Once every week while school is in session, EECS invites a distinguished researcher or practitioner in a computer science or electrical and computer engineering-related field to present their ideas and/or work. Talks are generally targeted to electrical engineering and computer science graduate students. This colloquium series is free and open to everyone.

Upcoming Colloquia

Extreme Compression in Convex and Non-convex Inverse Problems: Role of Geometry, Priors and Measurement Design

Monday, January 27, 2020 - 4:00pm to 4:50pm
Linus Pauling Science Center 125

Speaker Information

Piya Pal
Assistant Professor
Electrical and Computer Engineering
University of California, San Diego

Abstract

Inferring parameters of interest from high dimensional data is a central problem in signal processing and machine learning. Fortunately, many modern datasets possess low dimensional structure (such as sparsity, low-rank) which can be judiciously exploited to reduce the cost of sensing and computation. Starting from seminal works in compressed sensing and linear underdetermined estimation, there has been tremendous progress towards understanding how such low dimensional structure can be optimally exploited in a variety of convex and non-convex inverse problems with provable theoretical guarantees. Celebrated results (which, in many cases, rely on randomized measurements to establish probabilistic guarantees) indicate that in many of these problems, it is indeed possible to obtain reliable inference with a sample complexity that is proportional to the underlying (low) dimension.

Many inverse problems of practical interest (such as those arising in source localization, super-resolution imaging, channel estimation) possess additional geometry that is imparted by the physical measurement model, physical laws governing wave propagation, as well as statistical priors (such as correlation) on the unknown quantities of interest. In this talk, I will demonstrate how to tailor the design of “smart” sensing systems and develop corresponding reconstruction algorithms that can achieve significantly higher compression (henceforth termed extreme compression) than existing guarantees on sample complexity. Instead of randomized measurements, I will focus on the design of deterministic Fourier-structured measurement matrices (that naturally arise in many practical imaging problems) and exploit combinatorial designs (governed by the idea of “difference sets” in one and multiple dimensions) to attain such extreme compression. I will derive non-asymptotic probabilistic guarantees in this regime by developing new algorithms that carefully exploit the geometry of these smart samplers. Throughout my talk, I will draw examples from applications in radar and sonar signal processing, super-resolution optical imaging, neural signal processing and hybrid channel sensing.

Speaker Bio

Piya Pal is an Assistant Professor of Electrical and Computer Engineering, and a founding faculty member and faculty advisor of the Halıcıoğlu Data Science Institute at the University of California, San Diego. Her research interests include signal representation and sampling for high-dimensional inference, super-resolution imaging, convex and non-convex optimization, and statistical learning. Her research has been recognized by the 2019 Presidential Early Career Award for Scientists and Engineers (PECASE), 2019 Office of Naval Research Young Investigator Program (ONR YIP) Award, 2016 NSF CAREER Award, 2018 Qualcomm Fellow Mentor Advisor Award and the 2014 Charles and Ellen Wilts Prize for Outstanding Doctoral Thesis in Electrical Engineering at Caltech. She is also a two-time recipient of the Best Graduate Teaching Award in Electrical and Computer Engineering at UC San Diego (in 2017 and 2018). She and her students have received several best paper awards at conferences, including the Best Student Paper Award (first position) at the IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP) 2017 and the Best Student Paper Award (first position) at 2019 IEEE CAMSAP. She is an elected member of the IEEE SAM and SPTM Technical Committees. She received her B.S. degree in Electronics and Electrical Communication Engineering from the Indian Institute of Technology, Kharagpur in 2007 and her Ph.D. in Electrical Engineering from the California Institute of Technology (Caltech), Pasadena in 2013.

Tensor-Based High-Resolution Channel Parameter Estimation for Hybrid MIMO OFDM Systems in the Millimeter Wave Band

Monday, February 10, 2020 - 4:00pm to 4:50pm
Linus Pauling Science Center 125

Speaker Information

Martin Haardt
Professor
Communications Research Laboratory
Ilmenau University of Technology

Abstract

In this talk, we present a gridless channel estimation scheme for MIMO OFDM systems in the millimeter wave (mmWave) band that is based on R-D Unitary Tensor-ESPRIT in DFT beamspace. Compared to conventional ESPRIT based algorithms in element space, the beamspace approach can be applied to MIMO systems with hybrid architectures. Moreover, the proposed scheme significantly reduces the training overhead for communication systems operating in the mmWave band. The proposed procedure involves coarse and fine estimation steps.

