The latest in graduate research, which often is on the cutting edge of expanding theories into practice, was on display at the College of Engineering’s annual Graduate Symposium. This year, topics included mixed-reality for testing multiple robots, managing heat from fuel cells, predicting cancer cell movement with neural networks, and predicting water main breaks, among many others. The work was presented to prospective and fellow students, as well as visiting ECE alumni.
Faculty and ECE alumni judged the posters and presentations, and winners were chosen in each area of study. Several ECE students were recognized in their research divisions, and one presenter earned a Towner PhD Research Award.
The following graduate students earned awards for their research projects in areas of research associated with Electrical and Computer Engineering.
Applied Electromagnetics and Plasma Science:
Nikolaos Chiotellis – 1st prize, for "Metamaterial Bessel Beam Radiator." Advised by Anthony Grbic.
Fatemeh Akbar – 2nd prize, for "Scalable Phased Array Architectures with a Reduced Number of Tunable Phase Shifters." Advised by Amir Mortazawi.
Omar Abdelatty – 3rd prize, for "A Position-Independent Highly-Efficient Wireless Power Transfer Based on Coupled Nonlinear Resonant Circuits." Advised by Amir Mortazawi.
Controls, Dynamics, and Robotics
Zhen Zeng – 3rd prize, for "Semantic Mapping: Revisiting Put That There in Real Domestic Environment." Advised by Odest Chadwicke Jenkins.
Integrated Circuits, VLSI, and Microsystems:
Tal Nagourney – 1st prize, for "High Quality Factor Gyroscope Resonators Formed with Blowtorch Reflow Molding." Advised by Khalil Najafi.
Kyuseok Lee – 2nd prize, for "A 272.49 pJ/pixel CMOS Image Sensor with Embedded Object Recognition and Bio-Inspired 2D Optic Flow Generation for Nano-Air-Vehicle Navigation." Advised by Euisik Yoon.
Optics, Photonics, and Solid-State Devices
Soumitra Joy – 2nd prize, for "Artificial Plasmon: a design-friendly alternative for micro-biosensing and high speed millimeter-scale communication." Advised by Pinaki Mazumder.
Wenzhe Zang – 3rd prize, for "A High Speed, High Sensitivity, and Universal Graphene Vapor Sensor for Both Polar and Non-polar Molecules." Advised by Zhaohui Zhong.
Chengang Ji – 4th prize, for "Structural Color Filters: Fundamentals and Opportunities for Real- world Applications." Advised by Jay Guo.
Zumrad Kabilova – 5th prize, for "Charge transport in highly doped (010) β-Ga2O3 single crystals made by edge-defined film-fed growth." Advised by Becky Peterson.
Power and Energy
Sung Yul Chu – 1st prize, for "Transfer-Power Measurement: A Non-Contact Method for Fair and Accurate Metering of Wireless Power Transfer in Electric Vehicles." Advised by Al-Thaddeus Avestruz.
Stephanie Ross – 2nd prize, for "Impacts on the Local Power Network when Residential Loads Provide Energy Balancing Services to the Regional Network." Advised by Johanna Mathieu.
Md Salman Nazir – 2nd prize, for "Addressing Synchronization and Oscillations under Market-based Coordination of Distributed Energy Resources." Advised by Ian Hiskens.
Systems, Software Engineering and Computer Science
Mohammad Mahdi Khalili – 2nd prize, for "Effective Premium Discrimination for Designing Cyber Insurance Policies with Rare Losses." Advised by Mingyan Liu.
Signal and Image Processing, Computer Vision
Morteza Noshad – 2nd prize, for "Optimal Estimation of Information Measures and their Applications." Advised by Al Hero.
Megha Ghosh – 3rd prize, for "Predictive Models for Transitions in Brain States." Advised by Omar Ahmed.
The following alumni visited campus to judge presentations and meet with students:
Read more about the work of these and other presenters in the Book of Abstracts.
This is a college-level competition intended to highlight the innovation and creativity demonstrated by our Ph.D. students. Each winner received $2500 and had their name added to a perpetual plaque that is displayed in the lobby outside of Chesebrough Auditorium. Three students were awarded a Towner Prize, one representing ECE:
Principal Component Analysis (PCA) is a classical method for reducing the dimensionality of data by projecting them onto a low-dimensional subspace that captures most of their variation, and it has numerous applications ranging from environmental sensing to anomaly detection and visualization to name just a few. However, conventional PCA treats all data uniformly and does not exploit any knowledge we may have of the relative quality of each sample. If the noise variance of each sample is known, a more natural approach is to give less noisy samples more weight, i.e., to use a weighted PCA. Common choices for weights include binary weights (i.e., throwing away noisier samples) and inverse noise variance (i.e., maximum likelihood weighting). This work analyzes the statistical performance of weighted PCA for high-dimensional data drawn from a low-dimensional subspace and degraded by heteroscedastic noise (i.e., noise that has non-uniform variance across samples).David Hong is advised by Prof. Jeffrey Fessler.