1

EE

K / E

lectrical E

ngi n

ee

ri ng K

al ei d

osco

pe

20

02

The Electrical Engineering Department isthe largest department in the College ofEngineering at UW. We are dedicated tomaintaining an atmosphere of cooperationthat nurtures high-quality research andeducation, and which develops matureundergraduate and graduate engineers.Our department is in the midst of aperiod of dramatic growth and positivechange. We moved into a new building inFebruary 1998, and construction of anadjoining new CSE/EE building is underway.Since August 1998 we have grown throughthe addition of 14 outstanding new faculty.ALL of the assistant professors who havejoined the department from 1998-2001have received the NSF Early CareerDevelopment Award. Externally fundedresearch in the department is increasing ata rate close to Moore’s law—from $5million in 1998-1999 to approximately$12.6 million in 2000-2001, and over $14Mfor the first six months of 2001-2002.Our goal is to become one of the very topEE departments in the world, through thedelivery of outstanding and innovativeeducation and the conduct of cutting edgeresearch.

We are aggressively recruiting the verybest graduate students. If you are aprospective graduate student, I encourageyou to consider applying to our depart-ment. We have an extraordinary range ofnew and growing research projects thatprovide outstanding opportunities forgraduate education and professionalgrowth.

— Howard Chizeck,Professor and ChairDepartment of Electrical Engineering

E lectrical E ngineering K aleidoscope

C O N T E N T S W E L C O M E T O E E K 2 0 0 2

P U B L I S H E RElectrical Engineering Department of the University ofWashington

Howard Chizeck, ChairMari Ostendorf, Associate Chair for ResearchJohn Sahr, Associate Chair for Education

E D I T O R I A L S TA F F :Mary Ann Krug, Editor

D E S I G N S TA F F :Sarah Conradt, Designer

C O N T R I B U T I N G W R I T E R S :Mary Ann Krug, Rob Harrill, Tim Midgett, Howard Chizeck

E D I T O R I A L A S S I S TA N C E P R O V I D E D B Y :Tim Wade, Ann Fuchs, Bill Harris

Production by University of Washington Publications Services

©Copyright 2002, Department of Electrical Engineering,University of Washington

The opinions expressed by the contributors do not reflect officialpolicies of the University of Washington.

2002EEKA N N U A L R E S E A R C H R E V I E W

F E A T U R E D R E S E A R C H

E D U C A T I O N N E W S

P E R S P E C T I V E S

How Sweet the SoundScalable Audio Encoding

M.A. Krug 2

Powering Project NeptuneEE Researchers Will Develop and Test PowerInfrastructure for Seafloor Observatories

5

Life on A ChipThe Continuing Saga of the Human GenomeProject

M.A. Krug 8

Robotics Tools Teach SurgeryM.A. Krug 11

D E P A R T M E N T U P D A T E S

F.I.R.S.T. Robotics Competition6

Research Apprentices at FridayHarbor Help Devise Solutions toUnderstanding Brain Activity

10

ReflectionH. Chizeck 12

ED ITOR IAL CRED ITS

Milestones & Innovations16

Faculty Directory 14

By Popular Demand: RoboticsCourse Sequence Designedwith Students in Mind

7

NSF Grant Bolsters VLSI/CADTeaching 7

H O W S W E E TT H E S O U N D

2

EE

K /

Ele

ctri

cal

En

g in

ee

r in

g K

a le

i do

sco

pe

20

02

by Mary Ann Krug

Scalable audio encoding provides newoptions for listening, sharing music

3

EE

K / E

lectrical E

ngin

ee

ring K

aleid

osco

pe

20

02

H O W W E H E A RMusic lovers, take note: if fingernailsscratching a blackboard sound better toyou than music played over the internet,and the hefty cost of a high-speedconnection keeps you buying CDs, stopdespairing. A new method for processingaudio files could change the way weshare and enjoy music.

Professor Les Atlas and M.S. recipient Mark Vinton, both fromthe EE department’s Interactive Systems Design Laboratory,recently made a patent application for their new audio-encodingtechnique, which they describe as fine grain scalable audio encod-ing. It produces a high quality sound even at low bandwidths.

Their success with this novel approach to audio encoding is basedon a multi-university research initiative known as the Center forAcoustics and Auditory Research, funded through the Office ofNaval Research.

The objective of the collaboration focused on “how our earswork, and using that understanding to improve and advance otherareas of research,” says Atlas.

Audio encoding was one of the areas that the research initiativewas able to enrich. The popularity and convenience of sharingmany kinds of files over the internet broadened interest instreaming media for sharing music and video images. Unlike othertypes of files, such as text or html files that can stand some delaywhile being accessed, streaming media files must be continuallyaccessed during the download process.

Music must be encoded into digital format before it can be sentover the internet. As part of this signal transformation process,some qualities of the sound get lost. The qualities lost during thisprocess vary with encoding software. Most encoders placeinformation lost during the digitization step into frequencies thatare the least perceptible to humans.

Since the human auditory system is most sensitive around 3000Hz, this means lost information ends up in higher frequencies.When streaming media is sent over a typical phone line connec-tion, the information transmitted at higher frequencies gets lost.

The new encoding technique developed by Vinton and Atlas issimilar to MP3 files, but with one extra-and critical-step. That stepis called modulation frequency analysis, which orders the elementsof auditory recognition into a new level of abstraction.

“We pull sound apart ...from the slowest changes to the fastestchanges, and we take the changes [too fast] for us to hear anddon’t transmit them,” says Atlas.

The transmitted file ends-up devoting more content to the slowermodulations that are more important for human auditory recogni-tion, yielding music that sounds just a well with fewer bits. The

A sound wave travels in longitudinal fashion, displacing air

molecules as it travels. Some displacements cause air

molecules to cluster together, while some spread air

molecules apart, resulting in high and low pressure regions,

respectively.

When sound travels down the auditory canal to the

eardrum of the outer ear, high-pressure regions push the

eardrum inward, and low-pressure regions pull it outward.

This causes the eardrum to vibrate at the same frequency as

the sound wave. Our ability to detect frequency enables us

to distinguish the pitch of a sound.

