UBC Publications

UBC Publications

UBC Publications

Focus 2002

Item Metadata

Download

Media
focus-1.0115153.pdf
Metadata
JSON: focus-1.0115153.json
JSON-LD: focus-1.0115153-ld.json
RDF/XML (Pretty): focus-1.0115153-rdf.xml
RDF/JSON: focus-1.0115153-rdf.json
Turtle: focus-1.0115153-turtle.txt
N-Triples: focus-1.0115153-rdf-ntriples.txt
Original Record: focus-1.0115153-source.json
Full Text
focus-1.0115153-fulltext.txt
Citation
focus-1.0115153.ris

Full Text

Array Mlllllf ITTTTTTTT
INSTITUTE    FOR   COMPUlllG, INFORMATION   AND   COGNITIVE   S Y S T E M S   • I • C • I • C • S
THE      UNIVERSITY      OF      BRITISH      COLUMBIA
Sending the Right Signals
Matt Yedlin's ultra-wide band antenna has a
wide range of applications, from wireless data
transmission to landmine detection.
If you hear the strains of a cello drifting down from the rooftops of
Matt Yedlin's antenna lab, don't be surprised. His new ultra-wide
band antenna (UWBA) isn't the only thing sending signals from up
there. The new UWBA can emit pulses down to one billionth of a
second over huge bandwidths (up to 10 GHz), instead of the
narrow-bandwidth tones of conventional antennas. Operable from
100 MHz to 100 GHz, the UWBA has a range of over three orders
of magnitude. Previously, it took many antennas working together to
emit such pulses.
UBC has filed a patent application for the UWBA, developed by
Ph.D. student Kim Dotto, supervised by Yedlin.
Inside the newly constituted antenna range, which is one of the
few in Canada, Yedlin and Dotto are interpreting new data with a
fully functioning tomography apparatus used to image the interior of
solid objects. They have truly been resourceful, developing their
state-of-the-art antenna on a shoestring budget. To date, they have
spent less than three thousand dollars, a third of which went into a
computer data-acquisition card. To realize the potential of the
UWBA's rich application possibilities, Yedlin is seeking governmental
and commercial funding partners.
A diversity of applications
The UWBA has several powerful applications, including wireless
data transmission, non-destructive testing (sending energy in and
back through a structure to see what is inside, without affecting its
integrity), and ground-penetrating radar, including landmine
detection.
In the blossoming field of wireless telecommunications, the
UWBA enables interception and jamming-resistant transmission of
confidential data. The US Federal Communications Commission has
just approved the unlicensed low-power use of the spectrum above
ICICS member Matt Yedlin/'Cellist on the Roof.'
2.1 GHz, and while there is a project in Alabama also focused on this
application, Yedlin and Dotto's antenna provides a wider bandwidth.
An interdisciplinary researcher
The number of possible applications of the UWBA is a reflection
of Yedlin's diversity of interests. He holds an Honours B.Sc. in Physics
(he was the Lieutenant-Governor's gold medal winner in his graduating year), an M.Sc. in Neurophysiology and a Ph.D. in Theoretical
continued on page 7
Spring 2002 Vol. 13, No. 1
Safety in Complex Software Systems 3
Out of Mind - Out of Sight 4
It Shoots - It Scores 5
Working in Wood 6
Passing Notes 7
System-on-a-Chip Lab 8 In today's economic climate it is more
important than ever to acknowledge the
accomplishments and achievements of
our ICICS members. It is through their
research developments, collaborations
and industry partnerships that BC will
grow and thrive.
The spring issue of FOCUS features
the work of Jeffrey Joyce, who recently
returned to ECE from Raytheon, where
he led the safety analysis of the Canadian Automated Air Traffic System
(CAATS). Prior to Raytheon, Jeffrey was
an assistant professor in Computer
Science. We are happy to welcome him
back, as his work will have important
applications in the safety of all high-
integrity systems.
How we perceive and how we reason
form the basis of the work of Computer
Science faculty members Alan
Mackworth and Ron Rensink (joint
with Psychology). Alan and the research
group in the Laboratory for Computational Intelligence are world-leaders in
this exciting field. Ron has recently
secured $1.4 million from Nissan to
study human perception and attention.
