IEEE International Conference on Microelectronic Test Structures

ICMTS 2024 Tutorials

08:00		Registration opens
09:00		Tutorials Welcome J. Klootwijk (Tutorials Chair) Philips Research
09:10	T1	Fundamentals of Measurement and Test Structures Prof Yoshio Mita University of Tokyo, Japan Semiconductor electronics have continuously benefited from mass production due to miniaturization and gave variety of application fields. To increase semiconductor manufacturability, researchers and engineers had to know "what is occurring" in the microelectronic devices. To efficiently measure an extract, in other words, to know through measurements (door meten tot veten), researchers and engineers have developed a variety of test structures. Professor Yoshio Mita, who is known among students to give lectures in an enthusiastic manner, will introduce the history of test structures to inspire researchers and engineers for the future.
10:00		Break
10:30	T2	Interest of cMUT technologies : from materials and processes to applications Prof Daniel Alquier Université de Tours, France After introducing MEMS technologies available today, the talk aims to present the history as well as the developments of capacitive Micromachined Ultrasonic Transducer (cMUT) during the last 30 years. We will review device fabrication, investigate the crucial steps through the various processes available to achieve such structures and obtain vibrating systems. We will try to better understand how the technological stages impact on cMUT devices and how to ensure device functionality and reliability. Further, we will review the first application of cMUT devices i.e., medical imaging and, hence, present many others that cMUT’s can address successfully. cMUT prospective as well as key issues on new material developments based on SiC technologies and dissemination of cMUT in harsh environment applications, developed in our group, will be presented.
11:20		Break
11:30	T3	Memristive Technologies: Testing, Modeling and Applications Prof Themis Prodromakis University of Edinburgh, UK Artificial Intelligence (AI) is destined to transform our society, affecting every aspect of our lives. However, a key bottleneck towards the proliferation of the technology is the lack of efficient hardware that will allow us to embed AI everywhere – well beyond the cloud’s reach. Up until now, the processing and storage of data in electronics has relied on assemblies of vast numbers of transistors that have got smaller and smaller to meet the increasing demands of modern societies, but have nowadays reached their physical limits. A novel nanoelectronic technology, known as the memristor, proclaims to hold the key to a new era in electronics and AI, being both smaller and simpler in form than transistors, low-energy, and with the ability to retain data by ‘remembering’ the amount of charge that has passed through them – akin to the behaviour of synaptic connections in the human brain. In this tutorial Professor Prodromakis will present the attributes of memristive technologies that make this emerging technology attractive for a variety of applications, along with the tools and techniques developed by his group for exploiting this in wide range of applications –from bio-inspired memories to compressing sensing and AI hardware accelerators.
12:20		Break
13:40	T4	The Multifaceted Impact of Resistive Memories on Neuromorphic Systems Dr Elisa Vianello CEA-LETI, France Resistive random access memory (RRAM) technologies, commonly known as memristors, hold immense potential for revolutionizing neuromorphic and in-memory computing systems. These systems offer significant advantages, such as improved parallelism, reduced power consumption, and minimal latency, particularly in AI workloads. However, practical challenges arising from device variability hinder the implementation of such computational paradigms. In this presentation, we delve into the exploration of efficient and reliable nanosystems that leverage imprecise devices, blending the fields of neuroscience, computer science, electrical engineering, and device physics. We begin by investigating the incorporation of organizing computational principles inspired by the brain into neuromorphic circuits and architectures. Second, we introduce Bayesian inference, which, although not directly inspired by the brain, has compelling evidence of its utilization in biological intelligence. For example, Bayesian inference is employed in honeybees' decision-making processes, optimizing their food search. Bayesian neural networks, renowned for their ability to handle limited data and uncertainty, offer significant advantages in sensory processing tasks. To address concerns regarding computational intensity arising from their probabilistic nature, we propose harnessing the potential of resistive memory devices. For instance, the variability in memristor resistance can be exploited to represent weight probability distributions.
14:30		Break
14:40	T5	Fundamentals of Accurate Wafer-Level Characterization at the RF and mm-Wave Frequencies Dr Andrej Rumiantsev MPI Corporation In this workshop presentation, we will be discussing the fundamental principles of making accurate and reliable RF measurements at the wafer-level. We'll delve into the requirements for the probe system architecture, integration of the measurement instrumentation, and related system accessories. Our focus will be on the accuracy of the RF system calibration, which can be impeded by several effects, including unoptimized boundary conditions of calibration standards, unwanted modes propagating in the substrate, the parasitic coupling of calibration standards and RF probe with neighboring elements, specifics of the calibration algorithm used, the impact of temperature, system operator and the laboratory environment, and others. We'll review concepts and essential differences in widely used RF calibration methods and their sensitivity to various parasitic effects. Finally, we'll look at several examples of how to improve the confidence of measured data at the mm-wave frequency range.
15:30		Break
16:00	T6	Devices for Photonic Integrated Circuits: from Characterization to Test Dr Alexandru Romanescu SMART Photonics, Netherlands A process for manufacturing Photonic Integrated Circuits (PICs) is made available for Circuit Design through the means of a PDK (Process Design Kit), comprising a library of Photonic Devices. During the Technology Development phase, these devices are characterized, with the purpose of both optimizing their behaviour, as well as extracting all the information relevant to the design process. This information makes its way in the PDK as raw data, quantitative plots in the Design Manual or Compact Models. During the production phase, the functionality of these devices must be assessed for each manufactured wafer. Efficient test methods are developed to capture the critical-to-quality parameters of each device. Their purpose is to validate the manufactured material (with significantly less resources than a full characterization), as well as to offer statistical means of monitoring and root-cause-analysis during the production ramp-up phase. This course will propose a framework for developing a characterization and test plan, as well as methods to execute these plans, specific to photonic devices.
16:50		Close of Tutorials
17:30		Welcome Reception for All ICMTS Attendees