Vibrations from the eardrum get transferred to the hammer,

one of three tiny bones located in the middle ear. These

bones serve as a system of levers that amplify and transfer

vibrations to the inner ear. When the vibration reaches the

stirrup, the smallest of the bones, the force acting on it

becomes nearly 15 times that acting on the eardrum itself.

This amplification allows us to hear very faint sounds.

The system of bony levers transmits vibrations to the inner

ear, via the oval window, to the cochlea. The cochlea is a

tiny coiled tube containing fluid divided into two chambers

of roughly equal size. Inside the cochlea, thousands of tiny

hair-like nerve cells exist, each varying slightly in size. The

variations in size impart sensitivity to particular frequencies.

When the vibrations travel through the cochlea, they

displace the fluid inside it, which causes the hair-like nerve

cells to move. When a nerve cell encounters a wave with a

frequency it is sensitive to, it responds by resonating with a

larger amplitude of vibration. The amplitude of a wave

allows us to distinguish the loudness of a sound.

This increase in amplitude is passed from the nerve cell to

the auditory nerve and on to the brain in the form of

electrical impulses. As these impulses travel up to the brain,

they pass through structures called the “auditorymidbrain.” Auditory physiologists are finding that the

concept of modulation frequency, where certain neurons

signal slow changes and others signal fast changes, is

common in the auditory midbrain.

4

EE

K /

Ele

ctri

cal

En

g in

ee

r in

g K

a le

i do

sco

pe

20

02

radio stations to broadcast programming that accommodatesusers with slower connections without sacrificing quality for userswith high-end connections and equipment.

“Scalability means that it’s going to do the best it can, no matterwhat the available bandwidth is,” explains Atlas. In other words asthe amount of bandwidth shrinks, the encoder lowers the modula-tion frequency rate, and transmits only the most importantelements for sound recognition. When more bandwidth is avail-able, portions of the music with higher modulation rates (and lessimportant auditory recognition qualities) are transmitted.

Other commercial possibilities include remote server access toaudio files. A user could keep all music files on one centralinternet-based server, and retrieve them not only with computers,but cellular phones, beepers, and PDAs (personal digital assistants).

Atlas has high hopes for the new encoder, and is currentlyexploring commercial possibilities.EEK2002

Editor’s note: Mark Vinton has graduated and taken a position with DolbyLaboratories.

The new encoding technique developed byVinton and Atlas is similar to MP3 files, butwith one extra-and critical-step. That step iscalled modulation frequency analysis, whichorders the elements of auditory recognitioninto a new level of abstraction.

result? A remarkably clear, enjoyable musical experience over adial-in connection.

The ability to provide CD quality sound with relatively littlebandwidth provides several improvements for existing technolo-gies, as well as new commercial applications.

Because bandwidth is not a critical limitation with this approach,an increase in channel capacity effectively results-users withbroadband and DSL connections could now enjoy music withoutinterruptions caused by increased connection traffic. The inherentscalability feature of the new technique enables internet-based

5

EE

K / E

lectrical E

ngin

ee

ring K

aleid

osco

pe

20

02

“Seafloor observatories present apromising, and in some cases essential,new approach for advancing basicresearch in the oceans.”

—National Research Council Report

UNDER THE SEA. Submerged volcanic systems support afragile microbial biosphere, hiding secrets to origins of life andeven the formation of the earth and other planets.

Scientists everywhere are realizing that to understand the mostcomplicated natural phenomena requires a cross-disciplinaryapproach to scientific investigation and discovery. Such realiza-tions are prompting shifts in operational paradigms everywhere,including the earth and oceanic sciences. These new intellectualpursuits require longer, more extensive observations of deep-seaenvironments and natural events.

Technological advances in the field of electrical engineering helpmake these types of studies possible, providing “an extensive,remote, continuous and interactive sensor presence” withinoceanic natural laboratories.

Such a presence requires a hefty undersea infrastructure.

A research team led by Bruce Howe and Tim McGinnis from theUniversity’s Applied Physics Laboratory, Harold Kirkham andVatche Vorperian from the Jet Propulsion Lab of the CaliforniaInstitute of Technology, and EE Professors Chen-Ching Liu andMohammed El-Sharkawi were recently awarded a 3-year, $2million award from the National Science Foundation for thedevelopment of a power system for cabled ocean observatories,specifically Project NEPTUNE (North East Pacific Time-SeriesUndersea Networked Experiments). Their goal is to design aprototype for the NEPTUNE power system, including a 3,000 kmfiber-optic/power cable that will lie on the sea floor.

The goal of NEPTUNE is to establish a submarine network ofremote, interactive laboratories for experiments and observationson the Juan de Fuca Plate off the Pacific Northwest Coast.NEPTUNE will connect webs of sensors in, on, and below the seato scientists, students, and the public on land. The NEPTUNEpower system will deliver approximately 100kW to distributedscience nodes on the seafloor with extremely high reliability. Theirdesign of the new power system will incorporate standardtelecommunications cable, regulated power supplies, and a multi-layered protection system. EEK2002

EE Researchers will Develop and Test PowerInfrastructure for Seafloor Observatories

NEPTUNE cable system map. NEPTUNE will providepower and communications bandwidth via fiber-optic/powercables to junction boxes (nodes) on the Juan de Fuca Plateand surrounding areas. By connecting sensor networks tothe nodes, with extensions onshore and offshore into theinterior, scientists will, for the first time, have the capabilityto study a host of interrelated processes with high spatialand temporal resolution over long periods of time.(Graphic: CEV)

P O W E R I N G P R O J E C T N E P T U N E

6

EE

K /

Ele

ctri

cal

En

g in

ee

r in

g K

a le

i do

sco

pe

20

02

F . I . R . S . T . R O B O T I C S C O M P E T I T I O N

For the third year, UW EE will take part in the FIRST (For Inspiration and Recognition of

Science and Technology) robotics competition. Most high school teams in this international

competition have advisors from industry-but UW engineering students serve as the

technical partners for our teams. On March 28th to 30th, UW will host the Pacific North-

west Regional competition. Approximately 34 teams will compete-including one from Brazil.