We will also look at the work and
passions of Gary Schajer (ME) and Matt
Yedlin (ECE). Gary is an associate
member of the Centre for Advanced
Wood Processing Control in the Faculty
of Forestry. Matt has developed an
ultra-wide band antenna with applications in radar, telecommunications, and
tracking.
Michael Davies' term as head of ECE
will be profiled and we also have an
update on the opening of the new
System-on-a-Chip Laboratory headed
by Resve Saleh.
Rabab Ward, ICICS Director
Pro-Active-Michael Davies'Term as ECE Head
A new building,thriving research,and a commitment
to teaching and learning
Soon to complete a five-year term as head of Electrical and
Computer Engineering (ECE), Michael Davies is ready for a
long-overdue leave. Prior to being appointed department
head, he was associate dean of the Faculty of Applied
Sciences for ten years.
"It has been a stressful but very exciting period," says
Davies, "We have accomplished a lot, thanks to a great
group of faculty, administrative and technical staff." He cites
two critical and parallel issues the department has had to
contend with. The first is the burgeoning number of
students pursuing careers in electrical and computer en-
neering as the field becomes increasingly pertinent to all
aspects of life. Second, the demand for experts in the
private sector has made it more difficult to hire and
retain good staff.
r:;^—*»*
as ed"cators is to
as"'™'°ns of the young
meet the
P-p,eofBcandCanada
*»r. not because they aren,t
;-»med,butbecaJ    nt
**'* have enough p,aces/,
Faculty excellence
Davies says that since the economy has slowed
down, and the university has recognized that salaries
in ECE and Computer Science must be competitive,
the department has been able to fill vacancies with
excellent applicants. He also acknowledges the strong support ECE has received from alumni,
the Faculty of Applied Science, the Presidents Office, as well as the provincial government in
its plan to increase the number of places available in CS and ECE. "With the Dean's leadership
we have raised major development funds toward a new building, which will extend from the
north end of the existing MacLeod building."
Under Davies' term as head, research has been thriving. ECE members made a major
contribution to the ICICS proposal and its subsequent funding. The department is now
involved in six of the seven clusters. ECE has received five major CFI awards in addition to its
involvement with ICICS, and two Canada Research Chairs—with two more anticipated soon.
Michael Jackson received CFI funding for his work in advanced, high-speed microelectronics,
Tim Salcudean for multi-modal simulators and user interfaces, and Resve Saleh for his work in
System-on-a-Chip (see page 8). Robert Rohling and Babak Hamidzadeh have also received
CFI funding. Saleh was named NSERC-PMC-Sierra Industrial Research Chair and Peter
Lawrence has been awarded an NSERC-Design Chair.
Creative tinkering
In order to maintain the legacy of educational excellence, another challenge is to encourage
the passion for creative "tinkering" in undergraduates, says Davies. Students who entered
engineering in the past were hobbyists; they fixed their own cars, built their own stereos and
computers, and generally were much more practice-oriented. With the emphasis on grades and
competitive entry, engineering students today become very good in math and physics, but
sometimes at the expense of acquiring hands-on understanding.
"We have to inject a bit of curiosity back into the learning process," he says, noting that the
second-year Project Integration Program was designed to help students relate what they learn
academically to tangible engineering problems. "We also need to make sure that we are educating people in things that are appropriate for the coming decades and for the industries that will
give us prosperity in the future."
FOCUS Safety in Complex Software Systems
Jeffrey Joyce is redefining
safety in high-integrity software
systems and establishing
techniques to reduce risk.
Software systems perform an increasing
number of critical functions in our lives—many
we are not even aware of. We trust the software
in our cars to regulate the braking system, just as
we trust software to provide air traffic controllers with critical radar and flight data needed to
orchestrate the complex, three-dimensional web
of flight trajectories in the sky. "We depend on
computers to perform critical functions, yet take
it as a matter of course when our desktop
computer crashes," says Jeffrey Joyce, ICICS
member and professor of Electrical and
Computer Engineering.
Making dependable software
Joyce recently returned to UBC from
seven years with Raytheon, where he led the
safety analysis of the Canadian Automated
Ar Traffic System (CAATS) and a related
military system. Before joining Raytheon, he
was an Assistant Professor in Computer
Science at UBC. He notes that most of the
safety-related software produced by the
aerospace industry is developed using a more
rigorous, formal and expensive process than
those used to develop most other kinds of
software.
