Salthouse to Give First Gupta Lecture on November 19
Chistopher Salthouse, the new Dev and Linda Gupta Assistant Professor in the Electrical and Computer Engineering Department at the University of Massachusetts Amherst, will deliver the first Gupta Lecture at 4:00 p.m. on Thursday, November 19, in the Massachusetts Room of the Mullins Center on campus. A reception will follow the talk at approximately 5:00 p.m. The event is free and open to the public. The title of the lecture is “How to Implant a Fluorescence Microscope.”
The talk will explain the key medical role played by fluorescence microscopy and will describe how implantable fluorescence microscopes could expand that role even more. Dr. Salthouse will also explain how the integrated circuits that have revolutionized computation will solve the biomedical problems related to developing implantable fluorescence microscopes.
Dr. Salthouse received his B.S., M.S., and Ph.D. degrees in Electrical Engineering from the Massachusetts Institute of Technology. During his time at MIT, he developed integrated circuits for cochlear implants, devices that give hearing to the deaf. After finishing his Ph.D., Salthouse moved to Massachusetts General Hospital, where he worked as a research fellow for three years while developing novel fluorescence imaging instruments. In August of 2009, he joined the faculty of the Electrical and Computer Engineering Department at UMass Amherst, where he has founded the Biomedical Electronics Laboratory.
The Dev and Linda Gupta Professorship is an endowment funded through the generosity of Dev and Linda Gupta that supports a new faculty member who brings fresh ideas and areas of inquiry to the Electrical and Computer Engineering Department. The support is specifically intended to encourage risk-taking and entrepreneurial pursuits.
Dev Gupta, who earned his Ph.D. in Electrical Engineering from UMass Amherst in 1977, is an adjunct professor in the Electrical and Computer Engineering Department. He has built a reputation as an expert in broadband and network access infrastructure technology. His company, NewLANS, is working on the development of high speed wireless data transmission. Previously he founded Narad Networks, a privately held provider of broadband IP infrastructure services. (November 2009)
CASA Makes Big Splash with Weather Research
The Engineering Research Center for the Collaborative Adaptive Sensing of the Atmosphere (CASA) is making big waves in the short-wavelength world and beyond. An article co-authored by several CASA researchers, led by CASA Director David McLaughlin and entitled “Short-Wavelength Technology and the Potential for Distributed Networks of Small Radar Systems,” will be the cover story in the influential Bulletin of the American Meteorological Society (BAMS) in December, 2009. CASA was also featured prominently in a recent article about weather research in the prestigious Chronicle of Higher Education, the number one source of news, information, and jobs for college and university faculty members and administrators. The newspaper is subscribed to by more than 70,000 academics and has a total readership of 350,000.
Founded in 1919, the American Meteorological Society (AMS) has a membership of more than 14,000 professionals, professors, students, and weather enthusiasts. AMS publishes nine atmospheric and related oceanic and hydrologic journals — in print and online — sponsors more than 12 conferences annually, and offers numerous programs and services.
Here is an excerpt about CASA from the Chronicle of Higher Education article.
“The University of Massachusetts at Amherst is leading the Engineering Research Center for Collaborative Adaptive Sensing of the Atmosphere, a consortium of universities and industry and government bodies whose ambitious project of networked Doppler radar is delving into complex "mesoscale" phenomena like cloud systems and front formation.
Mesoscale refers to midrange weather systems, between large "synoptic" weather systems and microclimate phenomena such as storms. In one mesoscale project, researchers are figuring out how a type of radar used on naval ships for protection from missiles, the phased-array radar, can help them to detect such events as sharp wind shifts that predict storms.
In another, Vortex2, participants are deploying armadas of cutting-edge detectors to surround tornadoes and the supercell thunderstorms that form them, to gain a far more complete picture of the phenomena than is currently available.” (November 2009)
Swift to Receive Special Award from American Meteorological Society
Calvin Swift, an emeritus professor in the Electrical and Computer Engineering Department at the University of Massachusetts Amherst, will receive the Special Award from the American Meteorological Society at its Awards Banquet, to be held on Wednesday, January 20, 2010, at 7:00 p.m. in Atlanta’s Georgia World Congress Center during the society’s 90th annual meeting. The citation will read, "for sustaining over 30 years an exceptional, interdisciplinary project, resulting in continuous operational monitoring of hurricane surface winds, improved hurricane intensity advisories, and saving countless lives."
Dr. Swift earned his Ph.D. in Physics from the College of William and Mary in 1969. Among other positions, he was an aerospace technologist and group leader at the NASA Langley Research Center in Hampton, Virginia, from 1962 to 1981. He was a professor in the Electrical and Computer Engineering Department at the University of Massachusetts Amherst from 1981 until 2001, and has been an emeritus professor since then.
He is the originator of the concept of using the Stepped Frequency Microwave Radiometer to remotely measure rain rate and ocean wind speed in hurricanes. He developed with students an L-Band Synthetic Aperture Microwave Radiometer to remotely measure soil moisture and ocean temperature. His research on the dielectric constant of sea water provided the technical basis for the development of the Aquarius Satellite to remotely measure ocean salinity.
Among other honors, Dr. Swift was elected as a Fellow of the Institute of Electrical and Electronics Engineers (IEEE) in 1983. He also received the IEEE Centennial Award in 1984. In 1994, he received the Distinguished Technical Achievement Award from the IEEE Geoscience and Remote Sensing Society, and he was presented with the Distinguished Service Award from the IEEE Council on Oceanic Engineering in 1977.
