Keynote Speakers

Speaker/Institution	Keynote Information
Roberto La Rosa ST Microelectronics, Italy	Strategies, Techniques and Systems for Powering Low-Maintenance and Maintenance- Free Wireless Sensor Nodes In the swiftly evolving Internet of Things (IoT), wireless sensor network integration is pivotal to realizing a connected ecosystem. Market analysts forecast a monumental milestone of one trillion IoT devices by 2035. This growth is mirrored in the smart sensor market, which is projected to expand from $36.6 billion in 2020 to $87.6 billion by 2025, at a Compound An- nual Growth Rate (CAGR) of 19.0%. Simultaneously, the energy harvesting systems market is expected to grow from $440.39 million in 2019 to $817.2 million by 2025, with a CAGR of 10.91%. The challenge of powering such a vast array of IoT devices is daunting, considering the reliance on batteries and the associated maintenance and environmental costs. To illustrate, powering a trillion-node network would necessitate the daily replacement of 274 million batteries, even under the assumption of a ten-year battery lifespan. This lecture delves into the forefront of research and development in energy- autonomous systems, presenting innovative strategies and techniques that eliminate the need for batteries in electronic devices. The focus will be on the latest advancements in circuits and systems for wireless sensor nodes, aiming for sustainability with minimal or no maintenance requirements. The discussion will highlight the transformative potential of these technologies in achieving a self-sustaining IoT infrastructure, aligning with the goals of en- vironmental sustainability and operational efficiency, and charting a course for the future of smart sensor applications. Join us as we explore the intersection of innovation and practicality, paving the way for the next generation of energy-independent IoT devices within the IEEE CASS community.
Ricardo Reis UFRGS, Brazil	Traveling Inside a Chip An important way to understand the operation of computer and electronic circuits and systems is to analyze how these circuits and systems are implemented. We will take a tour inside various chips, showing solutions from the physical level to the system level, solutions observed through reverse engineering of microprocessors, such as the Z8000, and that are not described in the literature. It also shows the use of visualization to understand how EDA algorithms and tools work for physical synthesis.
Orazio Aiello Università degli Studi di Genova, Italy	Ultra-Low-Power and agile IC design: energy-autonomous and low-cost systems for distributed sensing The vision of a world where ubiquitous integrated electronic systems embedded in everyday goods (such as household appliances, surveillance equipment, and healthcare systems) are fully interconnected to collect, process, and exchange useful information, requires energy-autonomous systems for distributed sensing and data acquisition. The low- cost criterion necessitates technology/design portability, digital-like shrinkage over CMOS generations, tiny space, and minimal design effort. In this framework, the design effort for any system-on-chip that interacts with analog signals can be significantly decreased by taking benefit from the potential to utilize digital (automated) design flow even for analog building blocks. Most of the electronic equipment that connects with the environment can profit from the findings of this inquiry since most signals from the actual world are analog, whereas data processing is digital. In this framework, the keynote illustrates the concepts and the design flows that enable the implementation of analog functions by true digital circuits.
Manuel Delgado-Restituto IMSE-CNM-CSIC/President of IEEE CASS, Spain	Integrated Microsystems for the Treatment of Neural Diseases Implanted bidirectional integrated systems combining high-density neural recording and precise neuromodulation represent a rapidly advancing technology with great promise for the treatment of neural diseases and the implementation of brain-machine interfaces (BMIs). At the core of these systems are high-channel-count microelectrode arrays (MEAs), which provide access to detailed neural activity across large populations of neurons. Applications of these systems range from the treatment of pathologies such asepilepsy, Parkinson’s disease, or chronic pain, to the restoration of motor function for paralyzed patients through BMI systems that directly interface with prosthetic limbs or other external devices. However, significant technical challenges remain in realizing the full potential of these systems. High-channel-count MEAs generate massive amounts of data, requiringadvanced, high-performance circuits capable of real-time processing while ensuring minimal noise, and low power consumption to reduce heat generation. The long- term stability of electrodes also remains a concern, as tissue reactions and electrode degradation can compromise signal fidelity over time. On the other hand, neuromodulation, which involves delivering electrical stimulation to targeted brain regions, also faces technical hurdles. Designing circuits capable of delivering precise, localized stimulation without inducing tissue damage requires precise control over stimulation parameters —current, voltage, and pulse duration— and careful design of power management units. One promising approach is the integration of edge computing directly within the implant for closing the loop between recording and stimulation operations. The goal is to process on-chip raw neural data to extract critical features in real time and deliver controlled stimulation when and where needed, ultimately supporting more efficient and autonomous system operation. However, the integration of these closed-loop architectures poses challenges in miniaturization, thermal management, and accuracy, all of which are critical for long-term, stable implantation. This talk highlights the intersection of cutting-edge microelectronics, materials science, and neural engineering to address the key technical and biological challenges of bidirectional neural interfaces. Recent advances in energy-efficient designs, miniaturized implantable electronics, and conformable bio-materials are also presented. As these technologies mature, they will play a crucial role in treating neurological disorders and enabling seamless brain-machine interfaces, profoundly impacting both healthcare and neuroprosthetics.
Iole Moccagatta Intel Corporation, United States	The state of video compression standards: strong, dynamic, and embracing deep learning-based technologies The state of video compression standards is strong and dynamic, and more compression is coming in their future. To explain why will start by presenting deployment and adoption of the two most recent video compression standards: AV1 and VVC. Will then discusses new deep learning-based video compression technologies that standards development organizations are investigating and exploring for standardization. Will close highlighting some of these investigations that have reached the standardization stage.