Keynote Speeches
Steven Latré, IMEC, Belgium
How machine learning is (not) transforming communication networks: status and challenges
The integration of Artificial Intelligence (AI) into communication networks holds immense promise, particularly in the realms of control and management. This keynote will explore this potential through several success stories such as joint communication and sensing, MAC control, and dynamic spectrum management. We will illustrate how AI has already enhanced network efficiency, adaptability, and performance. However, this journey is not without significant challenges. The lack of high-quality data, the substantial power consumption associated with AI models, and the pressing need for efficient embedded AI solutions present considerable hurdles. This talk will provide a balanced perspective, highlighting both the achievements and the ongoing challenges in the deployment of AI in communication networks, offering insights into the future trajectory of this evolving field.
Bio:
Prof. Steven Latré, is VP R&D AI and algorithms and is leading the algorithmic research at imec, with a particular focus on machine learning. imec is the worldwide leading R&D hub for nano-electronics and digital technologies. He leads an interdisciplinary team of 1,000 researchers at the intersection of AI accelerators, advanced sensors and machine learning. This results in application domains such as compute and connectivity systems, life science technologies, automotive, greentech and smart industries. Next to this, he is also a part-time professor at the University of Antwerp. He received a Master of Science degree in computer science from Ghent University, Belgium and a Ph.D. in Computer Science Engineering from the same university. Before joining imec, he was leading the IDLab research group at the University of Antwerp, a 100+ research group focusing on wireless networks and machine learning. He is author or co-author of more than 150 papers published in international journals or in the proceedings of international conferences. He is the recipient of the IEEE COMSOC award for best PhD in network and service management 2012, the IEEE COMSOC Young Professional award 2015 and is a recipient of the Laureate award of the Academy of Sciences, Belgium.
Google Scholar Profile: https://scholar.google.com/citations?user=gY1AtYAAAAAJ&hl=en&oi=ao
Sinem Coleri, Koç Üniversitesi, Türkiye
AI Based Ultra-Reliable Wireless Networked Control Systems in 6G
Unlike previous generations of networks, which were primarily designed to meet the requirements of human communications, 5G networks enable the collection of data from the machines as well. As per the Ericsson Mobility Report, the estimated number of connected devices is projected to reach 26 billion in 2026. Looking forward to 6G systems, the focus shifts towards leveraging this data for a new spectrum of control applications, including extended reality, remote surgery and autonomous vehicle platoons. Designing communication systems for these control applications poses unique challenges. It requires meeting stringent requirements for delay and reliability, addressing the semantics of control systems and ensuring robust resource management. In the first part of this talk, we delve into ultra-reliable channel modeling and communication techniques based on extreme value theory and generative artificial intelligence (AI). Generative AI enables predicting the channel parameters with higher accuracy while incorporating various system inputs and providing adaptivity to dynamic scenarios. In the second part of the talk, we explore the benefits of employing optimization theory based, explainable and safe AI in radio resource management for the joint design of control and communication systems in 6G networks. These approaches offer a systematic methodology to enhance robustness and interpret decisions made by black-box AI models.
Bio:
Sinem Coleri is a Professor and the Chair of the Department of Electrical and Electronics Engineering at Koc University. She is also the founding director of Wireless Networks Laboratory (WNL) and director of Ford Otosan Automotive Technologies Laboratory. Sinem Coleri received the BS degree in electrical and electronics engineering from Bilkent University in 2000, the M.S. and Ph.D. degrees in electrical engineering and computer sciences from University of California Berkeley in 2002 and 2005. She worked as a research scientist in Wireless Sensor Networks Berkeley Lab under sponsorship of Pirelli and Telecom Italia from 2006 to 2009. Since September 2009, she has been a faculty member in the department of Electrical and Electronics Engineering at Koc University. Her research interests are in 6G wireless communications and networking, machine learning for wireless networks, machine-to-machine communications, wireless networked control systems and vehicular networks. Dr. Coleri has more than 150 publications with citations over 11000 ( Google scholar profile ). She has received numerous awards and recognitions, including N2Women: Stars in Computer Networking and Communications in 2022; TUBITAK (The Scientific and Technological Research Council of Turkey) Incentive Award and IEEE Vehicular Technology Society Neal Shepherd Memorial Best Propagation Paper Award in 2020; Outstanding Achievement Award by Higher Education Council in 2018; and Turkish Academy of Sciences Distinguished Young Scientist (TUBA-GEBIP) Award in 2015. Dr. Coleri currently holds the position of Editor-in-Chief at the IEEE Open Journal of the Communications Society. Dr. Coleri is an IEEE Fellow and IEEE ComSoc Distinguished Lecturer.