Last year students from UW and Bellarmine Preparatory High School designed Rainmaker

II, which finished 25th out of 331 at the National Competition in Orlando. This was better

than most university-backed entries. Our robot was dubbed “the six-wheeledwonder” on NASA TV. It received an award for “the most innovative locomotion

mechanism.” The faculty advisor of this activity, EE Assistant Professor Alexander Mamishev,

reports that the students “considered this enterprise a very valuable learning experience of

robotics, management, marketing, public relations, teaching, and life in general.”

This year, UW engineering students will team with students from Roosevelt High School in

Seattle. From the moment the rules are announced, the teams have only six weeks to

imagine, design and build a mammoth 130-pound remote-controlled robot. For this reason,

Team 824 calls itself: Students Working Against Time (S.W.A.T.) Robotics. For more informa-

tion, go to http://swat.ee.washington.edu/

Students working onRainmaker II.

7

EE

K / E

lectrical E

ngin

ee

ring K

aleid

osco

pe

20

02

Robotics course sequence designed with students in mindby Mary Ann Krug

B Y P O P U L A R D E M A N D :

“We wanted to give students the opportunity to build robots for a specific task,” saidaffiliate assistant professor Linda Bushnell of the newly designed EE course sequence inautonomous robotics.

Bushnell took up the challenge of modernizing the courses when she joined the depart-ment last year. A former U.S. Army Research Office Program Manager and former facultymember at Duke University, she approached the course design very pragmatically, seekingout and incorporating student feedback. While Bushnell admits experience in C program-ming is helpful, there are no formal prerequisites for her multi-disciplinary robotics coursesequence.

Instead of traditional exams, the courses have competitions between class teams. Studentteams build several robots for different purposes during the academic quarter. Each teammember is responsible for a particular aspect of the project, such as system integration,mechanical design, digital imaging, RF, or control systems.

The hands-on opportunity to build and design robots, along with interpersonal communi-cation and project management experience make the courses extremely popular, bothinside and outside of the department. Students from aeronautics and astronautics engi-neering, chemical engineering, mechanical engineering, computer science, and physicsfrequently enroll. According to EE student Dinh Bowman, the best thing about thecourses is that regardless of academic background students “get to work on amultidisciplinary project that is still in their area of interest.”EEK2002

The University of Washington Department of Electrical Engineer-

ing (EE) was recently awarded a $1.1 million NSF grant that will

bolster the research and teaching efforts of the VLSI/CAD group.

The group will use the funds to purchase computers, FPGA

boards, and other equipment for both VLSI/CAD teaching and

research laboratories.

“This is strictly an equipment grant, intended for [equipment

costs] not normally covered in other grants,” explained Associate

Professor Scott Hauck, the grant’s principal investigator. Co-

investigators include EE Professors David Allstot, Scott Dunham,

Carl Sechen, Mani Soma, and Gregory Zick, EE Associate Professor

Richard Shi, and CSE Professor Carl Ebeling.

The EE VLSI/CAD group teaches a large range of courses, including

a rigorous 3 course sequence: VLSI I, VLSI II, and VLSI III. Typical

enrollment is 120 students and remains high throughout the entire

N S F G R A N T B O L S T E R S V L S I / C A D T E A C H I N Gby Mary Ann Krug

sequence. On the teaching side, VLSI is the heaviest consumer of

computer resources in the department. In private industry the

computing and software resources provided would cost close to

$1 million per person.

The highly anticipated award required a substantial matching

contribution: approximately $290,000 will come from the univer-

sity itself.

Hauck is optimistic about the impact of the grant on the VLSI

teaching program. “This [grant] will give a greatly neededboost to the VLSI teaching program for the next 5 years,” he

said.

Autonomous robots in competition.

8

EE

K /

Ele

ctri

cal

En

g in

ee

r in

g K

a le

i do

sco

pe

20

02

L I F E O N A C H I P

Mary Ann Krug

The Continuing Saga of the Human Genome Project

Did last year’s announcement that a draftof the human genome was completeleave you wondering “Now what?”

If so, then read on...

Phot

os b

y Fe

ttah

Kos

ar.

9

EE

K / E

lectrical E

ngin

ee

ring K

aleid

osco

pe

20

02

CEGS also serve as springboards for cross-disciplinary intellectualactivity. At this CEGS, scientists from numerous life sciences andengineering disciplines collaborate. This CEGS currently draws onlife science experts from genomics, proteomics, microbiology andanalytical chemistry. It utilizes engineering experts from MEMS,nanotechnology, sensor fusion, detection, automation, and systemsintegration. EE faculty members Karl Böhringer, Denise Wilson,and Mark Holl currently collaborate with other CEGS scientists.

The feasibility of bringing this technology into existence is not sodistant as it seems. Like many other high-technology fields,advances in electrical engineering and computer technology aredriving progress in the Human Genome Project. Silicon chips thatmonitor cellular activity already exist (please see p.10 ). Meldrum’sown EE research lab, the Genomation Laboratory, has developedmicrosystems that enable automation of large-scale DNA se-quencing. The challenge will be to adapt existing areas of expertiseto solving new problems.

Meldrum believes that efforts of this CEGS are “extremely”compatible with research going on in the Seattle area, whichprovides a host of potential collaborators.

“We’re bringing together a lot of groups that normally don’t worktogether...as we move on and make progress, we want to bring inmore researchers, applications and technology,” explains Meldrum.EEK2002

EE Professor Deirdre Meldrum (PI) and Associate Dean MaryLidstrom (co-PI) received a $15 million award to start a Center ofExcellence in Genomic Science (CEGS0 from the NationalInstitutes of Health (NIH) National Human Genome ResearchInstitute (NHGRI).

“It’s a lot of money, but we have a lot to do,” says Meldrum.

This CEGS grant is one of three that were awarded nationally,becoming part of the Human Genome Project (HGP) during itssecond phase, 1998-2003. The goals of the first phase, including a‘draft’ sequence of the human genome, were accomplished ontime or ahead of schedule. The full sequence, scheduled forcompletion in 2003, promises to give scientists massive amountsof raw data.

