This page is created to inform any interested participants regarding the seminars to be conducted for the EE690 Graduate Seminars course. The schedule and the content will be updated throughout the semester.
2024-2025 Fall Graduate Seminars
Title: Enabling 6G Performance in the Upper Mid-Band and, DoA Estimation with Limited Array Sizes
Abstract: This talk will be about the transition from Massive MIMO (mMIMO) to Gigantic MIMO (gMIMO) as a key enabler for 6G networks, focusing on the upper mid-band spectrum (7-24 GHz). gMIMO enhances beamforming, spatial multiplexing, and near-field communication, enabling precise localization and sensing from a single base station. However, challenges such as hardware complexity, cost, and algorithmic efficiency must be addressed to realize its full potential. Open research challenges will be discussed, including near-field antenna placement, transceiver design, and energy efficiency. Addressing these challenges will enable transformative 6G applications in smart cities, autonomous vehicles, and beyond. This presentation highlights how gMIMO can shape the future of wireless networks. In the second part of the talk, a novel MIMO-OFDM radar system for Direction of Arrival (DoA) estimation in dense environments will be presented. By filtering targets in the delay and Doppler domains and applying a fusion technique, the system achieves high accuracy even with limited antenna arrays.
Bio: This talk will be about the transition from Massive MIMO (mMIMO) to Gigantic MIMO (gMIMO) as a key enabler for 6G networks, focusing on the upper mid-band spectrum (7-24 GHz). gMIMO enhances beamforming, spatial multiplexing, and near-field communication, enabling precise localization and sensing from a single base station. However, challenges such as hardware complexity, cost, and algorithmic efficiency must be addressed to realize its full potential. Open research challenges will be discussed, including near-field antenna placement, transceiver design, and energy efficiency. Addressing these challenges will enable transformative 6G applications in smart cities, autonomous vehicles, and beyond. This presentation highlights how gMIMO can shape the future of wireless networks. In the second part of the talk, a novel MIMO-OFDM radar system for Direction of Arrival (DoA) estimation in dense environments will be presented. By filtering targets in the delay and Doppler domains and applying a fusion technique, the system achieves high accuracy even with limited antenna arrays.
Title: Systems Engineering for Multi-Function and Cognitive Radars: From Concept to Mission
Abstract: Radar systems have transitioned from traditional architectures to multifunction and cognitive designs, enabling adaptive and intelligent sensing. Modern radar systems are designed to operate across diverse platforms and fulfil multiple roles, adapting to evolving threat environments and mission requirements. Their ability to dynamically allocate resources and optimize performance in real time makes them essential in critical missions.
This seminar provides an exploration of technical management and systems engineering through the lifecycle of advanced radar systems. It covers the entire development process, including initial concept definition, requirements management, system design, and algorithm development. Additionally, it covers integration, system validation, operational deployment, and long-term mission sustainment. It demonstrates how these complex system designs evolve from theoretical concepts to fully operational capabilities.
Bio: Ülkü Doyuran received her B.S., M.S., and Ph.D. degrees in Electrical and Electronics Engineering from Middle East Technical University. She began her career as a research assistant in the department, focusing on telecommunications and signal processing. Following that, she joined Aselsan Inc. as a radar systems engineer. With approximately twenty years in the industry, she specializes in systems engineering, engineering management, technical management, and algorithm development for radar technology. She is currently working as the Radar Systems Engineering Director, leading the development of advanced radar systems and technology in the field.
Title: From Micro-Nano Scale Research to Real-Life Impact: A Journey Through Translational Engineering
Abstract: Bridging the gap between micro-nano scale engineering and life science applications presents unique challenges, not only in technical development but also in experimental rigor and scientific mindset. In this talk, I will share insights from my journey—starting from my master’s studies at METU, through my PhD research; in microfluidics, microphysiological systems, and impedance-based single-cell analysis, to my transition into industry, where I now work on applying these technologies in biotechnology and bioprocessing. A key realization in this transition was the fundamental difference in experimental expectations: while in engineering research, demonstrating a single working device from a fabricated wafer is often considered a success, biological and clinical sciences demand reproducibility across technical and biological replicates, with experimental designs that account for inherent variability in living systems. This shift in approach, along with navigating interdisciplinary collaborations, regulatory constraints, and commercial viability, has been instrumental in my work. I will also discuss the broader challenges of translating research into real-world applications, including the intersection of engineering-driven problem-solving and biology-driven complexity, and the role of resilience in interdisciplinary careers. This talk aims to provide perspectives on how engineers can effectively contribute to and shape emerging fields that intersect with life sciences.