Google Scholar Profile: https://scholar.google.com/citations?user=XhV0a1UAAAAJ&hl=en&oi=ao
Sandra Cespedes, Concordia Université, Canada
From Earth's Orbit to the Cosmos: A Vision for a Multi-Planetary Space IoT System
Satellite communications have re-emerged as a cost-effective and globally accessible solution to expand terrestrial Internet of Things services, enabled by the advent of Low-Earth Orbit constellations of nano and small satellites. Direct-to-satellite IoT connectivity, known as DtS-IoT, is now feasible through emerging long-range, low- power technologies such as narrow-band IoT, LoRa, and 5G non-terrestrial networks architecture. DtS-IoT is expected to provide services for both near-Earth and deep- space IoT deployments. In near-Earth IoT, applications face challenges such as restricted data transfer speeds, significant delays in sparse constellations, costly synchronization methods, and an inability to prioritize different types of data traffic. Meanwhile, deep-space IoT applications encounter even more difficult circumstances, including signal disruptions from planetary occlusions, limited energy availability, and lengthy transmission times due to vast distances between celestial bodies. Despite the significant challenges, the potential of a future interplanetary Internet backbone offers promising solutions. In this talk, we will discuss the challenges and research opportunities of space IoT deployments supported by an Interplanetary Internet backbone. Then, I will introduce recent research advances focused on developing a Multi-Planetary Cognitive Delay/Disruption-Tolerant Space IoT system, aiming to provide adaptiveness, autonomy, energy efficiency, security, and the scalability required by space IoT networks..
Bio:
Sandra Céspedes is an Assistant Professor with the Department of Computer Science & Software Engineering, Gina Cody School of Engineering and Computer Science, Concordia University. Previously, she was an Associate Professor with the Department of Electrical Engineering and the Head of Research at NIC Chile Research Labs, Universidad de Chile, Santiago, Chile. Dr. Céspedes holds an honorary Adjunct Professorship at Universidad Icesi, Cali, Colombia. She is an Associate Researcher with the Advance Center of Electrical and Electronic Engineering (AC3E), Chile. Dr. Céspedes received her B.Eng. (2003) and Specialization (2007) degrees in Telematics Engineering, and Management of Information Systems, from Universidad Icesi, Colombia, and a Ph.D. (2012) in Electrical and Computer Engineering from the University of Waterloo, Canada. Her research focuses on the topics of routing and protocols design for restricted devices and the Internet of Things, vehicular communications systems and networking, and cyberphysical systems. She serves as an Associate Editor for the IEEE Internet of Things Journal and the IEEE Vehicular Technology Magazine.
Google Scholar Profile: https://scholar.google.com/citations?user=8RDQ4k8AAAAJ&hl=en&oi=ao
Oscar Mauricio Caicedo, Universidad del Cauca, Colombia
LLM for Network Management: opportunities and challenges
The Zero-touch Network & Service Management (ZSM) paradigm, a direct response to the increasing complexity of communication networks, is a problem-solving approach. ZSM can take advantage of recent advances in generative Artificial Intelligence. In this regard, Large Language Models (LLMs) and Short Language Models (SLMs) can be used to architect ZSM agents. These agents can be responsible for providing sel-management functions (self-healing, self-configuration, self-monitoring, self-optimization, and self-protection) intended to manage in an efficient, autonomous (with minimal or without human intervention), and intelligent way softwarized and virtualized networks as well as 5G and beyond networks. As LLMs and SLMs are in development, it is necessary to explore the opportunities and challenges of integrating them in functional areas of network management to achieve ZSM fully
Bio:
Oscar Caicedo [GS’11, M’15, SM’20] is a full professor at the Universidad del Cauca, Colombia, where he is a member of the Telematics Engineer-ing Group. He received his Ph.D. degree in computer science (2015) from the Federal University of Rio Grande do Sul, Brazil, and his M.Sc. in telematics engineering (2006) and his degree in electronics and telecommunications engineering (2001) from the Universidad del Cauca. His recent research interests include network and service management, network functions virtualization, software-defined networking, machine learning for networking, and network softwarization. He serves as Series Editor for the IEEE Communications Magazine Series on Network Softwarization and Management and Associate Editor of the IEEE Networking Letters and IEEE Latin America Transactions. He has served and continues to serve as Technical Program Co-Chair for IEEE-sponsored international conferences, including the IEEE Global Communications Conference and IEEE Latin-American Conference on Communications. Also, he is currently a TPC member of relevant conferences, including NetSoft, CNMS, NOMS, and a reviewer of distinguished journals, such as TNSM and ESWA.
Google Scholar Profile: https://scholar.google.com.co/citations?user=NZeJb3MAAAAJ&hl=en