The American Meteorological Society promotes the development and dissemination of information and education on the atmospheric and related oceanic and hydrologic sciences and the advancement of their professional applications. Founded in 1919, AMS has a membership of more than 14,000 professionals, professors, students, and weather enthusiasts. (October 2009)
ECE Paper Wins Dasher Award
A paper based on the teaching innovations of Professor William J. Leonard, a former lecturer in the Electrical and Computer Engineering Department and the 2009 College of Engineering Outstanding Teacher, has won the Benjamin J. Dasher Best Paper Award at the annual Frontiers in Education Conference, sponsored by the American Society for Engineering Education and the Institute of Electrical and Electronics Engineers. The winning paper was entitled “Mastering Circuit Analysis: An Innovative Approach to a Foundational Sequence,” co-authored by Leonard, ECE Department Head Christopher Hollot, and W. J. Gerace.
“This paper was based on Bill’s innovative pedagogue in UMass Amherst’s ECE introductory circuits courses,” explains Hollot.
As the paper explains: “The Department of Electrical and Computer Engineering at the University of Massachusetts Amherst has dramatically changed how Circuit Analysis is learned. Combining a traditional lecture/recitation format with secure, online tests, we have raised our expectations for students and improved student performance, while increasing the fraction of students who succeed, especially underrepresented minorities and women. We call the instructional approach ‘Mastery’ because students continue to work on a topic until they earn a perfect score on the corresponding test. Further, it shows great promise for transforming undergraduate education, especially those courses that provide foundational skills and prepare students for future learning.”
This year’s Frontiers in Education Conference theme was “Imagining and Engineering Future Computer Science, Engineering, and Technology (CSET) Education” and featured 81 sessions related to it. The conference was held in San Antonio, Texas. (October 2009)
Organizing Chaos and Saving Lives During Disasters
On September 26, 1998, a charter bus lost control in mountainous terrain in rural California, resulting in two deaths and many serious injuries. The initial report from emergency personnel reaching the scene exclaimed that "I have a bus on its side and bodies everywhere!" Now an electronic system created by Professor Aura Ganz of the Electrical and Computer Engineering Department at the University of Massachusetts Amherst can quickly make sense of such chaotic disasters, while saving many lives in the process.
Her DIORAMA system, based on Radio Frequency Identification (RFID) technology, can rapidly map out the exact location of the most severely injured victims, as identified by the first-response triage team, so ambulance personnel can find and evacuate critical cases ASAP. Soon DIORAMA will be able determine all the emergency resources in the area and classify local hospitals according to how many ER patients they can handle and what kind of injuries they are capable of treating. Moreover, it will also be able to route ambulances to the right hospitals by steering them around traffic snarls.
“There are so many large-scale disasters that happen with buses overturning, trains colliding, multi-car pileups, air crashes,” Ganz explains. “And it is very cumbersome for medical personnel to arrive at a scene filled with chaos and debris and the victims spread all over the place and carry out a fast and effective triage operation to save as many people as possible.”
Professor Ganz has developed and field-tested DIORAMA through a $400,000 grant from the National Institutes of Health (NIH) and with the help of several medical doctors and emergency management experts from the Beth Israel Deaconess Medical Center (BIDMC), Harvard University, and the Massachusetts Department of Public Health.
“The first problem is the issue of immediate response,” says Ganz. “What is happening in the critical minutes just after a disaster occurs, when every second is crucial to whether or not victims survive? This is the most challenging issue in emergency management of a mass-casualty disaster, and this is the problem NIH is funding me to solve. The aim of the project is to streamline the triage process and significantly reduce the time it takes to rescue victims.”
The term “diorama” usually denotes a replica or scale model of a landscape or some other scene, such as a battle. In this case, the actual DIORAMA model is a map of the disaster scene, pinpointing the location of all the medics and victims and accessible on the computer screen of the incident commander coordinating the whole emergency response operation.
RFID technology allows the storage and retrieval of data through electromagnetic interaction with a radio frequency compatible integrated circuit. An RFID system consists of RFID readers and RFID tags. Such systems are currently being deployed by numerous companies, the government, the military, the medical community, and other users as a radical new means of tracking and identifying products and people.
In DIORAMA, emergency vehicles will carry “emergency locators” (which, in effect, function as “readers”) that will track location and status information from various RFID tags worn by emergency personnel and attached to accident victims during the initial triage. These components create a mobile tool that can be speedily deployed at a disaster scene to enable an offsite incident commander to visualize the location and condition of the casualties as well as all the available resources.
DIORAMA gives the incident commander the small picture at the scene of the accident by pinpointing all the victims. Soon DIORAMA will also provide the large picture by mapping out rescue routes and identifying available emergency resources and nearby hospitals, their capacity, and their capabilities for dealing with specific kinds of injuries.
“Currently the incident commander has to find out most of this information by phone, if it’s available at all,” relates Ganz. “The incident commander is flying blind.”
To create this life-saving tool, Professor Ganz is collaborating with some real heavyweights in the medical emergency management field: Doctors Larry Nathanson, Jonathan Burstein, and Gregory Ciottone from BIDMC. Dr. Nathanson is the Director of Emergency Medicine Informatics at BIDMC and a faculty member of Harvard Medical School. Dr. Burstein is the Medical Director of the Emergency Preparedness and Response Program at the Massachusetts Department of Public Health and the Massachusetts State EMS Medical Director. He is the co-editor of Disaster Medicine, and an editor of the text Medical Response to Terrorism. Dr. Ciottone is the Director of the Division of Disaster Medicine at BIDMC and has served as the commander of a federal Disaster Medical Assistance Team.
“Our next step in developing the DIORAMA will also include dynamic traffic data,” says Ganz. “I am working with the UMass Transportation Center on this part of the concept. At that point, DIORAMA would take into consideration the up-to-the-minute traffic pattern when mapping out routes for ambulances taking victims to hospitals.”
