March 31, 1999
Contact: Chris Curran

[AWACS aircraft]

all aircraft photos courtesy of U.S. Air Force

Cincinnati -- The Air Force Research Lab at Wright-Patterson has awarded a University of Cincinnati engineer a $342,000 grant to help design and build electronic components for the F-22 fighter and the AWACS surveillance aircraft.

UC's research and development partners are TRW, Inc. of Cleveland and EDAptive Computing, Inc. of Dayton. EDAptive is a start-up engineering firm founded by UC alumnus Praveen Chawla in 1997. All but one of EDAptive's employees are UC alumni.

The dilemma facing the researchers is something called "legacy systems" and unlike the great legacies of America's war heroes, these legacy systems left behind a major headache for the military.

"A legacy system is any system that cannot be maintained," explained Perry Alexander, associate professor of computer engineering and the principal investigator on the project. "In many cases, it can't be replaced. There aren't any you can go buy. It's like you have an 8-track tape collection, and there aren't any 8-track tape players anymore."

Alexander's contribution to the project is the development of a programming language called VSPEC. EDAptive will use that language to develop a commercial computer-aided design (CAD) toolsuite for TRW. TRW will actually design and build the replacement parts. Alexander said the project is a great opportunity for Ohio engineers and industry to work together on a major technology transfer project.

"To have this impact statewide is very important," said Alexander. "Everybody wins. The university gets to do good research. EDAptive gets a product they can market, and TRW gets a solution for their problem." Chawla agrees that the project will be a boost to his company's mission. "Our mission is to transform innovations, either our own or others, into products ready for use or licensing."

The VSPEC language offers several advantages to engineers working on legacy systems. Instead of the time-consuming process of reverse engineering existing parts, VSPEC allows an engineer to simply describe the requirements for the component. That means you can replace a plane's electronics components without really understanding how the old part worked. You only have to know what it's supposed to do and how it's supposed to operate. Alexander uses this simple example.

"In an avionics system, when I pull the stick to cause the plane to go up, it better go up instantly, not three or four seconds later. Those are very important functions."

VSPEC also allows designers to generate tests of their new systems before production starts, a very cost-efficient way of avoiding mistakes. "We'll reduce TRW's costs and the Air Force's costs and any customer's costs in testing and design," said Alexander.

That capability makes VSPEC-based programs ideal for other electronics applications, including commercial microchip production. "Our ability to simulate components is being outrun by the complexity of the components," noted Alexander. "We can build these tiny, complex systems, but we can't test them or simulate them.

"Automotive engineers sit at multiscreen CAD systems. They know when they put the car together it's going to work. We don't. And it's very costly to fix problems once production is in high gear. "If a company needs a chip, they usually build a factory for that chip. It's extremely costly if they make a mistake."

Suprisingly perhaps, Alexander's high-tech language also has some very low-tech applications. "Some low-tech systems are critical like your car's computer. Teller machines are not high tech, but they had better work. Medical equipment...when you turn on the X-ray machine, you can't tolerate even a small possibility that it will fail. Even though those systems aren't high tech, those are the kind of systems we care about."

Alexander also cares about the economic impact the project is likely to have in Ohio. "One of my goals is to see this small company EDAptive grow into a big company. The project also gives my students good experience working with a real deliverables schedule. They're not hiding in the lab. Then, TRW and EDAptive can hire our students. They get good jobs in our state. I'm paid by the taxpayers of the state, and it's nice to be able to contribute something economically back."

Research News Archive
General News Archive
Public Relations Home Page
University of Cincinnati Home Page