Joyce would like to see what he has
learned from the aerospace industry incorporated into the curriculum of the new Software Engineering Option offered in the
Department of Electrical and Computer
Engineering. "Not every student is going to
end up developing safety-critical software,"
he says. "However, many of the basic
principles could be scaled down and adapted
to benefit almost any kind of software
development."
A key step in the development of a safety-
related software system is the written specifications of its requirements. For a complex
system such as CAATS, this could easily
involve tens of thousands of sentences. A
significant challenge in the development of
such systems is minimizing ambiguity and
inconsistencies in the requirements, especially
when specifications are expressed in everyday, or "natural," language. Part of Joyce's
research program at UBC is concerned with
developing practical techniques and tools to
address this challenge. These include the use
of "lightweight" formal description techniques that strike a balance between mathematical rigour and readability for non-
mathematicians .
The other main focus of his research is
the safety analysis of software-intensive
systems. In mechanical, chemical, nuclear,
and other non-software systems, safety
analysis typically emphasizes component
failure. Unlike the physical components of
these systems, software components do not
crack, rust or wear out. Instead, hazards in
software systems can usually be traced back
to its requirements. With many thousands of
requirements in a system, pinpointing safety-
critical ones poses a considerable challenge.
Industry-University collaboration
While at Raytheon, Joyce was also an
adjunct professor at UBC, supervising three
PhD students and one MSc student. Three
of his PhD students, Nancy Day, Michael
Donat and Ken Wong, were involved in an
industry-university collaboration funded by
the Advanced Systems Institute, Raytheon,
and MacDonald Dettwiler. The project,
called FormalWARE, investigated the
industrial application of formal description,
tool-based analysis and other techniques in
the development of safety-critical systems.
"The Raytheon experience was fantastic
and generated a program of research that will
probably keep me busy for a good part of the
rest of my career," says Joyce. "Now I can take
what I have learned in industry and turn it
into knowledge that can be used more widely."
E-mail Jeff Joyce at jeffj@ece.ubc.ca or call
him at 604-822-7281.
Spring 2002 Out of Mind-Out of Sight
Ron Rensink recently secured $1.4 million from Nissan to study human perception and attention.
The automobile has come a long way since
the Model T Automatic transmission, cruise
control, air conditioning, and CD players
have added pleasure and convenience to
driving. But with more traffic, increased
speed, and cockpit-like dashboards loaded
with monitors and controls for everything
from RPMs to music volume—driving has
become increasingly complex. Not to
mention the distraction of cell phones, busy
intersections, flashing lights, and ubiquitous
advertising billboards. We have a lot to pay
attention to.
Too much in fact, says Ron Rensink,
ICICS member and assistant professor in
Computer Science and Psychology. And with
the addition of intelligent systems (like the
ones the Laboratory of Computational
Intelligence is helping to develop) cars will
soon be able to give more information to the
driver, such as the condition of the road
ahead, or the presence of a vehicle in a driver's
blind spot. One of the questions Rensink is
trying to answer is how much information is
too much?
Nissan partnership
Rensink's groundbreaking work on
change-blindness, undertaken at Cambridge
Basic Research (CBR) in collaboration with
MIT and Nissan, showed that if our attention is focused elsewhere, we simply don't see
that the car ahead has braked, or that a
pedestrian has stepped off the curb. When he
heard that Nissan wanted to fund a new lab
after the closure of CBR, he invited his
former colleagues to visit UBC in the hope of
securing some of the funding. Instead, he got
all of it. "Not only do we have some of the
best people in the world here, we work well
together and collaborate on projects," he says.
"In the US, the research environment is more
competitive, which makes effective multi-
disciplinary collaboration more difficult."
Nissan has committed $1.4 million to
research at UBC over the next 18 months.
The partnership currently involves engineers
at Nissan's Vehicle and Transportation
T.
"The car is a good test case for future technology, such
as wearable computers and smart homes, because it
involves many of the same issues of how humans and
machines can best interact."
Research Laboratory in Yokohama and five
UBC researchers: Rensink, Karon MacLean
from Computer Science, and Alan Kingstone,
Jim Enns, and Vince DiLollo from Psychology. In March 2002, the group will expand
to include researchers from Human Kinetics,
Linguistics and Philosophy. Much of the
funding will support graduate and postdoctoral researchers.
This type of industry-university collaboration—partnering very basic and very applied
research—is exactly what ICICS fosters.
