ECE faculty receive CAREER Awards
By Lauren Eichmann
March 31, 2008
- Deming Chen plans to concentrate on the study of nano-centric design methodology for nanoscale FPGAs.
- Xiuling Li will research a new type of nanotechnology building block, a compound semiconductor nanotube.
- Matthew Frank will focus on system support for implicitly parallel programming, a strategy for performing large, complex tasks faster.
ECE assistant professors Deming Chen, Matthew I. Frank, and Xiuling Li recently received Faculty Early Career Development (CAREER) Awards from the National Science Foundation (NSF). CAREER awards are highly sought after and highly competitive and are presented to untenured faculty with assistant professor standing who are in the early stages of their careers.
Qualified candidates must successfully integrate research and education in their proposals, which if accepted, allow them the opportunity to receive as much as $400,000 in research funding over a five-year period. Since 1996, 24 ECE faculty members have been awarded the CAREER Award.
"There are many new professors these days," said Chen. "So, as a result, this award is getting quite competitive. I’m very grateful to win, as it’s my first try (at applying). It looks like NSF liked the idea, the topic, and the approach of my proposed research."
Chen, who is focusing on circuits and computer engineering related to computer-aided design, said each applicant is allowed three chances to apply and win the one-time award.
The NSF division of Electrical Communications and Cyber Systems (ECCS), under which Li applied, honors approximately 10 to 12 percent of applicants each year with the CAREER Award. "I was really happy to have received it," said Li.
Li, who has been at Illinois for a year now, said she is grateful to many of her senior colleagues who enthusiastically gave her feedback on her proposal. "They were all so willing to help. Those I contacted gave me tips on how to write it," she said. "Some of them even took the time to edit the proposal. I really appreciated that."
Chen, who said working on a proposal at this level requires a significant amount of time, also received extensive feedback from colleagues. "At least you have some support during your career as an assistant professor," said Chen of the award benefits. "Hopefully this funding can help me to focus on a specific area as a leading expert."
Chen, who won the CAREER Award through the division of Computing and Communications Foundations (CCF), said he plans to concentrate on the study of nano-centric design methodology for nanoscale FPGAs. "When we design the circuit, because it’s so complex, people need to use computer software to help," he explained. The goal, according to Chen, is to better optimize the design software tools - called electronic design automation, or EDA - so the target design can be smaller, faster, use less power, and provide more reliability.
As there are different types of integrated circuits (IC), including application-specific ICs, Chen said the FPGA chip is becoming more widely adopted, allowing for a design to be written on a configurable - or programmable - chip in the field. Such flexibility in circuit design has already proven successful, Chen said. One research firm reported 89,000 design-starts using FPGAs during 2007, and predicted 110,000 design-starts in 2010. "This is a major trend. People are trying to take advantage of this architecture for fast time to market and low design cost," said Chen. "The true potential of nanotechnology relies on the system integration of circuits, and the most current studies focus on the devices, however. If we don’t have tools to evaluate these systems early on, people are just designing the device without knowing the final end result of the device," he continued. "That’s where I come in."
He said there are four major tasks in nano-centric design methodology for nanoscale FPGAs: designing novel patterns for the nano-FPGA, nano-material modeling, nano-centric synthesis flow, and generating nano-FPGA evaluation and exploration engines. Applications of Chen’s research in this area include potential uses of nanotubes to build transistors and interconnects, and nanowires for new devices and circuit local routing structures.
Chen said his educational plan has a component on teaching elementary school children about science and engineering as part of an initiative to create a broader impact in society. In fact, he has already given short seminars about microchips in one local elementary school. "We want to reach out to these youngsters to train them so they will become the future leaders of nanotechnology," he said. On the graduate level, he also plans on developing a new course to incorporate nano-material and computer-aided design. "One of my goals is to train graduate students to be good engineers and good educators and mentors," Chen said. He is currently looking for an additional graduate student to help with the development of the research.
Li said her research funded by the CAREER Award will focus on establishing a new type of nanotechnology building block, a compound semiconductor nanotube. The most common form of nanotube is carbon nanotube; the semiconductor nanotube will present a completely different fabrication process and different functionalities, she said.
Li said her vision for this field includes two things: "First, you can use this as a nanotechnology building block - like carbon nanotubes, quantum dots, quantum wires - by itself as an active component to be part of the device. The other aspect is that you can roll up not only the basic layers, but also any other layer on top of it. So we can roll up the entire active device to turn it from 2-D into 3-D," Li said.
On the fundamental side, semiconductor nanotubes introduce one more degree of freedom, curvature, which would change how the electrons and holes are distributed, said Li. On the application side, such new nanotechnology could influence drug delivery and fuel leak detection in automobiles and spacecrafts. "You can make nano-resonators with biological materials attached to the inner wall and allow biological sensing or diagnostic," added Li.
Although the field is in its early stage, especially in the U.S., Li said she has students working in her Nanostructure Semiconductor Materials and Devices Group to study the formation process, electronic properties, and applications of the various types of semiconductor nanotubes.
In terms of educational outreach, Li believes in the importance of recruiting and retaining female engineering students at a very young age. "My belief is that if we really want to increase the female student enrollment here, and in general an interest in engineering, we have to start at a very early age," said Li. "One of the things I proposed is to take some cool science modules to elementary schools and have the faculty and graduate students try to stimulate their interest."
Li said she would also like to see a nanotechnology symposium for undergraduates at Illinois. Li proposes that once a semester, those studying nanotechnology would present their results as a way of exchanging ideas and stimulating discussion. "The field is developing so fast, and is becoming so multi-disciplinary," she said. "Especially on the application side, you need chemists, solid state physicists, engineers, and biologists to collaborate on these things. So the forum would not be limited to the (ECE) department."
Like Chen, Frank received his CAREER Award through the CCF division. He is focusing on system support for implicitly parallel programming, a strategy for performing large, complex tasks faster. "The problem that we’re addressing is that all the performance schemes that people are getting now come from multi-core computing," he said. "Dealing with the parallelism in a multi-core computer is actually quite hard for the programmer for a variety of reasons - one of which is that they have to design parallel algorithms now, and that involves a lot of experimentation."
Frank said that systems are needed which allow programmers to rapidly prototype applications. When they parallelize applications, there is a conflict between achieving parallel performance, as well as the kinds of things a software engineer would want - such as being able to handle errors and a reusable code for a variety of circumstances, Frank said. "I’m trying to develop tools that allow you to bridge that gap. They allow you to rapid prototype things, they allow you to do debugging and they allow you to write your code with the error-checking," he said.
"My initial reaction to receiving a CAREER Award was how I was incredibly relieved. The proposal takes a long time to write, and the prestige that comes with such an award is important."
Frank said his educational component focuses on developing undergraduate knowledge on the conversion of parallel programs. "The key here is as we move forward into the era that all the chips are multi-core, right now the hardware and systems don’t give you much support for writing parallel programs. You have to write them by yourself," said Frank. "That means that the undergraduates that we’re training - in order to be successful in the job markets - need to know (the systematic methodology that allows you to do the conversion between sequential and parallel programs)."
He said he hopes to provide students with a very concrete "set of recipes" for how to do such a conversion, which may be handled through courses he hopes to develop that concentrate on parallel programming.
Frank’s research group, Implicitly Parallel Architectures, is working with various applications groups to identify where parallelism is going to make a huge difference in their research. "We’re finding out their application needs and how, as computer architects, we can address those needs. We’re developing programming tools to make the transition to the multi-core era easier."
Editor's note: media inquiries should be directed to Brad Petersen, Director of Communications, at firstname.lastname@example.org or (217) 244-6376.