During the coarse estimation step, Unitary Tensor-ESPRIT in element space is applied to the array with a reduced size aperture to obtain initial information about the directions of arrival, the directions of departure, and the propagation delays of the dominant multipath components. Based on these estimates, a more accurate estimation of the angular profiles, propagation delays, and channel gains is performed in a second step by applying 3-D Unitary Tensor-ESPRIT in DFT beamspace in the spatial domains combined with the element space version in the frequency domain. We explain how to combine the received signals from different spatial sectors of interest and how to perform joint processing. The simulation results confirm the tensor gain of the proposed procedure in addition to the improved channel estimation accuracy.

Speaker Bio

Martin Haardt has been a Full Professor in the Department of Electrical Engineering and Information Technology and Head of the Communications Research Laboratory at Ilmenau University of Technology, Germany, since 2001.

After studying electrical engineering at the Ruhr-University Bochum, Germany, and at Purdue University, USA, he received his Diplom-Ingenieur (M.S.) degree from the Ruhr-University Bochum in 1991 and his Doktor-Ingenieur (Ph.D.) degree from Munich University of Technology in 1996.

In 1997 he joint Siemens Mobile Networks in Munich, Germany, where he was responsible for strategic research for third generation mobile radio systems. From 1998 to 2001 he was the Director for International Projects and University Cooperations in the mobile infrastructure business of Siemens in Munich, where his work focused on mobile communications beyond the third generation. During his time at Siemens, he also taught in the international Master of Science in Communications Engineering program at Munich University of Technology.

In 2018, Martin Haardt has been named an IEEE Fellow “for contributions to multi-user MIMO communications and tensor-based signal processing.” He has received the 2009 Best Paper Award from the IEEE Signal Processing Society, the Vodafone (formerly Mannesmann Mobilfunk) Innovations-Award for outstanding research in mobile communications, the ITG best paper award from the Association of Electrical Engineering, Electronics, and Information Technology (VDE), and the Rohde & Schwarz Outstanding Dissertation Award.

In the fall of 2006 and the fall of 2007 he was a visiting professor at the University of Nice in Sophia-Antipolis, France, and at the University of York, UK, respectively. From 2012 to 2017, he also served as an Honorary Visiting Professor in the Department of Electronics at the University of York, UK, and in 2019 as an Invited Professor at the Université de Lorraine in Nancy, France.

His research interests include wireless communications, array signal processing, high-resolution parameter estimation, as well as numerical linear and multi-linear algebra.

Prof. Haardt has served as an Associate Editor for the IEEE Transactions on Signal Processing (2002-2006 and 2011-2015), the IEEE Signal Processing Letters (2006-2010), the Research Letters in Signal Processing (2007-2009), the Hindawi Journal of Electrical and Computer Engineering (since 2009), the EURASIP Signal Processing Journal (2011-2014), as a senior editor of the IEEE Journal of Selected Topics in Signal Processing (JSTSP, since 2019), and as a guest editor for the EURASIP Journal on Wireless Communications and Networking as well as the IEEE JSTSP.

From 2011 to 2019 he was an elected member of the Sensor Array and Multichannel (SAM) technical committee of the IEEE Signal Processing Society, where he served as the Vice Chair (2015 – 2016), Chair (2017 – 2018), and Past Chair (2019). Since 2020, he has been an elected member of the Signal Processing Theory and Methods (SPTM) technical committee of the IEEE Signal Processing Society. Moreover, he has served as the technical co-chair of PIMRC 2005 in Berlin, Germany, ISWCS 2010 in York, UK, the European Wireless 2014 in Barcelona, Spain, as well as the Asilomar Conference on Signals, Systems, and Computers 2018, USA, and as the general co-chair of WSA 2013 in Stuttgart, Germany, ISWCS 2013 in Ilmenau, Germany, CAMSAP 2013 in Saint Martin, French Antilles, WSA 2015 in Ilmenau, SAM 2016 in Rio de Janeiro, Brazil, CAMSAP 2017 in Curacao, Dutch Antilles, SAM 2020 in Hangzhou, China, and the Asilomar Conference on Signals, Systems, and Computers 2021, USA.