Data that needs to be digested, analyzed, and put to use.

Researchers want to correlate human sequence information withimportant biological processes, so that complicated activities suchas cell proliferation, differentiation, programmed cell death(apoptosis), and disease progression and development (pathogen-esis) can be thoroughly understood.

Current research protocols involve large cell populations contain-ing individuals that vary greatly. When researchers measure cellularactivity of a population, they essentially observe the averageactivity of its individual cells. This accounts for why biopsies andpap smears sometimes come out normal when they actually aren’t—these tests use non-random samples of heterogeneous popula-tions. Researchers need to account for the differences in activitiesof individual cells to understand the behavior of an entire popula-tion.

Since complex biological processes involve so many differentevents over time, the ability to study multiple activities of indi-vidual cells is needed. Such deep levels of understanding requiretechnological capabilities that currently do not exist.

The CEGS co-directed by Meldrum and Lidstrom focuses ondeveloping and enabling technology for the study of individualcells. Specifically, the center will build devices that detect rare cellswithin populations and perform real-time analysis of processessuch as metabolism in individual cells.

These technological capabilities will take the form of individualmodules designed to measure activity for single cells. Imagine ahypothetical protein X, associated with progression of a certaindisease. A researcher might need to detect cells that produce theprotein, and monitor activities associated with it. These activitiesmight include DNA transcription, RNA translation, and numerousother factors before, during, and after the cell starts makingprotein X. The Center will design a module to study each of theseevents in real-time: life-on-a-chip.

According to Meldrum, the center intends to package theseindividual modules into kits, so researchers can select the charac-teristics they want to study and ‘mix-and-match’ the right modulesto suit their needs.

Researchers want to correlate human sequenceinformation with important biological processes,so that complicated activities such as cell prolif-eration, differentiation, programmed cell death...,and disease progression and development... canbe thoroughly understood.

Mary Lidstrom (L) and Deirdre Meldrum co-directthe College's new Microscale Life Sciences Center.

10

EE

K /

Ele

ctri

cal

En

gin

ee

rin

g K

ale

ido

sco

pe

20

02

Research Apprentices at Friday Harbor HelpDevise Solutions to Understanding Brain Activity

Mary Ann Krug

of Engineering Denice Denton, an expertin microfabrication and biocompatibilityissues. Zoology Professors DennisWillows and Thomas Daniel contributedtheir expertise on the animal species usedin the project, while Computer Scienceand Engineering Associate Professor ChrisDiorio contributed to chip design. Amongthe animal species selected to receiveimplantable sensors was Tritonia diomedea,a sea slug native to the Pacific Northwest.The selection of this animal was due to theextensive amount of knowledge thatalready exists about it; Willows has studiedthese creatures for years.

Conventional attempts to study brainactivity involved sensors strapped toimmobilized specimens, which preventedresearchers from gathering authentic dataon natural behavior. The sensors designedin this project are intended to be perma-nently implanted in animals, allowing neuraldata collection during normal activities.Unlike conventional sensors, which piercecells with glass electrodes, the electrodesof the new chips are made of silicon.Researchers hope that the silicon elec-trodes will hold the sensors in place, sothat the animals can be released into theirnatural environment for study. EEK2002

The inability of scientists to directly observe brainactivity remains a major obstacle to an empiricallybased understanding of behavior.

This spring, EE undergraduates NathanielJacobson and Jai Patel participated in amultidisciplinary research project aimed atproviding new tools for solving thischallenging research problem.

The goal of the project, which took placeat the world-renowned UW Friday HarborResearch Facility, was ì to build a complete,stand-alone implantable microsystem forrecording neural activity in freely behavinganimals.î EE faculty participation includedAssistant Professor Karl Bˆ hringer, aMEMS expert, and EE Professor and Dean

Above: A silicon chip implanted in theneural ganglia of Tritonia diomedea.

Top: Tritonia diomedea, a seaslugnative to the Pacific Northwest andthe model organism under study.

11

EE

K / E

lectrical E

ngin

ee

ring K

aleid

osco

pe

20

02

It looks like aerobics class, except theseparticipants only move their hands. Theinstructor makes a rotating motion withhis wrist and forearm, and his only pupilcopies it closely.

Closely, but not perfectly.

The instructor shakes his head, andrepeats the motion. The pupil concentratesharder, trying to mimic the movementexactly. When he succeeds, the instructorstops and nods quickly. The pupilrepeats the motion correctly one moretime, and nods slowly in understanding.

This exchange occurred during a meetingbetween members of EE Professor BlakeHannaford’s research team, including Prof.Jacob Rosen and graduate student JeffBrown, and medical surgical residents whotrain under Dr. Mika Sinanan in the UWMedical School Department of Surgery.The two groups are collaborating ondeveloping instrumentative surgical toolsand algorithms for objectively evaluatingsurgical skills that will improve theteaching of Minimally Invasive Surgery(MIS) to surgical fellows.

(MIS) has been practiced on a large scalefor about 10 years in the United States.MIS replaces traditionally more invasiveprocedures for common operations suchas gall bladder removal, and providestremendous benefits for patients. Unliketraditional surgery, which involves makinglarge incisions in the patient to accommo-date the surgeons hands, MIS incisionsconsist of small ports, through which longtools and a camera are inserted. Becauseof these smaller incisions, patient recoverytimes are much shorter: 1-2 days insteadof 1-2 weeks with traditional surgery. Theshorter recovery times and decreasedincidence of complications result inreduced healthcare costs.

The differences between MIS and moretraditional invasive techniques present aunique set of challenges for trainingsurgeons. In MIS procedures, surgeons lackdirect physical contact with patients,

R O B O T I C S T O O L S T E A C H S U R G E R Y

making it difficult to gauge the appropriateamount of force and torque to applyduring the operation. Surgeons also lack adirect line of sight, watching their progressthrough images projected onto a televisionfrom a tiny camera inserted into thepatient.

Consequently, teaching by expert surgeonsnecessarily becomes more abstract, andevaluations of student progress moresubjective. Currently expert surgeonsevaluate progress by commenting onvideotapes of procedures done by youngsurgeons. Still another challenge is distin-

Most improvement in technique wasachieved during the first 2-3 years ofthe 5-year surgical residency. Afterthat point, technical progress in-creases less dramatically and cogni-tive skills develop more fully.