A prototype for the user-friendly DIORAMA system has already been validated during a half-dozen field tests at UMass Amherst, UMass Medical School, and BIDMC this summer, but Ganz is currently applying for further NIH support to make the system even better. Given that the second round of R01 NIH funding is approved, she predicts that her system could be ready for testing in real disasters in as little as a year. (September 2009)
Boryssenko Writes Cover Story for High Frequency Electronics
Research Professor Anatoliy Boryssenko of the Electrical and Computer Engineering Department has published the cover story for the August 2009 edition of High Frequency Electronics magazine. Dr. Boryssenko’s story, entitled "Taxonomical and Heuristic Studies on UWB Antenna Design Strategies," borrows from biology to develop a method for evaluating wide bandwidth antenna structures and create rules for the discovery of useful new structures. As he writes in his article, “This paper attempts to assist in navigation through the realm of reported designs on ultrawideband (UWB) antennas through systemization of key design principles and ideas using two methodologies proven in other science and engineering fields.” The magazine’s Editorial Director, Gary Breed, puts the article in perspective during a long editorial on Boryssenko’s article: “These techniques provide an orderly, rules-based approach to discovering potential new antenna structures. We should expect it to work – engineering is inherently orderly and is definitely based on rules.” Link to article: High Frequency Electronics. (September 2009)
MIRSL Makes Cover of IEEE Journal
A paper based on the thesis research of former graduate student Dragana Perkovic, working with the Microwave Remote Sensing Laboratory (MIRSL), was chosen for the cover article of a prestigious journal published by the Institute of Electrical and Electronics Engineers (IEEE). The article is titled "Longshore Surface Currents Measured by Doppler Radar and Video PIV Techniques." It can be found in the August 2009 issue (Vol. 47, Issue 8, Part 2, Page 2787) of the IEEE Transactions on Geoscience and Remote Sensing journal. The authors are Perkovic, T. C. Lippmann of the Center for Coastal and Ocean Mapping at the University of New Hampshire, and Stephen Frasier, the director of MIRSL and a professor in the Electrical and Computer Engineering Department.
The main thrust of the article is to compare, contrast, and test two innovative techniques for measuring the all-important currents that flow and churn along shorelines while transporting sediments and pollutants in the surf zone.
As the article notes, “These flows are affected by local bathymetry, as well as spatial and temporal changes in incident-wave energy and direction. The ability to capture the spatial and temporal variations in mean flows in the surf zone is key to predicting the morphological change of the near-shore topography and shoreline.”
The article explains that much of the field research in the nearshore zone is done with in situ measurement techniques, using instruments such as pressure sensors and current meters that are fixed on pipes jetted into the sandy bottom or instruments that are mounted on moving platforms such as drifters. But these techniques are difficult to employ because of the large number of instruments required, and because they present a hazard to recreational swimmers, surfers, or boaters. These instruments are also difficult to install and maintain in the harsh nearshore environment for a given length of time.
Recently, remote sensing technology has been applied to studies of nearshore processes. Remote sensors are less invasive, are generally easier to deploy and maintain, and offer wider areal coverage than typical arrays of in situ instruments. However, because the remote measurements are indirectly related to the quantity of interest, field verification is required to establish the validity of the measurements and to understand their limitations.
In this research, mean longshore surface currents within the surf zone were measured using two remote sensing techniques: microwave Doppler radar and optical video. Doppler radar relies on small-scale surface roughness that scatters the incident electromagnetic radiation so that velocities are obtained from the Doppler shift of the backscattered radiation. Video relies on texture and contrast of scattered sunlight from the sea surface, and velocity estimates are determined using particle imaging velocimetry (PIV). This paper compares video PIV and Doppler radar surface velocities over a 1-kilometer by .5-kilometer area in the surf zone of a natural beach.
“The good spatial and temporal agreement between the two remote measurement techniques, which rely on very different mechanisms,” as the article concludes, “suggests that both are reasonably approximating the true mean longshore surface velocity.” (August 2009)
Cisco Hiring Lots of Our Students
Cisco Systems, Inc., whose corporate headquarters are in San Jose, California, has hired six recent graduates from the College of Engineering in its current group of new employees. The company has a program called "Cisco Choice" for hiring engineers directly out of college and, out of the 200 new engineers hired across the nation, UMass Amherst led the count with nine students. Besides the six from engineering, there were also three new hires from the Computer Science Department. “Just thought you guys would appreciate the fact that Cisco hired more students from UMass than any other school in the U.S.,” said Austin Cormier, who graduated from our Electrical and Computer Engineering Department, in a note to our faculty. “You guys are obviously doing something right!” The other five from the college were Ivan Berkovich, Scott Richard, Alex Trefonas, Doug Frazier, and Danxiang Li. Cisco also hired 12 interns, seven from engineering and five from computer science.
Cisco Systems, Inc., founded in 1984, is the worldwide leader in networking for the Internet. Today, networks are an essential part of business, education, government, and home communications, and Cisco Internet Protocol-based (IP) networking solutions are the foundation of these networks. Cisco hardware, software, and service offerings are used to create Internet solutions that allow individuals, companies, and countries to increase productivity, improve customer satisfaction, and strengthen competitive advantage. Cisco has 66,558 employees worldwide. (July 2009)
Another Engineering Phenom Honored at Commencement
Jamaica Plain resident Ivan Bercovich, an electrical engineering major and honors student, is the third engineering student who will be presented with 21st Century Leaders Awards during the 139th Annual Commencement on Saturday in McGuirk Alumni Stadium. Among many other activities, Bercovich is the leader of a team of students that designed and built a “Personal Head-Up Display,” or HUD, that effectively works as an OnStar navigation system for pedestrians. Bercovich and the HUD were recently featured in the Boston Globe (Globe). He has also done research on a system to beat the stock market by using the electrical engineering principles of “signal processing” as a guide.
“Visitors to a new area struggle to find their way around,” says Bercovich about the problem addressed by the HUD. “Often they can be standing right in front of the building they are looking for without even realizing it. It would be very helpful if there was a system which would place large virtual labels above buildings the user was looking at, as well as information specific to buildings in the user's field of view.”