A Cross-Layer Approach to Retrofitting Robotic Vehicle Controllers

Thursday, February 13, 2020 - 10:00am to 11:00am
KEC 1007

Speaker Information

Chung Hwan Kim
Researcher
NEC Labs America

Abstract

Robotic vehicles (RVs) such as unmanned aerial vehicles are a type of cyber-physical system for autonomous transportation and missions. With their increasing adoption, RVs are facing threats of cyber and cyber-physical attacks that exploit their attack surfaces. Although many RVs are critical to human safety and the environment, it is difficult to make them secure against such attacks due to new challenges that are not addressable by traditional approaches. Many of these challenges originate from the limited hardware resources and semantic unawareness of RV controllers in security mechanisms.

In this talk, I will present an overview of my recent work that discovers and removes two new attack surfaces in RVs. More specifically, I will first discuss an attack surface caused by the absence of memory isolation in RVs. I will show how my work retrofits the system architecture of RV controllers and eliminates more than 75% of the attack surface under hardware and real-time constraints. I will then describe a new type of semantic vulnerabilities in RV controllers and a novel tool to detect the vulnerabilities in commodity RVs through automated testing. I will conclude this talk with a discussion of my on-going and future research on security, resilience, and privacy of cyber-physical systems.

Speaker Bio

Chung Hwan Kim is a security researcher at NEC Labs America in Princeton, New Jersey. He received his Ph.D. in Computer Science from Purdue University in 2017 and joined NEC Labs in the same year. His research interest lies in solving security and reliability problems in a broad variety of computing systems from cyber-physical systems to enterprise and cloud systems. His research seeks to achieve this by using and developing a unique combination of techniques in program analysis and operating systems. His work has been nominated as a top 10 finalist for the CSAW Best Applied Research Paper Award (2018) and received a Business Contribution Award from NEC Labs America (2015). He has served as a program committee member of NDSS’19 and is serving as a program committee member of DSN’20.

More than Figure-of-Merit ADC Design Perspective

Friday, February 21, 2020 - 11:00am to 12:00pm
KEC 1003

Speaker Information

Chi-Hang Chan
Assistant Professor
University of Macau

Abstract

In the past few decades, ADC researchers have spent numerous efforts to push the energy efficiency toward the fundamental boundary based on Hartley-Shannon law. On the other hand, when looking at the ADC designs nowadays, it can be recognized that they already optimized enough where even 2x or 4x power reduction does not make much difference in the system point of view. On the other hand, such low power designs often leave burdens to the peripheral circuits or trading off specifications not shown from the FoM, which eventually leads to diminishing returns in practice. In this talk, several issues will be first overviewed, then followed by more-than-FoM design examples from our group.

Speaker Bio

Chi-Hang Chan was born in Macau S.A.R., China, in 1985. He received the B.S. degree in electrical engineering from University of Washington (U.W. Seattle), USA, in 2008, and the M.S. degree and Ph.D. from the University of Macau, Macao, China, in 2012 and 2015, respectively. He currently serves as assistant professor at SKL-AMSV at University of Macau, Macao and Secretary of IEEE Solid State Macau Chapter. He received the Chipidea Microelectronics Prize and Macau Science and Technology Development Fund (FDCT) Postgraduates Award during his Master and PhD study. He has also received Macau FDCT Award for Technological Invention in 2014, 2016 and 2018 for his outstanding academic and research achievements in microelectronics. He is the recipient of the 2015 Solid-State-Circuit-Society (SSCS) Pre-doctoral Achievement Award and co-recipient of ESSCIRC 2014 best paper award. He has authored and co-authored 13 Journal Solid State Circuit (JSSC) and 10 International Solid State Circuit Conference (ISSCC) papers in the data converter and clock circuit design field. His research interests include high speed Nyquist, noise shaping ADCs and low jitter clock circuits.

 

NOTE: This is not part of the EECS colloquium series.

