential to become one of those rare capabilities that “weave themselves into the fabric of everyday life until they are indistinguishable from it,” in the immortal words of Mark Weiser. Wearable Cognitive Assistance (WCA) achieves this capability via the convergence of AI, Edge Computing, wireless networks, and wearable devices. In this talk, I will describe the evolution of WCA through three phases: vision, reality, and future challenges. Since 2014, we have built close to 20 applications of this genre. Most recently, we have been able to leverage cloud-based LLMs such as ChatGPT for lowering the barrier to creation of WCA applications. In this talk, I will share the lessons that we have learned from this decade-long experience. The possibility of improving human productivity in many domains through WCA is tantalizing, but some challenges remain.

Bio: Satya’s multi-decade research career has focused on the challenges of performance, scalability, availability and trust in information systems that reach from the cloud to the mobile edge of the Internet. In the course of this work, he has pioneered many advances in distributed systems, mobile computing, pervasive computing, and the Internet of Things (IoT). Most recently, he has been viewed as “The Father of Edge Computing” for his seminal 2009 paper, and his pioneering contributions to the foundations of edge computing. Satya is the Jaime Carbonell University Professor of Computer Science at Carnegie Mellon University. He received the PhD in Computer Science from Carnegie Mellon, after Bachelor’s and Master’s degrees from the Indian Institute of Technology, Madras. He is a Fellow of the ACM and the IEEE.

Keynote 3 : 11:30-12:30

Towards Autonomous, Precision, and Sustainable Farming

Speaker: Prof. Jie Liu Research Institute for Artificial Intelligence, Harbin Institute of Technology, China National Key Laboratory (Director)	Moderator: Prof. Longfei Shangguan University of Pittsburgh

Abstract: In this talk, we look at food security for human society and study how artificial intelligence and IoT can help with improving farm productivity and efficiency. Using information from remote sensing, weather forecasting and ground sensors, combined with advanced algorithms, we aim at making intelligent decisions based on real-time data as well as human expertise. We also designed unmanned farm vehicles to reduce farming labor intensity and to improve operation precision. We will discuss challenges and future roads towards industrializing this old form of economy.

Bio: Jie Liu is the Director of National Key Lab on Smart Farming Technologies and Systems in China. He worked at Xerox PARC and Microsoft from 2001 to 2019. He was a Principal Research Manager at Microsoft Research and a partner of the company. His research interests are AI for IoT, mobile and embedded systems, and energy efficient computing. He was the Steering Committee Chair for CPS-IoT Week, IPSN, and SenSys. He received IEEE TCCPS Distinguished Leadership Award and 7 Best Paper Awards from top conferences. He is an IEEE Fellow and an ACM Distinguished Scientist.

Lunch Break : 12:30-14:00

Keynote 4 : 14:00-15:00

Enabling Ubiquitous Sensing for Smart Health Applications

Speaker: Guoliang Xing Professor, Director of AIoT Lab Department of Information Engineering, The Chinese University of Hong Kong	Moderator: Prof. Yunxin Liu Tsinghua University

Abstract: A key enabling technology for next-generation smart health applications is ubiquitous sensing systems that monitor people’s health and their surrounding environments anytime, anywhere, without compromising privacy. In this talk, we present our recent work addressing key challenges in enabling ubiquitous sensing for smart health.

Wi-Fi-based sensing is promising due to the prevalence of off-the-shelf Wi-Fi devices. However, existing approaches rely heavily on channel state information (CSI), which is limited to only 6% of operational Wi-Fi devices. To overcome this, we introduce BeamSense, a novel system using compressed beamforming reports (CBR) available in operational Wi-Fi devices. BeamSense employs a multi-path estimation algorithm to map bidirectional CBR to a multi-path channel, enhancing the generalizability of CSI-based algorithms while maintaining high accuracy.