Bio: Dr. Furkan Gökçe earned their Bachelor's and Master’s in Electrical and Electronics Engineering from METU (2015, 2017) and a PhD in Translational Bio Engineering from ETH Zurich (2023). Their research focused on micro-nano systems (BioMEMS), microfluidics and single-cell analysis, and microphysiological personalized chemotherapy screening. Now a New Business Development Manager at Amphasys AG, where they develop and commercialize micromachined single-cell analysis technologies for life sciences and bioprocessing. Their expertise spans microtechnology, biosensors, and product strategy, bridging research with industry to create market-ready solutions in bioprocessing and life sciences.
Title: Microwave Breast Cancer Detection and Screening: Roadmap from research to application
Abstract: Breast cancer is the most commonly diagnosed cancer in women and has the highest incidence of cancer in women. 1 out of 7 women is diagnosed by breast cancer every year. Early diagnosis of cancer is utmost importance and unfortunately the existing solutions on the market have such tremendous harms on the human body as they mostly use harmful x-ray that there are even academic studies that claim that some devices even increase the probability of the cancer. X-ray mammography cannot be used as screening tests for women younger than 40 years old and cannot be used more than one time in a year.
“Early detection is only possible with breast cancer screening of young women early and frequently.” It has been shown that there is a high dielectric contrast between the normal and malign tissues of the breast in microwave frequencies. This gives a very important opportunity to use microwaves for the early detection, imaging and screening of breast cancer.
The breast cancer-screening and diagnosis device that we developed, namely SAFE (Scan And Find Early), is a device used for imaging tumors in breast tissue using harmless electromagnetic waves (microwaves) without using the X-ray. Our technology is patented in Turkey and USA. “Our disruptive novel SAFE equipment can be used in all ages, as many as required, even in pregnancy” The initial market segment of SAFE is as an early breast cancer-screening device for women in high-risk breast cancer group. For that market segment there is no solution other than physical examination by hand. During the talk, we will give the details of roadmap from research to market.
Bio: İbrahim Akduman received the B.Sc., M.S., and Ph.D. degrees in electronics and communication engineering from Istanbul Technical University, Istanbul, Turkey, in 1984, 1987, and 1990, respectively. He was visiting professor at the New York University Tandon School of Engineering, Brooklyn, NY, USA, in 1991; King’s College London, London, U.K., in 1995; the New Jersey Institute of Technology, Newark, NJ, USA, in 2000; and the University of Göttingen, Göttingen, Germany, in 2001. He was the Dean of the Electrical and Electronics Engineering Faculty, Istanbul Technical University, from 1999 to 2001 and from 2020to 2023; and Vice President, from 2002 to 2004. He is currently with Istanbul Technical University, as a Full Professor, where he is also the Head of Medical Device Research, Development and Application Laboratory. He is shareholder of a company as well, where he is involved in research and developing products for medical application of electromagnetic fields. His current research interests include microwave tomography, microwave spectroscopy and microwave hyperthermia in medicine. He is a Humboldt fellow and received the Turkish Scientific and Technological Research Council Young Scientist Award in 2000.
Title: Beyond Pretraining Loss: Evaluating Value of Pretraining Data for Large Language Models at Scale
Abstract: The performance of Large Language Models (LLMs) is fundamentally dependent on the quality and quantity of their pretraining data. Understanding the impact of data choices and scale is crucial for optimizing model development and deployment. This talk will begin by highlighting the importance of accurately estimating the effect of data characteristics on LLM performance. We will then delve into scaling laws, which provide valuable insights into predicting pretraining loss as a function of pretraining data size and model parameters. However, pretraining loss alone does not fully capture the complex interplay between pretraining data and downstream task performance. To address this, we will explore our recent work extending beyond pretraining loss to directly predict downstream metric performance at scale. This approach provides a more comprehensive evaluation of value of pretraining data, enabling a nuanced understanding of how data influences real-world applications.
This talk will draw upon findings in our recent ICLR 2025 paper: Scaling Laws for Downstream Task Performance in Machine Translation.
Bio: Berivan Isik is a research scientist at Google, working on efficient and trustworthy AI. Her current interests are efficient training of large models, data valuation and scaling laws for LLMs, and unlearning. She completed her PhD at Stanford University in 2024. Her research was supported by Stanford Graduate Fellowship, Google Ph.D. Fellowship, and a Meta research grant.