Admittedly, the HUD doesn’t give its wearer the panache of Fred Astaire in a top hat. It looks like a space-age welder’s helmet and contains such instruments as gyroscopes, compasses, a GPS, and accelerometers. Bercovich and company hope to miniaturize the device. But the HUD is so brilliantly conceived that it won first place in the Senior Design Competition run by the UMass Department of Electrical and Computer Engineering and second place in the regional Senior Design Competition staged by the Institute of Electrical and Electronics Engineers.
Bercovich has also been using the tricks of his own trade to work out a sound stock-market investment strategy that will make sharp downturns, such as the current recession, irrelevant in the long run. His premise is that investors have to make long-term commitments to their stocks, 20 years in this case, regardless of short-term volatility in the market. Bercovich’s method is based on a scientifically proven engineering technique known as “signal processing,” meaning the analysis, interpretation and manipulation of signals. In Bercovich’s research, the “frequency” of the signal is represented by the up and down graphing of closing stock-market prices over the last century.
“I’m using the same engineering technique that one would use to process a signal,” says Bercovich. “So the signal will have a pattern, or frequency, and it will have noise. In my research, we look at the fluctuations in the market as frequency and the market volatility as the noise. So, when I am studying volatility I am really trying to determine how noisy the market is. It’s much the same as trying to receive a radio signal when it is thunder-storming.” (May 2009)
A head-up display for wayward travelers
It takes a certain amount of nerd courage to don the massive headset recently assembled by a group of UMass seniors. But their augmented reality, "personal head-up display" (www.ecs.umass.edu/ece/sdp/sdp09/wolf/) will make you the go-to guy for directions around campus - even the big city.
The personal HUD is ugly: It looks like a skateboarding helmet with a solar oven tacked to the front and a circuit board screwed on top.
The wearer sees a full-color, high-resolution, 3D readout superimposed on a 2D plane of view - a display that changes as you turn and move your head.
Inside the HUD, you see the names of the buildings you are looking at, seemingly projected against those structures. The building names and their coordinates were entered by the UMass personal HUD investigators. The HUD can also derive information from text files, Google Earth data, and other sources, said UMass electrical engineering major Ivan Bercovich.
It's like a wearable version of Wikitude (www.mobilizy.com/), the augmented reality guidebook for the T-Mobile G1. Wikitude superimposes Wikipedia data about landmarks over the live picture in the G1's camera view. Wikitude relies on the phone's compass, accelerometer, and GPS transponder to determine what you're looking at.
The UMass students used similar technologies. They also added OpenGL graphics software and a microprojector from 3M. Microprojectors have revolutionized AR headsets, Bercovich said recently.
Invoking a bit of Ray Kurzweil-futurist talk (for which I am a sucker, by the way), Bercovich noted further miniaturization is necessary.
"We are right on the singularity of not being able to do it (miniaturize personal HUDs) and being able to do it," he said. (May 2009, Boston Globe, Mark Baard)
Tang Receives Distinguished Faculty Award
On Thursday, April 16, Professor Emeritus Ting-wei Tang of the Electrical and Computer Engineering Department will receive the Distinguished Faculty Award from The University of Massachusetts Amherst Alumni Association as part of the 2009 Distinguished Alumni Award ceremonies at the State House in Boston. University of Massachusetts President Jack Wilson and UMass Amherst Chancellor Robert C. Holub will address those in attendance prior to the awards presentation.
Professor Tang received his bachelor’s degree from National Taiwan University, and his master's and Ph.D. degrees from Brown University. He joined the UMass Amherst faculty in 1968. Tang's research field is in semiconductor device physics and numerical simulation of semiconductor devices. He is a Life Fellow of the Institute of Electrical and Electronics Engineers, elected for his contributions to the hydrodynamic transport modeling of semiconductor devices.
Professor Tang was the recipient of the Outstanding Senior Faculty Award and the Outstanding Teaching Award from the College of Engineering in 1989 and 1991, respectively. In 2000, he was awarded the UMass Amherst Chancellor’s Medal. He and his wife, Shirley Tang, founded the annual Tang Endowment Lecture Series, which brings leaders of engineering-based companies to campus to interact with students and faculty and present a major talk. Shirley Tang is a retired academic advisor for the United Asia Learning Resource Center.
The Distinguished Alumni Award is the highest honor bestowed by the UMass Amherst Alumni Association on alumni, faculty, and friends. Recipients of this award have translated their UMass Amherst experience into distinguished achievement in the public, business, or professional realms and bring honor to UMass Amherst and to their field of endeavor. (April 2009)
MIRSL Part of Historic Study to Understand Tornadoes
The Microwave Remote Sensing Laboratory (MIRSL) is playing a critical role in an historic study to explore the origin, structure, and evolution of tornadoes, a project which will take place from May 10 to June 13 across the central United States. The project, named the Verification Of Rotation in Tornadoes EXperiment 2 (VORTEX2), is the largest attempt in history to study tornadoes and will utilize more than 50 scientists and 40 research vehicles, including 10 mobile radars. VORTEX2 is funded by the National Science Foundation (NSF) and the National Oceanic and Atmospheric Administration (NOAA), and involves scientists from NOAA, 10 universities, and three non-profit organizations. The main objective of the MIRSL work is “to understand better the dynamics and kinematics of severe convective storms and the tornadoes they sometimes spawn.” MIRSL will operate two mobile Doppler radars during the project: The University of Massachusetts mobile W-band radar, and its mobile, polarimetric, X-band radar.
Researchers will sample the super-cell thunderstorms that often form over more than 900 miles of the central Great Plains. Areas of focus include southern South Dakota, western Iowa, eastern Colorado, Nebraska, Kansas, the Texas panhandle, and western Oklahoma.