These researchers will be investigating the
scientific basis of intelligent automobile
interfaces. One aspect of their work will be to
determine the limits of human perception
and cognition in order to establish the
effectiveness of an interface. They also want to
determine the extent to which new kinds of
sensory input, such as sound or touch, can be
used to convey information.
Mindsight
Rensink's work in visual attention has led
to other unexpected discoveries. Some people
can actually sense or "feel" change without
seeing it. The phenomenon, which Rensink
calls mindsight, is a conscious, non-sensory
experience: the "gut feeling" that something
is happening. In trials of flickering images,
subjects were asked to indicate when the
image changed. Such a large number of
participants asked if they should hit the space
bar when they felt something change as well
as saw it, that Rensink became intrigued. The
people who felt change were also able to
identify control trials, where images flicker
but don't change.
Too much information
Another effect he discovered was that we
can only attend to one thing at a time. Even
though we can keep track of four or five
things at once, we simply can't see two
changes occur simultaneously. So how do we
define units of attention? "These are all
abilities and limitations that we have to
consider when we are designing technology,"
says Rensink. He acknowledges that his work
is a bit metaphysical for some colleagues.
"When you present an idea that upsets
journal reviewers, it's a good sign that you are
on to something big."
E-mail Ron Rensink at rensink@cs.ubc.ca
or phone him at 604-822-0598.
FOCUS It Shoots-It Scores!
The Dynamics of Computational Intelligence
Soccer-playing robots developed by Alan Mackworth's
group can perceive, strategize, co-operate—and
score goals.
What defines intelligence? Linguists would
say language. Anthropologists would argue
the ability to use tools. Computer Science
Professor and ICICS member Alan
Mackworth believes that computation is the
essence of intelligence. As Canada Research
Chair in Artificial Intelligence and founding
director of UBC's Laboratory for Computational Intelligence (LCI), his opinion carries
the weight of considerable expertise.
For Mackworth and his LCI colleagues,
the primary characteristic of an intelligent
agent—whether human, animal, or robot—is
that it has purposes or goals and acts appropriately to achieve them. In other words, it is
proactive. It must also be reactive, by
responding appropriately to changes in its
environment. "If an agent exhibits both
proactive and reactive behaviour, then we
believe it meets the necessary and sufficient
conditions for intelligence," says Mackworth.
"The ability to use tools and language comes
about as a result of being proactive and
reactive."
Laboratory for Computational Intelligence
Although Mackworth admits that
everyone might not agree with this paradigm,
it is essential in designing computer systems
and robots that can perceive, reason, and
interact in dynamic environments in real
time. This is a far cry from earlier GOFAIR
The essence of control theory
is that you get feedbackfrom the
world and integrate that into
the reasoning process,"
says Mackworth.
robots developed as single agents in a static,
controlled environment. Traditional AI
concentrated on the proactive process, notes
Mackworth, whereas a reactive process, such
as cruise control, involves a control system.
The challenge is integrating AI and
control theory in a scientifically rigorous way.
Mackworth and colleagues started UBC's
program in AI and robotics in 1984 and it
was the first to be funded by the Canadian
Institute for Advanced Research (CIAR). It
evolved into the Institute for Robotics and
Intelligent Systems (IRIS), a national Networks of Centres of Excellence under the
umbrella of Precarn, a not-for-profit, industry-led consortium that helps to bridge the
"innovation gap" between basic research and
commercial applications.
Soccer-playing robots
The real test of computational intelligence
is in multi-agent systems, where the actions of
each agent—either human or machine—and
input from the environment affect reasoning
and decision-making. Take soccer for example.
With funding from IRIS, Mackworth and
colleagues developed soccer-playing robots.
The Dynamo project uses remote-control toy
vehicles for the physical agents, an off-board
camera and computers for perception,
reasoning and control. LCI's robot team "The
Dynamites" analyse the play 60 times per
second and convey visual information, such as
speed, direction, and position, back to the
computers' controllers, in effect, planning their
strategy as the game progresses. What started as
a lab experiment with two robotic trucks has
become an annual international event.
Robocup, sponsored by Sony, now has several
leagues and up to eleven robots per team.
Constraint Nets
All fun aside, these high-tech toys were
designed to test constraint-based theories.
The Dynamo robots run on Constraint Nets,
a software system developed by Mackworth.