The Blue Dragon, a device for measur-ing the properties of endoscopic toolsand for objectively evaluating asurgeon's performance.

guishing between technical skills andcognitive development of young surgeons.For example, if a procedure has 30 steps init, and an inexperienced surgeon is havingproblems completing the operationeffectively, is it due to a lack of surgicalskill, or is it because they have a hard timeremembering the precise sequence ofsteps 21-24? Current training techniquesmake it difficult to evaluate these kinds ofquestions.

“We started out designing special surgicalinstruments that measure the forces asurgeon is applying during the operation,”says Hannaford. The sensors on theinstruments collect large amounts of dataon the mechanical forces exerted by thesurgeons. The data is evaluated usingstatistical techniques, including HiddenMarkov modeling.

Hannaford’s team collected data fromexpert and inexperienced surgeons, whooperated on an animal model system, andcreated a quantitative basis for comparingtheir respective skill levels. Comparingdata generated by surgeons of differentlevels of expertise provides a moreobjective method of evaluating skill leveland progress.

Their preliminary data revealed someinteresting facts about development ofsurgical skills. Most improvement intechnique was achieved during the first 2-3years of the 5-year surgical residency.After that point, technical progressincreases less dramatically and cognitiveskills develop more fully. The EE research-ers and their collaborators in the depart-ment of surgery are currently planning alongitudinal study of surgical residents inthe department of surgery.

Editor’s Note: Blake Hannaford, Jacob Rosen, andMika Sinanan recently received a 4-year, $1.4million grant to develop minirobots for telesurgeryfor battlefield surgery.

12

EE

K /

Ele

ctri

cal

En

g in

ee

r in

g K

a le

i do

sco

pe

20

02

Reflectionby Howard Chizeck

We know now the “real” starting date of the twenty-first century. Theterrorist attacks of September 11, 2001 changed our nation. Someexperienced personal grief and pain. Most were shocked and shaken bya sudden loss of safety, by our vulnerability and the realization thatsome people want to kill us badly enough to sacrifice their own lives inthe process.

On 9/15 a memorial was scheduled atthe Seattle Center to honor the victimsof the 9/11 attacks. There were nospeeches or formal programs. Crowdsbrought flowers and stood vigil for 3days and 3 candle-lit nights, accompa-nied by flutes, native drummers, Spanishguitars, and funeral trumpets. The vigilwas scheduled for only 3 hours. (Photoby Carson Jones)

Initial responses to these events werefollowed by attempts to understand themeaning of the attacks and their conse-quences. We may not yet know how theseevents will change our individual lives andcommunities, but there has clearly been achange in national perceptions and inresource allocations. Any major change inperceptions and resource allocations is aturning point for our society.

We have been thrust into war by forcesopposed to the globalization, free ex-change of ideas and societal change thatarise from technology. The use of commer-cial airliners and passengers as weapons ofmass destruction was directed against akey component of global civilization-the airtravel system. The initial attempt againstLAX (aborted by quick thinking by borderguard in Port Angeles, Washington), thedestruction of 9/11, and the later attemptto use a shoe bomb to murder thepassengers of a trans-Atlantic airliner wereall attacks against the rapid movement ofpeople across national and culturalborders.

The attack on the twin towers was also anattack on the system of world trade. It wasmeant to permanently impair the worldeconomy. Ironically, our often-questionedy2k preparation greatly mitigated thesystemic effects of this violence. Brokeragehouses and financial institutions that werephysically destroyed in the World TradeCenter were quickly returned to opera-tion because of backup and recoveryprocedures that were developed for y2k.

We are engineers. Our business is knowledge,understanding, reason, and problem solving.How can we constructively respond to thischanged world?

Attention has turned to technologicalpreventions, responses and counter-measures to terrorism. Many involveElectrical Engineering. Of course, the verybest technology is useless if the trainingand management of those who use it isinadequate.

Sensor technologies under developmentto detect food-borne bacteria and virusesmay be applicable for the rapid detectionof bioweapons. Other sensors may identifychemical weapons, explosives and radioac-tive threats. Recent innovations ingenomics and proteomics have thepotential to facilitate rapid vaccinedevelopment, and may yield the broad-spectrum antiviral agents that will beneeded to protect our population fromunexpected attacks by biological weapons.

To prevent future use of civilian aircraft asweapons, a multitude of ideas for researchand development have been proposed,including: hardened and alarmed cockpitdoors; remote control of aircraft; emer-gency broadcast of cabin and cockpitinformation and images; video monitoringof planes while on the ground; and securityscanning of passengers and luggage.

Similar protective measures and responsescan be developed for cargo transport.Detection of threats in shipping contain-ers, transponders and satellite-based

13

EE

K / E

lectrical E

ngin

ee

ring K

aleid

osco

pe

20

02

monitoring systems, remote control andother technologies will help to prevent theuse of ships and trucks as weapons ofmass destruction.

Research in computer face-matching,retinal and palm print recognition andDNA-based identification technologies allhave the potential to avert future terroristacts. But these technologies, as well asdata-mining of credit and cash transactioninformation, monitoring of computer andcommunication systems and other newsurveillance technologies raise challengingquestions regarding civil liberties andprivacy.

technologies will be accelerated. In thelonger term, we must apply our skills toglobal human problems where we canmake a difference. Victory in this war ofthe new millennium will ultimately restupon our success in attacking the criticalproblems that generate despair on somuch of our planet: the lack of economicsecurity and opportunity, health andnutrition, environmental quality, under-standing and tolerance of cultural differ-ences, and individual freedoms.

We face new challenges for engineeringeducation. There is critical need to teachour students the fundamentals of systemssurvivability. Education in engineeringdesign methods must address the robust-ness and resilience of systems. In addition,the accreditation-mandated educationalpriority of “Engineering Ethics” now has anexpanded meaning.EEK2002

Attention has turned to techno-logical preventions, responses andcounter-measures to terrorism.Many involve Electrical Engineering.

