released March 14, 2000

Contact
Karen Green
Public Information Officer
kareng@ncsa.uiuc.edu
217.265.0748 phone
217.244.7396 fax

CHAMPAIGN, IL -- Researchers at the University of Illinois at Urbana-Champaign recently simulated the electromagnetic scattering effect of a full-size aircraft at X band microwave frequencies, the first time such a complex real-world problem has been solved in a computer simulation.

Weng Cho Chew, a professor of electrical and computer engineering in the U of I's Center for Computational Electromagnetics, and Senior Research Scientist Jiming Song used a 128-processor Silicon Graphics Origin2000 supercomputer at the National Center for Supercomputing Applications (NCSA) to complete his team's simulation. The simulation computed electromagnetic scattering at a frequency of 8 gigahertz, a microwave frequency termed the X band. The research is funded by the U. S. Air Force Office of Scientific Research through the Multidisciplinary Research Program of the University Research Initiative (MURI).

Electromagnetic scattering refers to how microwaves are scattered when they come in contact with an object—in this case an airplane. Some of the waves are scattered while others are bounced back, coding the radar waveform with information on the size, shape, and speed of the object. Chew's group developed new computational algorithms that greatly speed up the solution of integral equations that arise in analyses of scattering and radiation problems. The new algorithms make it possible to solve problems with as many as 10 million unknown variables. Conventional techniques could handle no more than a few tens of thousands of unknowns, and analysis of scattering in a full-size aircraft at X band would have taken years of compute time instead of hours.

"Last fall, using the algorithms, we were able to simulate a spherical shape with 10 million unknowns," said Chew. "In the simulations we just completed [in February], the new milestone is that we have adapted this process to solve a real-world problem. In the past scientists were able to solve these problems only on a scale model aircraft, which of course is much smaller. Also, simulating an aircraft is much harder than simulating a sphere because of its complex geometrical shape."

The process has obvious implications for the U.S. Air Force, which is constantly trying to develop aircraft that can hide their true size and shape through the scattering effect. Chew believes that the ability to solve a real-world problem of this type also has significant implications for the entire field of electrical engineering. Electromagnetic waves play a major role in all kinds of electronic systems and devices, he said, and algorithms that can accommodate millions of unknown variables could be used in a variety of computer experiments, including simulations of the myriad component combinations that affect the behavior of computer chips.

Chew's team computed electromagnetic scattering of an aircraft that measured about 400 wavelengths long at the X band microwave frequency, where one wavelength equals approximately 4 centimeters, making the aircraft roughly 16 meters long. Since one wavelength is essentially the smallest level of detail that is embodied in a radar return, the simulation allowed the research team to examine the aircraft at a resolution of about 4 centimeters on a computer—higher than ever before. The new algorithms also eliminate many of the bottlenecks associated with these types of problems. According to Chew, computer simulation experiments using the new algorithms could potentially replace scattering experiments on real aircraft.

"Experiments in the real world are very expensive, but the cost of computer experiments is going down." said Chew. "What this means is that without incurring a lot of design costs engineers can change the shape of an aircraft to alter its radar return. Also, engineers and scientists can coat a virtual aircraft with different materials to test its stealth. Many different designs can be experimented with on the computer."

Computer simulations could also become the norm in a wide range of electrical engineering research areas, Chew predicted.

"The ability to solve problems with millions of unknowns on a desktop using fast algorithms will alter the working and design habits of electrical engineers," he said. "Design of communication systems, MEMS (micro electro mechanical systems), remote sensing equipment, and optoelectronic devices could all be affected."

The National Center for Supercomputing Applications is the leading-edge site for the National Computational Science Alliance. NCSA is a leader in the development and deployment of cutting-edge high-performance computing, networking, and information technologies. The National Science Foundation, the state of Illinois, the University of Illinois, industrial partners, and other federal agencies fund NCSA.

The National Computational Science Alliance is a partnership to prototype an advanced computational infrastructure for the 21st century and includes more than 50 academic, government and industry research partners from across the United States. The Alliance is one of two partnerships funded by the National Science Foundation's Partnerships for Advanced Computational Infrastructure (PACI) program, and receives cost-sharing at partner institutions. NSF also supports the National Partnership for Advanced Computational Infrastructure (NPACI), led by the San Diego Supercomputer Center.