Sensing in low-light and dark environments holds significant potential in health domains but faces challenges like excessive noise and low resolution. We present Mozart, which uses off-the-shelf Time-of-Flight (ToF) depth cameras to generate high-resolution, texture-rich maps in dark scenarios by manipulating ToF measurements’ phase components. Mozart uses an unsupervised learning approach to produce high-resolution maps, offering real-time, low-cost, and high-performance sensing in the dark.

We also discuss ADMarker, a system for detecting Alzheimer’s Disease (AD) digital biomarkers in natural living environments. ADMarker integrates multi-modal sensing technologies and a three-stage federated learning architecture to address challenges like limited data labels and heterogeneity. Clinical trials show ADMarker accurately detects digital biomarkers, enabling early AD identification and providing a longitudinal tracking platform for patients and caregivers.

Lastly, we discuss several technologies that have been successfully commercialized, including a smart home system that uses novel sensors and large language models (LLM) to manage health risks for the elderly.

Bio: Guoliang Xing is currently a Professor and the director of the AIoT Lab in the Department of Information Engineering at The Chinese University of Hong Kong. Previously, he was a faculty member at Michigan State University between 2008 and 2017. His work received five Best Paper Awards and seven Best Paper Finalists/Runner-Ups at leading international conferences, including MobiSys, MobiCom, ICNP, IPSN, IoTDI, and SenSys. His research interests include Internet of Things (IoT), Autonomous Driving, Smart Health, Edge Computing, and wireless networking. Several mobile technologies developed in his lab were successfully transferred to the industry. He received the U.S. NSF CAREER Award in 2010 and the Withrow Distinguished Faculty Award from Michigan State University in 2014.  He is a Fellow of IEEE.

Keynote 5 : 15:00-16:00

Privacy on the Edge: From Federated Computations to Personal Agents

Speaker: Dr. Marco Gruteser Research Scientist, Google	Moderator: Prof. Zhenjiang Li City University of Hong Kong

Abstract: Intelligent systems that tackle society’s biggest challenges increasingly rely on personal data originating from the edge. Computing at the edge, via federated learning and analytics, also presents opportunities to enhance privacy while deriving insights from such decentralized data. In this talk, I will share our experience with architectures that embody privacy preserving principles and mechanisms such as early data minimization, ephemeral aggregation and differential privacy to protect sensitive information during data collection and processing. I will further describe how such systems scale with confidential federated computations and open questions that remain as we increasingly rely on large models and a multitude of agents.

Bio: Marco Gruteser is a Research Scientist at Google leading teams focused on private analytics and learning technology for location data. Previously he served as the Peter D. Cherasia Faculty Scholar and Professor of ECE at Rutgers University’s WINLAB, leading research in mobile computing with an emphasis on connected vehicles. He has also chaired the Association of Computing Machinery’s Special Interest Group on Mobile Computing (ACM SIGMOBILE) as well as served as general and program chair of its MobiCom and MobiSys conferences, respectively. He has held research and visiting positions at the IBM T. J. Watson Research Center and Carnegie Mellon University. His recognitions include an NSF CAREER award, a Rutgers Board of Trustees Research Fellowship for Scholarly Excellence, a Rutgers Outstanding Engineering Faculty Award, ten keynotes, as well as multiple award paper awards at ACM MobiCom and ACM MobiSys. He received his PhD degree from the University of Colorado in 2004 and is an ACM Distinguished Scientist.

Coffee Break : 16:00-16:30

Panel on IoT + Generative AI / LLMs : 16:30-17:20

Panelists

Speaker: Mahadev Satyanarayanan Carnegie Mellon University	Speaker: Jie Liu Harbin Institute of Technology

Speaker: Guoliang Xing The Chinese University of Hong Kong	Speaker: Yunxin Liu Tsinghua University

Speaker: Marco Gruteser Google

Moderator

Moderator: Prof. Mi Zhang Ohio State University, USA

Closing Remarks : 17:20-17:30

Chenren Xu and Mi Zhang