High Performance Hybrid ADCs & Digital Assisted Techniques

Friday, February 21, 2020 - 2:00pm to 3:00pm
KEC 1003

Speaker Information

Yan Zhu
Associate Professor
University of Macau

Abstract

SAR (Successive Approximation Register) ADCs rely on the switch-capacitor circuit to perform the binary/non-binary-searched feedback to the input signal. Excluding its peripheral interface circuities, its operation is fully dynamic and contains mostly the digital feature; thus, it demonstrates the best conversion efficiency as compare to the other conventional ADC architecture. This talk uses a SAR ADC as base, introducing several SAR-reconfigured architectures that can breakthrough its performance bottleneck regarding the speed and resolution. Such SAR-based hybrid ADCs achieve similar design specifications of a conventional pipelined, while still maintains the excellent power efficiency like a SAR ADC.

Speaker Bio

Yan Zhu (S’10- M’17-SM’19) received the M.Sc. and Ph.D. degrees in electrical and electronics engineering from the University of Macau Macao, China, in 2009 and 2011, respectively. She is now an associate professor with the State Key Laboratory of Analog and Mixed-Signal VLSI, University of Macau, Macao, China. She received Best Paper award in ESSCIRC 2014, the Student Design Contest award in A-SSCC 2011, and 4 times Macao Scientific and Technological R&D Awards for outstanding Academic and Research achievements in Microelectronics. She has published 70+ technical journals and conference papers in her field of interests, and holds 3 US patents. Her research interests include low-power and wideband high-speed Nyquist A/D converters, PLL and machine learning for image processing.

 

NOTE: This is not part of the EECS colloquium series.

Robust Methods for Topology Estimation in Unsupervised Learning

Monday, February 24, 2020 - 4:00pm to 4:50pm
Linus Pauling Science Center 125

Speaker Information

Shay Deutsch
Assistant Adjunct Professor
Mathematics Department
University of California, Los Angeles

Abstract

Learning graph connectivity has broad-ranging applications from 3D reconstruction to unsupervised learning. In this talk I will introduce a new method to learn the graph structure underlying noisy point set observations assumed to lie near a complex manifold. Rather than assuming regularity of the manifold itself, as customary, we assume regularity of the geodesic flow through the boundary of arbitrary regions on the graph. The idea is to exploit this more flexible notion of regularity, captured by the discrete equivalent of the isoperimetric inequality for closed manifolds, to infer the graph structure.

In a broader perspective, when studying the topology of the graph networks, we would like to learn new representations that capture not only local connectivity, i.e., nodes that belong to the same local structure, but also similarity which is based on their structural role in the graph. I will discuss a new approach and vision towards learning a good trade-off between these local and structural types of similarities that includes diverse possible applications including point clouds, biological networks and social networks.

Speaker Bio

Shay Deutsch received a B.Sc. in Mathematics from the Technion—Israel Institute of Technology in 2007, an M.Sc. in Applied Mathematics from Tel Aviv University in 2010, and a Ph.D. in Computer Science from the University of Southern California (USC) in 2016. He is currently an Assistant Adjunct Professor in the Mathematics Department at the University of California, Los Angeles (UCLA). His research work is in the union of transfer learning, graph signal processing and graph networks, where his research is dedicated to developing robust methods for unsupervised learning. His most recent research efforts focus on developing cohesive relations between embedding topology and graph networks using uncertainty principles on graphs.

Past Colloquia

Frank O'Mahony
Monday, October 17, 2016 - 4:00pm to 4:50pm
David Hendrix
Monday, October 10, 2016 - 4:00pm to 4:50pm
Robert Adams
Monday, October 3, 2016 - 4:00pm to 4:50pm
Kofi A.A. Makinwa
Monday, June 13, 2016 - 4:00pm to 4:50pm
Nan Sun
Monday, May 9, 2016 - 4:00pm to 4:50pm
Alan Hui Zhao
Monday, April 25, 2016 - 4:00pm to 4:50pm
David J. Allstot
Monday, April 25, 2016 - 9:00am to 10:00am
Subhasish Mitra
Monday, April 18, 2016 - 4:00pm to 4:50pm

Pages