Collins Decodes Problems, Keeping Galileo Safe in Outer Space
By Ken Keatley

Twice--once before launch, once millions of miles from earth--
Oliver Collins has rescued the $1.3 billion deep space probe
Galileo from mission-threatening peril.  
     "I'm not quite sure I'd say I'm a troubleshooter for
Galileo," said Dr. Collins, 30, an assistant professor of
electrical and computer engineering in the School of Engineering.
"It's gotten into trouble at interesting times for me."
     Dr. Collins--whose erudite manner and keen intellect mask
his unlikely avocation as a steed-riding, Wyoming-roaming cowboy-
-has been blessed with immaculate timing. Each time he finds a
solution, there is a problem just waiting for it.
     "It's extremely rare, even in engineering, for people to be
working on an academic problem and then to have somebody actually
build a machine based on the results," Dr. Collins said. "The
probability is less than once in a lifetime, so it's very nice to
have it happen twice."
     His first once-in-a-lifetime experience took place in 1988,
while he was working on a Ph.D. in electrical engineering at the
California Institute of Technology. 
     Delayed after the 1986 explosion of the space shuttle
Challenger, Galileo was again threatened. Because of
Challenger-inspired changes, the upper stage rocket that would
propel Galileo on its more than 2 billion-mile mission was not as
powerful as originally planned, and the change in orbit would
have produced a decrease in information transmission capability.
     As a result, Jet Propulsion Laboratory scientists had to
find a way to get the radioisotope thermoelectric generators that
power Galileo to transmit the data collected from Jupiter by
using less power. It required an upgrade to the spacecraft's
communications coding system, and just such an upgrade--a new
convolutional decoder--was the subject of Dr. Collins' thesis
project.
     "There was a fundamental problem that had kept people from
implementing these particular codes, and I was able to solve that
problem," he explained. "So I got a chance to see what I worked
on for my thesis put into use, which was really quite a lot of
fun."
     Galileo was launched on Oct. 18, 1989, deployed from the
space shuttle Atlantis on the first leg of a six-year journey
that will ultimately probe the secrets of the planet Jupiter.
     But 18 months into the mission, Galileo was again plagued
with a crisis that could have prevented the data being recorded
from being transmitted to Earth. Prior to a flyby of the asteroid
Gaspra, Galileo's fragile, umbrella-like, high-gain antenna 
became stuck, and unable to unfurl properly.
     JPL scientists determined the cause of the problem_some of
the graphite ribs in the antenna were jammed in the closed
position_but not a way to fix it. 
     Quite conveniently, Dr. Collins, by then a young faculty
member at Hopkins, was working on a new coding scheme--partly
funded by JPL--called determinate state convolutional codes, that
would further boost transmission efficiency, even with a balky
antenna.
     "Just about the time that I had some results and was ready
to start thinking about writing a paper on it, Galileo's antenna
didn't open," he recalled. "I kept getting progressively more
urgent phone calls from JPL: 'Could you please send us the paper,
now?'
     "I got it to them in time, and it turned out to be the best
solution for upgrading the spacecraft's software," he said.
     In April Dr. Collins won the Marconi Young Scientist Award
for his research on determinate state convolutional codes and
long constraint length decoders. Honored in Washington at a
black-tie dinner, he received an engraved bronze medal and a
$10,000 award from the Marconi Foundation.
     Also this year, his research titled "The Subtleties and
Intricacies of Building a Constraint Length 15 Convolutional
Decoder" garnered him the 1994 Browder J. Thompson Memorial Prize
Paper Award from the Institute of Electrical and Electronics
Engineers. The award was presented June 30 during the IEEE
International Symposium on Information Theory in Tron-heim,
Norway. The prize included a $1,000 award.
     The Thompson Prize puts the Department of Electrical and
Computer Engineering in select company, as only a handful of
universities have won it more than twice. Former department chair
William Huggins, now professor emeritus, was a Thompson Prize
honoree in the 1940s.
     Dr. Collins continues to conduct research in new coding
techniques and means for implementing them, with the end result
of achieving very large scale, low cost data collection and
improvements in cellular telephone and terrestrial satellite
communications. 
     "In a sense, I've been spoiled," Dr. Collins said. "It is
ridiculously improbable to have two things actually used in a
large scale way in the course of four or five years. Now, the
challenge is to try to come up with something else."