The director of MIRSL is Stephen Frasier, a professor in the Electrical and Computer Engineering Department. MIRSL’s participation in VORTEX 2 is integral to its success. MIRSL’s mobile W-band radar has the highest spatial resolution of all existing mobile Doppler radars. This radar is critical to VORTEX2 in that it provides the best possible opportunity to map the wind field in and just above the important friction layer of tornadoes. It also has the best chance to document the structure of multiple, sub-tornado-scale vortices, which are thought to cause much of the localized, extreme damage in some tornadoes.
MIRSL’s mobile, polarimetric, X-band radar will provide information on whether scatterers within tornadoes are composed of hydrometeors or of debris. Such knowledge is essential to separate target motion from air motion. It will also be used to estimate the type of hydrometeors present (rain, hail, etc.) in all regions of supercells.
"An important finding from the original VORTEX experiment was that tornadoes happen on smaller time and space scales than scientists had thought," said Stephan Nelson, NSF program director for physical and dynamic meteorology. "New advances from VORTEX2 will allow for a more detailed sampling of a storm's wind, temperature, and moisture environment and lead to a better understanding of why tornadoes form--and how they can be more accurately predicted."
NSF has contributed $9.1 million to VORTEX2. The original VORTEX program, operated in the central Great Plains during 1994 and 1995, documented the entire life cycle of a tornado for the first time in history. Recent improvements in severe weather warning statistics may be partly due to the application of VORTEX findings. VORTEX2 will build on the progress made during VORTEX and further improve tornado warning skills and short-term severe weather forecasts. Data collected from VORTEX2 will help researchers understand how the large-scale environment of thunderstorms is related to tornado formation, according to Louis Wicker, meteorologist at NOAA's National Severe Storms Laboratory and a VORTEX2 co-principal investigator.
Participating scientists and students are from organizations throughout the United States and three countries. The researchers come from MIRSL, the Center for Severe Weather Research, Rasmussen Systems, NOAA National Severe Storms Laboratory, OU/NOAA Cooperative Institute for Mesoscale Meteorological Studies, NSF-sponsored National Centers for Atmospheric Research, Penn State University, University of Oklahoma, Texas Tech University, Lyndon State College, University of Colorado, Purdue University, North Carolina State University, University of Illinois, University of Nebraska, and Environment Canada and the Australian Bureau of Meteorology. (April 2009)|
Distinguished Journal Chooses Polizzi Article as “Editors’ Suggestion”
An article written by Eric Polizzi of the Electrical and Computer Engineering Department (ECE) was just published by the distinguished journal Physical Review B and then chosen for one of the precious few selected as “Editors’ Suggestion” papers. The editors and referees choose only five to 10 of the 80 to 150 papers accepted each month for this honor. “As a service to both our readers and authors,” notes the publication, “we are formally listing a small number of PRB papers that the editors and referees find of particular interest, importance, or clarity. These Editors' Suggestion papers, when published, are listed prominently on prb.aps.org and marked with a special icon in the print and online Tables of Contents and in online searches.” As ECE department head Christopher Hollot says, “This is quite an honor for Eric, our department, the college, and the university; something to blow our horn about.”
Physical Review B is the largest and most comprehensive international journal specializing in condensed matter and materials physics, publishing important papers on a wide range of topics. It is ranked number one in total citations in condensed matter physics. The title of Dr. Polizzi’s paper is “Density-matrix-based algorithm for solving eigenvalue problems,” published on March 16.
The abstract reads: “A fast and stable numerical algorithm for solving the symmetric eigenvalue problem is presented. The technique deviates fundamentally from the traditional Krylov subspace iteration based techniques (Arnoldi and Lanczos algorithms) or other Davidson-Jacobi techniques and takes its inspiration from the contour integration and density-matrix representation in quantum mechanics. It will be shown that this algorithm—named FEAST—exhibits high efficiency, robustness, accuracy, and scalability on parallel architectures. Examples from electronic structure calculations of carbon nanotubes are presented, and numerical performances and capabilities are discussed.” (March 2009)
M5 Lets Students Do Their Thing
Theory is something you can’t touch, take apart, or put back together. But electrical and computer engineering theory is applied to hands-on hardware every day in the lower level of Marcus Hall in a place called M5 (short for Marcus Hall, Room 5). M5 consists of 13 rooms chockablock with circuits, chips, parts, voltmeters, power supplies, oscilloscopes, audio/video equipment, microcontrollers, and many other things near and dear to the hearts of engineers. It’s a place, as one student jokes, where electrical and computer engineering students can “do their Rube Goldberg thing.”
What the Electrical and Computer Engineering Department (ECE) offers its students through M5 is something extremely rare. M5 means free access to electronic components, specialized test equipment, a design-oriented reference library, open hours staffed by undergraduates, a “junk room” with old electronics for students to use for parts or reverse-engineering, an instructional lab dedicated to electronics hardware and computing, an audio engineering workstation, a hub where students can gather and bond, and a place for ECE students to call home.
M5 is the brainstorm of ECE Department Head Christopher Hollot and ECE Professor Baird Soules. Soules cites a Carnegie Report that came out in December of 2008 and reinforced M5’s mission to balance theory and application in the curriculum.
“So we’re trying to offer a place with opportunities to apply engineering knowledge in a concrete fashion,” he notes. “M5 gives students a place to design and build things even before they actually cover the theory.”
Or, as Hollot says: “The desire to tinker is a terrible thing to waste. Traditionally, the urges to take stuff apart and figure out how it works were hallmarks of engineers-to-be. However, the miniaturization of electronics in modern commercial products has made this ‘open Sesame’ activity inaccessible. M5 aims to lower these barriers.”
What Soules calls the “intuitive knowledge” learned by students while building things and taking them apart also allows them to apply the analytical knowledge they get from learning theory in the classroom. Students who’ve gotten hooked on M5 agree.