The idea behind Constraint Nets is to think
of both the agent body and the controller as a
dynamical system, or a model of computing
where the output affects the input in a closed
loop. This involves integrating a hierarchy of
control systems from the robot-body and
continued on page 7
Laboratory of Computational Intelligence (LCI) Researchers
Christina Conati
Nandode Freitas
Holger Hoos
Jim Little
David Lowe
Karon MacLean
Dinesh Pai
David Poole
RobertWoodham
Adaptive
Probabilistic
Dynamic
Computational
Object
Haptic interfaces,
Reality-based
Preferences,
Computer vision,
interfaces,
machine learning,
constraint
vision, stereo and
recognition,
multisensory
modelling, multi
planning and
remote sensing
computer-aided
computational
optimization, local
motion
motion tracking
interaction
modal interfaces
probabilistic
instruction
statistics
search algorithms
systems
Spring 2002
5 Working
in Wood
Chunks of wood, antique saw blades and a
motley collection of frogs adorn Mechanical
Engineering Professor Gary Schajer's office.
His passion for the lumber industry began
over twenty years ago, when he worked with
Weyerhaueser Company in Tacoma after
finishing his PhD. His passion for frogs goes
back much further. Schajer began his research
and teaching career at UBC in 1989, and is
now director of the Renewable Resources Lab
and an associate member of the Centre for
Advanced Wood Processing in the Faculty of
Forestry. "Anything we can do to save raw
material has a huge payback," says Schajer.
Finding ways to reduce waste—and better
ways to use it—is the crux of his research.
Enhancing woodcutting processes
To keep sawdust and waste to a minimum,
industrial saw blades are made as thin as
possible. However, thin saw blades are
flexible and unstable, making straight cutting
more difficult. And what is saved in sawdust
is often lost in the planing of crooked boards.
Clearly, the business of woodcutting is far
from cut and dried. In fact, understanding
the details of the woodcutting process is an
extremely complex task, says Schajer, because
wood is a natural material with very high
variability. Differences in grain, density
moisture content, and knots make it difficult
to interpret experimental results and determine what aspects are significant.
Recent developments in sawing techniques have improved the amount of lumber
produced from raw logs. However, these
techniques are still not well understood.
Guided saws, in which the blade is secured
on either side, involve highly non-linear
interactions between the saw body, the saw
teeth and the wood, and these are much
more difficult to analyse. Schajer is working to
discover the real factors that determine the
operation of this type of saw blade in order to
explain its effectiveness and improve design.
Schajer is also developing laser sensors that
separately measure both top and bottom
surfaces of sawn lumber to better determine
With the BC lumber industry struggling,
Gary Schajer's research in process             ^tf^^fe
control is more critical than ever.
%F
*
^8                                   v*
"The biggest economic issue in the wood industry is
that about 75 percent of the cost is in raw material."
thickness and quality. Information about
individual surfaces helps to detect certain
types of machine problems, thereby enhancing quality control on both product and
machinery. And since more wood is planed
off the top of a piece than the bottom, it also
determines which side should be "up," again
reducing waste. In this project, Schajer is
working with Professors Tom Maness and
Rob Kozak of the Wood Science Department,
MASc student Javier Gazzarri, and industrial
partner Herniary Opto Electronics, a BC firm.
Particleboard: value-added waste
"The main feature of particleboard is that
it is made from sawmill floor sweepings," says
Schajer. Consequently, the material is
irregular, and cut surfaces are very rough and
prone to chipping—not attractive qualities in
a consumer item. The cost of finishing cut
surfaces can be greater than all other costs
combined. The challenge is to find better
ways to cut and process particleboard. This
involves determining the right mix of woods,
particle size, the best glue, and the optimum
temperature and pressure for processing.
Working with Schajer for his PhD studies
is Darrell Wong who is also a research
engineer with Fornitek Canada Corporation
and former assistant director of the Centre for
Advanced Wood Processing.
Measuring residual stress
Not all Schajer's work is in wood. He also
works on improved methods for measuring
residual stresses. These hidden stresses,
created during most manufacturing processes,
can seriously affect fatigue life and dimensional stability of a material—not something
you want to consider when boarding an
airplane. Residual stress is measured by
removing a small sample of material and
measuring the nearby strain changes. Schajer
developed a six-element strain gauge that is
included in the official ASTM standard test
method specification.