Advances in wireless and optical communi-cations hardware and network design canenhance communication and data networksecurity, reliability and resilience. We werefortunate that the attacks of 9/11 werenot coordinated with attacks on ourInternet infrastructure, communicationssystems, electric power and natural gasdistribution, or our ground transportationinfrastructure. The robustness and stabilityof these complex networks is now acritical issue. Recent trends toward “just intime” manufacturing and low inventorylevels have heightened our vulnerabilitiesto disruption.

Beginning with World War II, Americanuniversities have been called upon to servethe national defense. The research andeducational programs of engineeringcolleges have been profoundly shaped bythe technological priorities of the military.Once again we are called to service. In thenear term, the development of certain

14

EE

K /

Ele

ctri

cal

En

g in

ee

r in

g K

a le

i do

sco

pe

20

02

Afromowitz, MartinProfessorMicrotechnology/SensorsPh.D., 1969 Columbia UniversityNIH Career Development Award

Aggoune, MohammedAffiliate Assistant ProfessorPower SystemsPh.D., 1988 University of Washington

Allstot, DavidProfessorVLSI and Digital SystemsPh.D., 1979 University of CaliforniaIEEE Fellow, IEEE Circuits & Systems SocietyDarlington Award, IEEE W.R.G. Baker Award

Atlas, LesProfessorSignal and Image ProcessingPh.D., 1984 Stanford UniversityNSF Presidential Young Investigator Award

Bilmes, JeffAssistant ProfessorSignal and Image ProcessingPh.D., 1999 UC-BerkeleyNSF Career Award

Böhringer, KarlAssistant ProfessorMicroelectromechanical Systems (MEMS)Ph.D., 1997 Cornell UniversityNSF Career Award

Bushnell, LindaAffiliate Assistant ProfessorControls and RoboticsPh.D., 1994 UC-Berkeley

Chen, Tai-ChangResearch Assistant ProfessorMEMS and MicrofabricationPh.D., 1997 University of Washington

Chinowsky, TimResearch Assistant ProfessorSensors, Instrumentation, Analog ElectronicsPh.D., 2000 University of Washington

Chizeck, Howard JayProfessor/Chair EEControls and RoboticsSc.D.., 1982 MITIEEE Fellow

Christie, RichAssociate ProfessorEnergy SystemsPh.D., 1989Carnegie Mellon UniversityNSF Presidential Young Investigator Award

Crum, LawrenceResearch ProfessorMedical UltrasoundPh.D., 1967 Ohio University

Dailey, DanResearch Associate ProfessorIntelligent Transportation SystemsPh.D., 1988 University of Washington

Damborg, MarkProfessorEnergy SystemsPh.D., 1969University of Michigan

Darling, R. BruceProfessorDevices, Circuits and SensorsPh.D. 1985Georgia Institute of Technology

Denton, DeniceProfessor/ Dean of EngineeringMicroelectromechanical Systems (MEMS)Ph.D., 1987 MITNSF Presidential Young Investigator Award,Harriet B. Rigas AwardAAAS Fellow

Dunham, ScottProfessorMaterials and DevicesPh.D., 1985Stanford University

El-Sharkawi, MohammedProfessorIntelligent Systems and EnergyPh.D., 1980 University of British ColumbiaIEEE Fellow

Hannaford, BlakeProfessorBioroboticsPh.D., 1985University of California, BerkleyNSF Presidential Young Investigator Award,IEEE EMBS Early Career Achievement Award

Hauck, ScottAssociate ProfessorVLSI and Digital SystemsPh.D., 1995 University of WashingtonSloan Research FellowshipNSF Career Award

Helms, WardAssociate ProfessorCircuits and SensorsPh.D., 1968 University of Washington

Holl, MarkResearch Assistant ProfessorMEMS. Micro-total analytical systemsPh. 1995 University of Washington

Homola JiriResearch Associate ProfessorPhotonic Devices & Optical SensorsPh.D., 1993 Academy of Science of the CzechRepublic

Hwang, Jenq-NengProfessorSignal and Image ProcessingPh.D., 1988 University of Southern CaliforniaIEEE Fellow

Jandhyala, VikramAssistant ProfessorElectromagnetics, Fast Algorithms, DevicesPh.D., 1998 University of IllinoisNSF Career Award

Kim, YongminProfessor/Chair of BioengineeringDigital Systems, Image Processing & MedicalImagingPh.D., 1982 University of Wisconsin-MadisonIEEE FellowIEEE EMBS Early Career Achievement Award

Kirchhoff, KatrinResearch Assistant ProfessorMultilingual Speech Processing, MachineLearningPh.D., 1999 University of Biefeld

Kuga, YasuoProfessorElectromagneticsPh.D., 1983 University of WashingtonNSF Presidential Young Investigator Award

Li, QinResearch Assistant ProfessorElectromagneticsPh.D., 2000 University of Washington

Liu, Chen-ChingProfessor & Associate DeanPower SystemsPh.D., 1983 UC-BerkeleyNSF Presidential Young Investigator, IEEE Fellow

Liu, HuiAssociate ProfessorCommunications and Signal ProcessingPh.D., 1995 University of Texas, AustinNSF Career Award, ONR Young Investigator

Mamishev, AlexanderAssistant ProfessorElectric Power Systems, MEMS, SensorsPh.D., 1999 Massachusetts Institute ofTechnologyNSF Career Award

Marks, RobertProfessorSignal and Image ProcessingPh.D., 1977 Texas Technical UniversityIEEE Fellow, OSA Fellow

McMurchie, LarryResearch Assistant ProfessorVLSI and Digital SystemsPh.D., 1977 University of Washington

Meldrum, DeirdreProfessorGenome AutomationPh.D., 1993 Stanford UniversityPresidential Early Career Award (PECASE),NIH SERCA Award

Nelson, BrianResearch Associate ProfessorPlasma PhysicsPh.D., 1987 University of Wisconsin Madison

Ohlson, JohnAffiliate ProfessorCommunicationsPh.D., 1980 Stanford University

E E F A C U L T Y

Ostendorf, MariProfessor & Associate Chair for ResearchSignal and Image ProcessingPh.D., 1985 Stanford University