“Last week I found an Amplifier for an iPod and decided to recreate it by rewiring the iPod’s connector cable,” says freshman electrical engineering major Josh Lowe. “And yesterday I came down here to M5 and rewired it and got it to play through some speakers. In the process, I took apart a speaker to see how it works. Later I’ll branch off that and apply it to something else. Everything down here is a work in progress.”
Lowe also built a robotic claw using surplus parts he found lying around M5. Meanwhile, electrical engineering freshman Dan Bercht has been working on a couple of projects. One uses an old Atari joystick to control the speed of a tape player and to create the kind of musical gyrations and oscillations that DJs produce in nightclubs. Another project uses a microcontroller that senses distance between a sensor and its physical surroundings. Then that data is sent to another program called Processing to make a virtual representation of the physical surroundings. “It’s kind of like rendering a 3D image of a real life object,” as Bercht says.
What can you do with that? “Not much,” he readily admits. “It’s just to see how it works. That’s a big part of the draw here, coming down to M5 to find out how something works. You just grab the parts you need and go on the Internet to find out how to do what you want to do.”
Lowe and Bercht got addicted to M5 after they took Professor Soules’ freshman course in Electronics and Computing Laboratory. The course takes students from the most elementary circuits through digital circuits, then introduces them to an open-source microcontroller board called Arduino, around which they can do programming. Now M5 has hired Lowe and Bercht as class assistants and as two of the 10 student staffers who operate M5, which is open 50 hours per week under student staff supervisor Jessica Lau. Lowe and Bercht not only provide one-on-one help for students doing lab work in the class, but Soules pumps them for lab topics they felt were missing when they took the course. Soules also teaches an Electronic Music Lab for sophomores, juniors, and seniors. The distinctive element in both classes is that they are based on all the very specialized equipment in M5.
“In general, M5 gives our students much more hands-on access to the hardware they’re learning about in class than they could have possibly had in the past,” explains Soules. “Our corporate friends tell us they want their engineers to be well-rounded, with the theoretical knowledge as well as the hands-on skills to apply that theory.”
Even a brilliant idea like M5 requires money to run it. M5 needs funds to pay its 10 student assistants, money to grow the facility and deepen the pool of electronics hardware, and support to purchase some wonderful but pricey educational kits for students to build.
“Whenever you build something, you’re learning valuable bench-top skills,” says Soules. “You develop understanding of the various components and their roles. The kits will help students understand the theory they’ll be taking later in class.”
It’s not all work and no play at M5. “People are not just working on projects,” says Bercht, “but they’re working on homework or studying or just hanging out. So we’re really getting to know each other in ways we couldn’t before, because we didn’t have a place to call home. It’s a great place for brainstorming.”
Lowe wishes he had known about M5 when he was juggling acceptances to multiple schools. “Before I transferred here, I was trying to decide among three different colleges,” he says. “If I had been told then that there was a place like M5 where I could come in and do all this hands-on stuff and build things, I would have picked UMass instantly.” (March 2009)
Pishro-Nik Gets College's 23rd NSF CAREER Grant
Hossein Pishro-Nik is working on the kind of system that automobile manufacturers have always dreamed of creating: a wireless communication network to prevent cars from crashing into each other. The National Science Foundation has awarded Dr. Pishro-Nik a $400,000 grant from its prestigious Faculty Early Career Development (CAREER) Program to create the theoretical and mathematical framework for this kind of anti-crash system. Pishro-Nik is a faculty member in the Electrical and Computer Engineering Department at the University of Massachusetts Amherst.
“The idea is to make cars equipped with wireless communication capabilities so they communicate with each other,” explains Pishro-Nik. “We use this network to prevent accidents and also send traffic-congestion information to drivers. It is predicted this new capability can significantly improve the safety and efficiency of the transportation system.”
Here’s an example of the kind of wireless communication Pishro-Nik is studying. Suppose two vehicles are approaching an intersection and both have the wireless capability to communicate with each other. Let’s say one vehicle runs the red light at the same time as the other is about to reach the intersection, so there is a high chance of collision. Because these two vehicles are automatically communicating, the second driver gets a warning that the first car is running the light and brakes to prevent a potentially fatal accident.
In fact, Pishro-Nik and Daiheng Ni, a faculty member in the Transportation Engineering Group, have worked with two different student teams that have built actual working prototypes of just such a wireless communication system.
According to Pishro-Nik, today’s transportation systems are facing significant challenges. Traffic accidents are the most common cause of death in the 15 to 34 age group by a wide margin. In terms of congestion, in the past three decades, the average delay has tripled. Such congestion wastes about 40 percent of travel time on average, unnecessarily consumes about 2.3 billion gallons of fuel annually, and adversely affects the environment. These critical transportation issues are not only taking a high toll on the economy, they are also changing the way people live, travel, and work.
Many countries are planning the deployment of vehicular ad hoc networks (VANETs) that can dramatically improve the safety and efficiency of transportation systems. Nevertheless, there is currently no rigorous mathematical framework for vehicular ad hoc networks. Pishro-Nik’s NSF research aims to develop a solid theoretical framework for wireless-enabled transportation systems by combining communication theory and traffic flow theory.
What kind of issues is he studying? First, there is the line-of-sight problem.
“One of the things I’m trying to develop here is a theory of obstructive wireless communication,” says Pishro-Nik. “These networks work in high-frequency ranges. They cannot go through buildings. So if you are coming to an intersection in a city, and there is a building between you and another vehicle approaching the intersection from an adjoining street, your vehicle cannot communicate directly with that car. You would have to have multiple links. Your vehicle would have to communicate with that car through another vehicle or have a communication hub at an intersection so that both vehicles can link through that.”
Another issue? Time is of the essence for transmitting emergency warnings in traffic. Safety applications, such as accident warning messaging, are highly delay sensitive. Thus, there should be direct low-delay communication between vehicles to help prevent accidents. Another difficult question is how many warning signals are too many. Pishro-Nik advises that you don’t want to warn drivers unless they are in immediate danger. You don’t want to give them constant warnings in traffic or else the “boy who cried wolf syndrome” sets in.