Schajer's challenge is finding a balance
between simplicity and realism. "I agree with
Einstein, who said 'theoretical models should
be as simple as possible, and no simpler.'"
E-mail Gary Schajer at schajer@mech.ubc.ca
or call him at 604-822-6004.
FOCUS Yedlin, continued from page one
Seismology. After a Stanford Geophysics Post
Doctorate, Yedlin went to Conoco Inc. as a
Research Geophysicist. In 1989, he received
a tenure track position at UBC, equally split
between ECE and Geophysics.
Yedlin observes that while the value of his
interdisciplinary training is clear, administratively the mechanisms have been less than
adequate. While many extol the value of
interdisciplinary research, they may not
realize that the process takes longer and that
such researchers are often thought of as "fence
sitters." Yedlin asserts that to have true
interdisciplinarity, a new paradigm needs to
be adopted to reduce the obstacles. Director
Rabab Ward expects that within ICICS, "...in
the future, researchers like Matt Yedlin will be
seen as the movers and shakers behind
collaborative research initiatives."
Philosophy
All professors deal with stress, and when
asked how he deals with his, Yedlin gives an
enigmatic grin. His colleagues know Matt
Yedlin as a smiling face. He enjoys teaching,
and earned the Killam Teaching Award in
1994. Married for twenty years, he regularly
attends ballroom dance classes at UBC with
his wife. He has enjoyed playing cello since
he was nine, and has a sixteen-year-old son.
"Know when to ask for help" is one piece of
advice he offers, but for new faculty Yedlin
recommends mentoring. He feels it is the key
to reaching a successful balance between
"being" (coping), "doing" (getting work
done) and "having" (achieving
start-up funding). Often the
"being" is merely assumed in
the rush to do and have.
Yedlin believes this is the root
of stress, and where mentoring
plays a vital role in a new
professor's success. His advice
to new faculty comes in the
form of a question: "Where are
you going to be ten years from
now?" Playing cello on the
roof of a very progressive
antenna lab is not a bad answer.
E-mail Matt Yedlin at matty@ece.ubc.ca
or call him at 604-822-8236.
I-GI-GS- Passing Notes
^   ^^^^^jl   Anne Condon (CS)
recently won the University College of Cork
JL '    '^L     distinguished Alumnus
Award for 2001—the
fci highest honour that
University bestows on an
alumnus. Anne is currently co-chair of the
Computing Research Association's Committee on the Status of Women in CS and ECE.
Dean of Science, Maria
Klawe (CS) received the
Science and Technology
Champion of the Year
award from the BC
Science Council for her
"outstanding contribution in promoting
and fostering BC Science and Technology."
Charles Laszlo (ECE) has won the Gold
Medal from the Canadian Council of
Professional Engineers.
Farrokh Sassani (ME) has been elected a
Fellow of the American Society of Mechanical Engineers.
Gary Schajer (ME) has been elected a Fellow
of the American Society of Engineers.
Vinod Modi (ME) is the first Canadian to
receive an Honorary Membership in the
Japanese Rocket Society.
Jose Marti (ECE) has
been elected a Fellow of
the IEEE for his work in
developing electromagnetic transient programs
I   for transmission line
modeling and real-time simulation.
MatthewW.Choptuik (Physics) received
the Rutherford Memorial Medal (Physics)
from the Royal Society of Canada.
Cristina Conati (CS), Sidney Fels (ECE),
Bryan Gick (Ling.), and Alan Kingston
(Psych.) received the Early Career Scholar
Award of the Peter Wall Institute In
Advanced Studies.
Peter Lawrence (ECE)
has been awarded an
NSERC Chair in Design
for his work in sensing
and signal processing for
telerobotic systems.
Janet Werker (Psych.)has been elected a
Fellow of the Royal Society of Canada.
The third annual PMC-Sierra Science Fair
Foundation Fun Run takes place Sunday
June 2, 2002. For registration and
information visit www.ScienceFairs.bc.ca,
or email scifair@scienceworld.bc.ca, or call
the Fun Run Hotline at 604-443-7552.
V
^^J
iBKIl'C
a
^Ju^^~* -  Hl fifc
on the roof: Matt Yedlin (c.) with graduate students Kim Lam (I.) and Kim Dotto (r.)
