UVA ECE congratulates Ph.D. students Chengshuai Shi and Beichen Wang, who have earned a McVey Fellowship. The Department's award honors Professor Eugene S. McVey's seminal contributions to the field of electrical engineering and his dedication to teaching and mentoring UVA students and junior faculty.Shi conducts research in machine learning applied to signal processing and wireless communications as a member of the Laboratory for Intelligent Communications and Networking led by Cong Shen, an assistant professor of electrical and computer engineering.
Shi's research improves dynamic spectrum access, working alongside Shen and Jing Yang, associate professor of electrical engineering and computer science at Penn State, who earned a National Science Foundation grant for spectrum and wireless innovation enabled by future technologies (SWIFT).
The team is creating a novel online learning-based framework for low-cost electronic devices to efficiently and effectively access shared spectrum. Shi uses feedback from the interaction of the network and users to improve a machine learning tool called multi-armed bandits, which helps devices quickly find open spectrum minus the trial-and-error of pulsing signals to find an open channel.
“My interests in using machine learning tools to benefit signal processing and wireless communications started in my undergraduate years,” Shi sad. “I was introduced to both the capability of machine learning and the importance of wireless communications. Professor Shen's research perfectly meets my interests in leveraging and designing machine learning techniques to build modern wireless communication systems.”
Shi has tackled several practically motivated problems with efficient designs. For instance, he incorporated recent advances from federated learning to multi-armed bandits and proposed a general framework that can learn simultaneously from heterogeneous local devices to optimize the overall system performance. “With more efforts in this direction, I have also identified more interesting problems, which keep motivating me to explore further,” Shi said.
Shi has published research in conference venues and journals considered among the most prestigious in their respective fields. Shi first-authored two papers published in NeurIPS and two published in AISTATS; he has also first-authored papers in AAAI, IEEE ISIT and IEEE Globecom. Notably, Shi's AISTATS 2021 paper was selected as an oral presentation, an honor reserved for the top three percent of all submissions.
Beyond conference proceedings, Shi has first-authored papers published in the IEEE Journal on Selected Areas in Information Theory and IEEE Transactions on Signal Processing. Because of Shi's exceptional research achievement, he earned the highly competitive Bloomberg Data Science Ph.D. Fellowship for 2021-22, working to find better ways to rank recommendations from online searches.
“Chengshuai is without question a star in the making,” Shen said. “He is truly passionate about his research, which he pursues as a valued colleague and team leader.”
Shi continues McVey's tradition of service and mentoring. Shi serves as a teaching assistant for Shen's wireless communication course and is guiding a new Ph.D. student working in a related area of machine learning. He also organized a reading group for Shen's students.
Beichen Wang is earning his Ph.D. as member of the photonics lab led by Xu Yi, an assistant professor of electrical and computer engineering who also holds a courtesy appointment in physics.
Wang discovered photonics as an undergraduate at Nankai University, one of the top universities in China.
“Our mentors encouraged us to get a sense of what research feels like,” Wang said. “Photonics was really interesting — like how to manipulate light inside optical fiber, to control the acceleration of light. Later I started to look at different directions in the photonics area.”
The photonics field was flocking around research in microresonator-based frequency combs, also called a microcomb. This photonic device efficiently converts photons from single to multiple wavelengths. Wang was enticed by Yi's innovations in optical frequency division, which offer a pathway to make a chip-size, low-noise system that is continuously tunable over a very large range.
As the microcomb technology matured, Wang began to look at its potential applications. “We have studied the physics inside the microresonator — how light can be compressed into a short pulse — for practical use” Wang said.
Wang is especially enthusiastic about photonic applications in the microwave field, which could play critical roles in 6G technology.
“I have used photonic techniques to measure extremely high-frequency electronic signals, up to hundreds of gigahertz,” Wang said. “Photonic techniques enable us to manipulate how electronic signals propagate in their media, such as generating an arbitrary waveform for an electronic signal in time domain.”
Wang's interests are a perfect fit for Yi's pursuit of the next big technological leap in radar and global positioning systems, for which he earned a $2.4 million three-year grant from the Defense Advanced Research Projects Agency's GRYPHON program, which stands for Generating RF with Photonic Oscillators for Low Noise. Yi's GRYPHON team seeks to translate the purity and stability of high-frequency optical signals into the microwave regime where defense capabilities for positioning, navigation and timing typically operate.
Wang has gained a reputation as a self-starter. As one of Yi's first group members, Wang built the very first experimental setup for Yi's lab, which has become the go-to platform for the team's peer-reviewed and published research. Wang serves as the student manager of Yi's lab, responsible for ordering lab equipment and constructing experiment setups. Wang embraces his role as mentor and coach to Yi's group members who are onboarding and starting their own research projects.
Within 15 months of starting his Ph.D. program, Wang finished his first major experiment. He developed a new method for optical frequency division to coherently connect high sub-THz frequency and a low electronic frequency, publishing his paper, Vernier Frequency Division with Dual-microresonator Solitons, in Nature Communications.
Wang is also a multi-tasker. He has conducted collaborative research with members of Professor Andreas Beling's group, published in Light: Science and Applications, and demonstrated a technique to precisely synthesize electrical waveforms published in Photonics Research. His theoretical work became the foundation for Yi's Air Force Office of Scientific Research Young Investigator Award.
The McVey Fellowship is a great encouragement to students such as Wang who are pursuing academic careers.
“After graduation, I would like to expand my research within the photonics field,” Wang said. “I would like to work on projects that compliment my work on frequency combs and microwave photonics, as a I prepare for a career in academe.”