Although, Pishro-Nik’s research is theoretical in nature, he is in an excellent position to validate his mathematical results. He and his collaborators at the Transportation Engineering Group at UMass Amherst have built a VANET test-bed and have access to a large set of real traffic data.
“I’m looking at the big picture,” he says. “I’m examining complex networks with thousands of vehicles and trying to answer the most important questions about what these communications networks can provide and what is the optimum design to provide it. I’m looking at the whole system. We have two networks interacting here: the communications network and the transportation network. How can they work together to get the most benefit?” (March 2009)
Taming the Wild and Wooly Stock Market
Jamaica Plain resident Ivan Bercovich, a senior electrical engineering major at the University of Massachusetts Amherst, is using the tricks of his own trade to work out a sound stock-market investment strategy that will make sharp downturns, such as the current recession, irrelevant in the long run. His premise is that investors have to make long-term commitments to their stocks, 20 years in this case, regardless of short-term volatility in the market. Bercovich’s method is based on a scientifically proven engineering technique known as “signal processing,” meaning the analysis, interpretation, and manipulation of signals. In Bercovich’s research, the “frequency” of the signal is represented by the up and down graphing of closing stock-market prices over the last century.
Bercovich’s research, which he is doing for his senior honors capstone thesis, was inspired by an article in Money Magazine written by Senior Editor Walter Updegrave. In his article, Updegrave posed a question that is much debated in the financial community.
“They are debating whether it is better to invest a sum all at once, or release it into the market in smaller slices to erode some of the risk,” says Bercovich. “Although there are a lot of articles that choose either side, it is rare to find numerical support for these claims. My objective is to perform a series of statistical analyses and come up with a more quantitative reason to choose one strategy over the other.”
With the help of his advisor, Professor Dennis Goeckel of the Electrical and Computer Engineering Department, the Argentinean-born Bercovich recognized that the ups and downs of the stock market are actually time-varying quantities that act like signals. They can therefore be analyzed and interpreted through signal processing, which quantitatively analyzes radio and telecommunication transmissions and many other kinds of signals, including sensor data such as electrocardiograms. With that idea in mind, Bercovich has gathered almost 100 years of figures from the Dow Jones Index and 50 years from the S&P 500 Index, all of which he is now analyzing as if it were an electrocardiogram of our stock market. In many ways it is.
“I’m using the same engineering technique that one would use to process a signal,” says Bercovich. “So the signal will have a pattern, or frequency, and it will have noise. In my research, we look at the fluctuations in the market as frequency and the market volatility as the noise. So, when I am studying volatility I am really trying to determine how noisy the market is. It’s much the same as trying to receive a radio signal when it is thunder-storming.”
In order to compare the “one lump sum” strategy of investment with the “little bit at a time” strategy, which people in the financial community refer to as “dollar cost averaging,” Bercovich is testing the two approaches against each other to see how they would perform over the past century of stock-market records.
“I will invest a lump sum on a random day and then keep the stock for 20 years,” he explains, “and I repeat this experiment 10,000 times. Then I see how the distribution of my returns comes up. Next I do the corresponding experiment, which is the dollar cost averaging, and repeat the same experiment 10,000 times. The idea is to compare these two strategies.”
After he completes this exhaustive analysis, he can determine quantitatively which approach is better. What Bercovich is using are equations that mathematicians and engineers have developed to predict signal fluctuations. His goal is to help people in the financial world translate theory to real-world decisions. He is combining his background in mathematics and electrical engineering, especially signal processing, and several courses in finance to connect these fields and interpret financial data from an engineering perspective.
One goal Bercovich hopes to reach is protecting investors from the so-called St. Petersburg Paradox, a classical economics problem in which a naïve decision criterion, based on unsound mathematics, suggests a course of action that no rational person would be willing to take because of false expectations and unforeseen results.
“After I analyze this particular problem,” says Bercovich, “this research could then be extended to explore the systematic analysis of financial data in all circumstances for which expectation yields deceiving results.” (February 2009)
Will the Internet Blow Our Minds?
While waxing eloquent about his upcoming Distinguished Faculty Lecture on February 23rd at 4:00 p.m. in the Massachusetts Room of the Mullins Center, Electrical and Computer Engineering (ECE) Professor Weibo Gong mentions the research being done in various places that could create what Time Magazine once called “super rats” and he calls “human rats.” When somebody jokes that there are already far too many human rats, Gong laughs heartily. But then he notes that this glib exchange expresses a key point in his lecture, entitled “Will the Internet Soon Outsmart Humans?” One sign that humans are still slightly ahead in our virtual race with the Internet is our sense of humor. As of right now, the Internet can’t joke. But all that could change. Professor Gong believes it’s only a matter of time before the Internet has the last laugh.
Gong says that in one area the human brain still outdoes the prodigious ability of Internet search engines to rifle through content. That’s because the brain engages in “creative searching,” intelligently connecting different types of knowledge stored in different regions. The Internet cannot yet duplicate such creativity. Our ability to joke is one sign of creative searching. But Gong argues that science already possesses the means to tap the enormous capacity of the Internet to make creative connections efficiently, that this will likely happen soon, and we should be prepared for the consequences.
“Our intelligence is based on the vast labyrinth of neurons composed through the learning process into many many sub-networks,” explains Gong. “As humans, we have the ability to connect these networks by doing creative searching, intelligent searching that makes associations through intelligent assumptions.”
Search engines such as Google, by contrast, work through what Gong calls a “Random Walk” through the Internet. In this case, it’s a race walk because of the tremendous power and speed of search engines.
“But soon the Internet, which mimics the brain in so many ways, will be able to out-perform the brain in many many functions because of the Internet’s ability to synthetically reproduce much denser networks with new kinds of hardware such as end nodes,” says Gong.