Mackworth, continued from page five
low-level motor control to high-level controllers. "The trick is that the constraints are changing while you are trying to solve them," he says.
Mackworth believes that a theory of computational intelligence will not only
help to build intelligent systems, but to understand human behaviour and
apply that knowledge to developing better technology.
"Understanding cognitive systems is extremely important," he says, noting
the work of ICICS colleague Ron Rensink. This research involves several ICICS
clusters, with collaborators in psychology and philosophy. "Constructing
intelligent models of the user is probably the best way to design computers that better support
human activity."
E-mail Alan Mackworth at mack@cs.ubc.ca or call 604-822-4893.
Spring 2002 System-on-a-Chip Lab makes media waves
UBC a world-leader in
microchip design
The high-profile launch of UBC's new
$3.2-million System-on-a-Chip (SoC)
Research Laboratory last November has put
ICICS researchers and Vancouver on the
map as world leaders in the evolution of
microchip design. Spearheaded by ECE
Professor Resve Saleh, it is currently home to
seven faculty members, two visiting professors, and twenty-five students. As major
players in Canada's System-on-a-Chip
Research Network, lab members are working
to streamline silicon chip design to keep pace
with the demands of technology.
One billion transistors
Currently, a microchip houses 20 to 30
million transistors. By the end of the decade,
designers want to be able to put over a
billion transistors on one chip. This technology is essential for developing new generation products such as mobile phones and
computers that use Bluetooth™ wireless
technology for personal area and internet
connectivity.
Funding partners for the lab include the
Canadian Microelectronics Corporation
(CMC), PMC-Sierra, the Canadian Foundation for Innovation (CFI), the BC Advanced
Systems Institute (ASI), the BC government's
Knowledge Development Fund, NSERC,
Micronet, and Prince Rupert's City Tel.
The SoC faculty members include (from left) Alan Hu of Computer Science, and David Pulfrey, Andre
Ivanov and Resve Saleh (all of Electrical and Computer Engineering). Members not present: Hussein
Alnuweiri, Mark Greenstreet and visiting researchers Dr. Yong Cho and Dr. Yong Lee.
The idea behind SoC is to build upon
previous designs so that each chip doesn't
have to be built from scratch—an expensive
and time-consuming process. The focus of
the lab's work is mixed-signal design,
verification and testing. SoC researchers have
recently embarked on a Bluetooth™
wireless chip design in partnership with
CMC andTality Corporation. "By analogy,
in order to build a new car engine you need
a car to put it into," says Saleh. "We are
essentially using this chip as a design
platform for our research."
A leading expert on SoC, Saleh is chair of
the technical advisory committee for CMC,
conference chair for the 2002 International
Symposium on Quality in Electronic
Design, and associate editor of IEEE. He has
fielded many interviews lately. At the time of
this one he was heading to Ottawa, to be
interviewed for an upcoming PBS documentary on SoC, narrated by Morley Safer.
Stay tuned for more news!
Res Saleh can be reached at
res@ece.ubc.ca or 604-822-3702.
>l*OI*OS* Institute for Computing, Information and Cognitive Systems www.icics.ubc.ca
UBC's Institute for Computing,lnformation and Cognitive Systems (ICICS) is an umbrella organization
that promotes collaboration between researchers from the faculties of Applied Science, Arts,
Commerce, Dentistry, Education, Forestry, Medicine, Pharmacy, and Science. ICICS supports the
collaborative computer-oriented research of more than 120 faculty members and over 500 graduate
students in these faculties. ICICS researchers attract approximately $ 15 million in annual grants and
contracts. Their work will have a positive impact on us all in the future.
Return Address:
ICICS, University of British Columbia
289 - 2366 Main Mall,Vancouver, BC,V6T 1Z4
CANADA
Writers:   Mari-Louise Rowley,
Pro-Textual Communications;
Kirsty Barclay, ICICS Technical
Programs Writing Advisor
Design:   William Knight, wilyum creative
Photos:   Janis Franklin, UBC Media Group
Greg Morton, UBC Media Group
Editor:   Kirsty Barclay
Office:   University of British Columbia
289-2366 Main Mall
Vancouver, BC,Canada,V6T 1Z4
Tel:   (604)822-6894
Fax:   (604)822-9013
E-mail:   info@icics.ubc.ca

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/cdm.focus.1-0115153/manifest

Comment

Related Items