Peckol, James S.Full-time LecturerDigital SystemsPh.D., 1985 University of Washington

Poovendran, RadhaAssistant ProfessorCommunications Networks & SecurityPh.D., 1999 University of MarylandNSF Rising Star Award, NSF Career Award

Ramon, CeonResearch Scientist, LecturerElectromagnetics & Remote SensingPh.D., 1973 University of Utah

Riskin, EveAssociate ProfessorSignal and Image ProcessingPh.D., 1990 Stanford UniversityNSF Presidential Young Investigator, SloanResearch Fellowship

Ritcey, JamesProfessorCommunications and Signal ProcessingPh.D., 1985 UC-San Diego

Rosen, JacobResearch Assistant ProfessorBiorobotics, Human-Machine InterfacesPh.D., 1997 Tel-Aviv University

Roy, SumitAssociate ProfessorCommunications and NetworkingPh.D., 1988 UC-Santa Barbara

Sahr, JohnAssociate Professor & Associate Chair forEducationElectromagnetics and Remote SensingPh.D., 1990 Cornell UniversityNSF Presidential Young Investigator AwardURSI Booker Fellow URSI Young ScientistAward

Sechen, CarlProfessorVLSI and Digital SystemsPh.D., 1986 University of California, BerkleyIEEE Fellow

Shapiro, LindaProfessorSignal and Image ProcessingPh.D., 1974 University of IowaIEEE Fellow, IAPR Fellow

Shi, C.J. RichardAssociate ProfessorVLSI and Digital SystemsPh.D. 1994 University of WaterlooNSF Career Award

Soma, ManiProfessorMixed Analog-Digital System TestingPh.D., 1980 Stanford UniversityIEEE Fellow

Spindel, Robert C.ProfessorSignal Processing, Ocean AcousticsPh.D., 1971 Yale UniversityIEEE Fellow, ASA Fellow, MTS Fellow, A.B. WoodMedal, IEEE Oceanic Engineering SocietyTechnical Achievement Award

Strunz, KaiAssistant ProfessorElectric Power Systems &Power ElectronicsPh.D., 2001 Saarland University

Sun, Ming-TingProfessorSignal and Image ProcessingPh.D. 1985 UC-Los AngelesIEEE Fellow

Thorsos, EricResearch Associate ProfessorSurface Scattering, Underwater AcousticsPh.D., 1972 MIT

Troll, MarkResearch Associate ProfessorBiophysical/Electronic Devices, OpticsPh.D., 1983 UC-San Diego

Tsang, LeungProfessorElectromagnetics and Remote SensingPh.D., 1976 Massachusetts Institute ofTechnologyIEEE Fellow, OSA Fellow

Wilson, DeniseAssociate ProfessorCircuits and SensorsPh.D., 1995 Georgia Institute of TechnologyNSF Career Award

Winebrenner, DaleResearch Associate ProfessorElectromagnetic & Remote SensingPh.D., 1989 University of Washington

Yee, H.P.LecturerElectric Power Systems & Power ElectronicsPh.D., 1992 University of Washington

Yee, Sinclair S.ProfessorCircuits and Sensors, PhotonicsPh.D., 1965 UC-BerkleyIEEE Fellow

Zick, GregoryProfessorVLSI and Digital SystemsPh.D., 1974 University of Michigan

E M E R I T I

Alexandro, FrankControls & RoboticsPh.D., 1961 University of Michigan

Albrecht, RobertControls & RoboticsPh.D., 1961 University of Michigan

Andersen, JonnyCircuits & SensorsPh.D., 1965 MIT

Bjorkstam, John L.Controls & RoboticsPh.D., 1958 University of Washington

Clark, Robert N.Ph.D., 1969 Stanford UniversityIEEE Fellow

Dow, Daniel G.Ph.D., 1958 Stanford University

Guilford, Edward C.Ph.D., 1960 UC-Berkeley

Haralick, RobertSignal & Image ProcessingPh.D., 1969 University of KansasIEEE FellowIAPR Fellow

Hsu, Chih-ChiPh.D., 1957 Ohio State University

Ishimaru, AkiraElectromagnetics and Waves in Random MediaPh.D., 1958 University of WashingtonIEEE Fellow, OSA Fellow, IOP Fellow, IEEEDistinguished Achievement Awards:Geosciences and Remote Sensing Society;Antennas & Propogation Society

Johnson, David L.Ph.D., 1955 Purdue University

Lauritzen, Peter O.Power ElectronicsPh.D., 1961 Stanford University

Lytle, Dean W.Ph.D., 1957 Stanford University

Meditch, James S.Ph.D., 1969 Purdue University

Moritz, William E.Ph.D., 1969 Stanford University

Noges, EndrikPh.D., 1959 Northwestern University

Peden, IrenePh.D., 1962 Stanford UniversityNSF “Engineer of the Year”, Member, NationalAcademy of Engineering, IEEE Fellow, IEEEHarden Pratt Award, U.S. Army OutstandingCivilian Service Medal

Porter, Robert B.Ph.D., 1969 Northeastern UniversityASA Fellow, OSA Fellow

Redeker, Charles C.Ph.D., 1964 University of Washington

Sigelmann, Rubens A.Ph.D., 1964 University of Washington

15

EE

K / E

lectrical E

ngin

ee

ring K

aleid

osco

pe

20

02

We apologize for any errors, omissions ormisspellings in 2002 EEK. We would like toextend special appreciation to the facultyand staff who assisted in producing thispublication and to the sponsors whosegenerosity made it all possible.

Congratulations to those faculty members who recently received

promotions. Scott Dunham and Deidre Meldrum were promoted

to full professor; Denise Wilson, Hui Liu, and Richard Shi are now

associate professors with tenure. Jiri Homola was recently

promoted to research associate professor. The department

welcomes our newest faculty, assistant professor Kai Strunz in

April 2002.

EE announces changes in leadership roles: Mari Ostendorf is

replacing Les Atlas as Associate Chair for Research; John Sahr

becomes Associate Chair for Education, taking over from Blake

Hannaford; R. Bruce Darling takes over from John Sahr as Gradu-

ate Coordinator; Eve Riskin becomes our first Undergraduate

Research Coordinator.