Title: Phased Array Antennas: Evolution, Breakthroughs, and Future Trends
Abstract: Phased array antenna technology is a key enabler of modern telecommunications, radar, and remote sensing, allowing for adaptive and electronically steerable beamforming. Its ability to dynamically control signal direction has made it indispensable in applications ranging from 5G networks to advanced radar systems. While the foundational principles of phased arrays were established in the early 20th century, practical implementations only emerged by the mid-20th century, primarily for military applications due to high costs and technological limitations. It was not until the late 2010s that phased array technology entered mainstream civilian markets, driven by advancements in semiconductor technologies, system-on-chip (SoC) integration, and cost-effective manufacturing. This talk will trace the evolution of phased array antennas, outlining key technological advancements that have led to today’s state-of-the-art implementations. The discussion will then shift to the latest innovations, performance breakthroughs, and future trends, exploring the key technologies set to drive the next generation of phased array systems.
Bio: Doğanay Doğan received his B.S. degree in Electrical and Electronics Engineering from Middle East Technical University in 2008, followed by M.S. and Ph.D. degrees from the same department in 2011 and 2023, respectively. In 2008, he began his career as an antenna design engineer at ASELSAN's Radar and Electronic Warfare Systems (REHİS) Business Sector. From 2018 to 2023, he served as the Head of the Phased Array Antenna Design Department at REHİS, and since 2023, he has held the position of Hardware Design Director. Over the course of his 17-year career, he has been extensively involved in projects focusing on antenna design, measurement, beamforming/shaping, and numerical electromagnetic analysis.
Title: Ultrahigh Field Magnetic Robotics
Abstract: With remote magnetic steering capabilities, magnetically actuated robotic system have proven their potential in minimally invasive medical procedures. However, integrating the magnetic actuation system into clinically available medical imaging systems and tracking these robots during operations is still a significant challenge. Magnetic resonance imaging (MRI) scanners have recently been proposed as magnetic robotic platforms, combining medical imaging and magnetic actuation in a single system. Besides providing ionizing radiation-free, high-quality, three-dimensional soft tissue images, MRI scanners are also human-sized electromagnets operating at high magnetic fields.
In this talk, I will introduce the basic magnetic actuation concepts at ultrahigh fields and explain how MRI scanners are used as magnetic actuation platforms for medical robotic systems, such as wireless capsule robots, guidewires, and catheters. Furthermore, I will introduce novel magnetic tracking methods at ultrahigh fields. Finally, I will conclude the talk by discussing the future directions of magnetic robotic system medical procedures.
Bio: M. Efe Tiryaki received his B.S. degree from Middle East Technical University, Ankara, in 2016 in mechanical engineering and physics and his M.S. degree from ETH Zurich in robotics in 2018. Then, he completed his Ph.D. work on “MRI-powered Magnetic Microrobotics” at Max-Planck Institute for Intelligent Systems and ETH Zurich in 2023. He is currently a postdoctoral researcher at the Max-Planck Institute. His research focuses on autonomous medical robotic systems in MRI scanners and developing novel magnetic actuation and tracking methods for minimally invasive medical operations.
Title: When Math Meets Applications: The Power of Computational Imaging
Abstract: Computational imaging is transforming the way we acquire and interpret data, pushing beyond the physical constraints of traditional optical systems. By leveraging advanced numerical methods and machine learning, this interdisciplinary field has enabled breakthroughs in microscopy, medical imaging, satellite remote sensing, and astronomy. Recent progress in large-scale 3D imaging has further expanded its reach, improving both resolution and robustness in challenging real-world scenarios. In this talk, I will explore key aspects of computational imaging, focusing on my recent work in tomography, phase retrieval, and image segmentation. Through these techniques, we can tackle fundamental challenges in reconstruction and interpretation, paving the way for novel applications and interdisciplinary collaborations. By examining these techniques, I will highlight ongoing challenges and opportunities for further development, fostering discussions on the future directions of imaging science.
Bio: Selin Aslan joined Koc University in Fall 2023 as an Assistant Professor in the Department of Mathematics. Prior to this, she was a postdoctoral appointee at Argonne National Laboratory, USA, where she collaborated with an interdisciplinary team of applied mathematicians, computer scientists, and physicists to develop new algorithmic and computational capabilities for next-generation microscopy instruments. Her primary research interest lies in developing algorithmic and computational solutions for large-scale imaging problems, with a special emphasis on inverse problems and optimization. Her work spans areas such as phase retrieval, tomography, large-scale data, and deep learning.