EE undergraduate students Rejo Jose, Hans Isern, and Jeff Chen

received the “AT&T Labs Student Enterprise Award,” given by IEEE

through AT&T labs and the AT&T foundation to support IEEE

student branch programs. Their project, “High Voltage Energization

Status of Underground Cables” safely determines the energization

state of underground cables using non-intrusive methods.

Associate Professor Jeff Bilmes received a $650,000 equipment

donation from IBM, which includes 10 RS/6000 44P Model 270

workstations. The machines will be used for computationally

intense forms of computer speech recognition.

16

EE

K /

Ele

ctri

cal

En

g in

ee

r in

g K

a le

i do

sco

pe

20

02

L A T E S T C A R E E R A W A R D R E C I P I E N T S

A R E I N G O O D C O M P A N Y

Congratulations to our latest NSF CAREER award winners!

Jeff Bilmes, Alex Mamishev, Radha Poovenderan and Vikram

Jandhyala are the latest EE faculty members to receive this

prestigious award. These new recipients are in good company: all 9

EE assistant professors hired from July 1998 through June 2001 are

recipients of this national award, which is given by the National

Science Foundation (NSF) in support of overall career develop-

ment of young scientists. It combines extensive, broad based

support for research and education of the highest quality. This level

of support exemplifies the importance that the National Science

Foundation (NSF) places on the early development of academic

careers dedicated to stimulating the discovery process in a

research environment enhanced by enthusiastic teaching and

enthusiastic learning.The IEEE has announced that Professor Carl Sechen has been

named an IEEE Fellow, effective January 2002. His citation reads

“For contributions to automated placement and routing in integrated

circuits.” The UW EE Department now has 19 IEEE Fellows. A list,

including their citations, appears at http://www.ee.washington.edu/

welcome/IEEE_Fellows.htm.

Congratulations to our recently appointed Professors Emeriti:

Robert Albrecht, Frank Alexandro, Jonny Andersen and Robert

Haralick.

Congratulations to Sang-Il Lee and Sermsak Jaruwatanadilok who

took 2nd and 3rd prize at the recent International Union of Radio

Science US National Meeting in Boulder CO. Mark Curry, who

took 2nd prize the previous year. Half of all the students winning

these URSI awards in the United States for the last two years have

been from our department.

We are pleased to report that our new building performed

admirably during the 6.8 magnitude earthquake of February 28,

2001. There was no significant damage and there were no injuries.

The old EE building did suffer damage, but not enough to acceler-

ate its demolition one month later (as planned). That site will

soon be the Paul G. Allen Center for Computer Science &

Engineering. The building will provide 75,000 square feet of new

space for the Department of Computer Science & Engineering and

10,000 additional square feet of space for the Department of

Electrical Engineering.

The Electrical Engineering department proudly congratulates its

2001 doctoral degree recipients: Al-hussein Abou-zeid, Ph.D., Selim

Aksoy, Ph.D., Hamed Alazemi, Ph.D., Supavadee Aramvith, Ph.D.,

Mohammed Azadeh, Ph.D., George Barrett, Ph.D., Mark

Billinghurst, Ph.D., Todd Chauvin, Ph.D., Aik Chindapol, Ph.D.,

Timothy Chinowsky, Ph.D., Georgios Chrysanthakopoulos, Ph.D.,

Mark Curry, Ph.D., Giri Devarayanadurg, Ph.D., Randall Fish, Ph.D.,

Jihun Joung, Ph.D., Jae-Byung Jung, Ph.D., Changick Kim, Ph.D.,

Seongwon Kim, Ph.D., Gang Liu, Ph.D., Ravi Managuli, Ph.D., Desik

Echari Nadadur, Ph.D., Garet Nenninger, Ph.D., David Palmer,

Ph.D., John Rockway, Ph.D., Kalev Sepp, Ph.D., Tatjana Serder, Ph.D.,

Izhak Shafran, Ph.D., Xinyu Wang, Ph.D., Dongxiang Xu, Ph.D.,

Hujun Yin, Ph.D., Jongtae Yuk, Ph.D.,

M I L E S TO N E S A N D I N N OVAT I O N S

Information and Investment Criterion:www.WRFCapital.com

WRF Capital provides seedfinancing and active professionalsupport to technology-basedNorthwest start-ups. But unlikeany other venture capitalcompany,we plow back a majorportion of our gains to theNorthwest “seedbed” institutionswhere so much technologicalprogress originates. That’s ourprogram. And it works! WRFCapital currently has 13 activeinvestment projects. Ourgrowing endowment asset basenow exceeds $50 million. Andto date, over $87 million hasbeen earned for or donated to

the research institutions of theNorthwest. (Primarily theUniversity of Washington.)WRF Capital: investing andreinvesting in Northwesttechnological progress.

WRF Capital and AmnisCorporation: David Basiji andBill Ortyn had an idea: find away to analyze cells, proteinsand the genetic code with ordersof magnitude more speed andsensitivity than currentlypossible. The capability wouldbe of inestimable value, leading

to new life-saving drugs andpowerful medical diagnostics.By fusing new ideas with thelatest technologies in optics,lasers and image processing, theyknew it could be done. ThenWRF Capital, in association withStratos Product Development,stepped in to help them foundAmnis and finance thedevelopment and prototypingof their device and process. TheAmnis ImageStream will be a giantstep ahead for science. Theinvestors will profit. And so willthe research institutions of theState of Washington.

Growing forward,plowing back

The win / win missionof WRF Capital:

Bill Ortyn, left, and David BasijiAmnis Corporation

E X T E R N A L F U N D I N G

G R A D U A T E E N R O L L M E N T T R E N D S

E N R O L L M E N T ( A U T U M N 2 0 0 1 )

■ 268 Graduate Students (19% female, 14.2% minority, 47.6% US)■ 545 Undergraduates (20% women)

D E G R E E S G R A N T E D I N 2 0 0 0 - 2 0 0 1

■ Ph.D. 26■ M.S. 53■ B.S. 164