All these philosophical thoughts are the sorts of intelligent connections that Gong makes routinely while thinking quite profoundly about subjects such as the potential of the Internet to outwit people.
“Weibo actually stops you in the hallway and uses you as a sounding board for his theoretical ideas,” says Christopher Hollot, the head of the Electrical and Computer Engineering Department. “In that sense, this talk will be Weibo being Weibo. He wants to use the whole campus community as a sounding board for his ideas.”
Gong refers to the Internet as a forest with millions of individual trees that represent the individual networks of knowledge. It’s a sort of big, leafy Wikipedia. But the Internet has not yet developed to the extent that it can connect these individual trees through creative searching and intelligent leaps of thought. So the Internet still cannot see the forest for the trees. But, Gong believes, we are quickly reaching the point when science will harness the Internet to outstrip our human brain power.
“The Internet is like nuclear energy,” says Gong. "As it develops, it can be used for tremendous good or for catastrophic harm. But it will be developed. Whether for good or for evil is out of the hands of the engineers and scientists who develop it. As Harry Truman said about Albert Einstein’s campaign for a sane nuclear policy and control of atomic bombs: ‘It’s none of his damn business.’”
The Internet was originally created to connect the people in the same tree, in the same field of knowledge. Now it is connecting all the trees, but not necessarily in an intelligent way. It connects them with the power of search engines, which randomly walk through the whole forest but don’t make intelligent leaps of logic while doing it.
“If we all want the internet to be as useful and effective as possible, bringing it to the point in which it can connect the trees through intelligent leaps of logic, we’re all going to have to work together in an interdisciplinary way,” says Gong. “We can see the technical possibility now to break the age-old barriers between the trees.”
A member of the ECE faculty since 1987, Gong has been an adjunct professor of Computer Science since 2001. Gong earned a master’s degree in control theory from the University of Science and Technology of China in 1981, and a master’s degree in engineering in 1985 and a doctorate in applied mathematics from Harvard University in 1987. Gong’s research interests include security of communications networks, modeling and control of communication networks, optimization methods, queuing theory, stochastic dynamic systems in electrical engineering, and cognitive searching methods. He received the Institute of Electrical and Electronics Engineers (IEEE) Transactions on Automatic Control George Axelby Outstanding Paper Award in 1997, was elected an IEEE fellow in 1998, and received the College of Engineering Senior Faculty Award in 2002. Students from his lab have become faculty members of Columbia University, New York University, Central Florida University, and various high-tech companies, including Google and Cisco.
Professor Gong’s lecture is free and open to the public, and a reception follows his talk. (February 2009)
Polizzi Gets College's 22nd NSF CAREER Award
Eric Polizzi of the Electrical and Computer Engineering Department has received a prestigious $400,000 CAREER grant from the National Science Foundation (NSF) to create a new suite of computer simulation methods to tackle the challenges created by designing, modeling and testing nano-devices that become more miniaturized every year. Dr. Polizzi’s research can be applied to simulations ranging from material sciences and chemistry to nano-electronics and bio-nanotechnology.
As the NSF panel noted when reviewing Polizzi’s project, “This is a must-fund proposal, because it would provide a unique and important tool to the entire Moore’s Law research community and beyond.”
“We are creating nano-devices that are smaller and smaller,” says Polizzi about his NSF project. “To make such devices as silicon nanowire, carbon nanotube transistors, nanoribbons, or a hybrid combination of those, it requires a lot of experimental research. Simulation becomes more and more important because it’s flexible and much less expensive than experimentation.”
Polizzi received his grant from the NSF Faculty Early Career Development (CAREER) Program. CAREER is the NSF’s most prestigious program for young faculty members, and it “recognizes and supports the early career development activities of those teacher-scholars who are most likely to become the academic leaders of the 21st century,” according to the NSF. Polizzi is the 22nd faculty member from the College of Engineering to be awarded the CAREER or its equivalent.
In nano-electronics, modeling and simulation have become critical for supplementing experimental approaches while designing, characterizing, and testing new devices. One objective of Polizzi’s research is perfecting the computational modeling required for creating future generations of high-speed, high-function, electronic devices. The project aims at ever higher levels of detail and realism in these simulations.
“These simulations must be fast,” sums up Polizzi, “and they must be accurate.”
Polizzi is using his diverse background in fundamental physics, mathematical modeling, device engineering, and computer science to create high-performance computing tools from a multidisciplinary perspective. Before coming to UMass Amherst in 2005, Polizzi was a senior research scientist in computer science and a postdoctoral research associate in electrical and computer engineering at Purdue University. He has also worked as a research and teaching associate in applied mathematics at the National Institute of Applied Sciences in Toulouse, France. He earned his B.S. and M.S. in physics and his Ph.D. in applied mathematics at the University of Toulouse.
“So I start with addressing the quantum physics, mathematical modeling, numerical techniques, algorithmic methodologies, and their implementation on parallel computing platforms, and finally I can perform the simulation,” says Polizzi about his scientific modeling process.
Miniaturization has provided opportunities for the electronics industry to improve the speed and efficiency of the transistor operations and computing power of processors. Nowadays, transistor operations are performed at the nanometer scale, but aggressive downscaling has caused many devices to reach their quantum limits and create many problems, including nanoscale fluctuations and leakage processes induced by quantum-mechanical tunneling. The continuing miniaturization of the devices will require drastic changes in the design and operation of the basic building blocks in computer technology.
Polizzi’s methods will allow an order-of-magnitude speedup in the modeling stage, which is extremely important for the designers of devices, circuits, and chips while running their simulations numerous times in search of the best design. The suite of modeling methods being developed by Polizzi will also be essential for understanding the fundamental physics governing the operation of novel nano-devices that are under development and promise to have a profound impact on the flagging world economy. (February 2009)
Schaubert Article makes Cover of Microwave Journal
Daniel Schaubert, the director of CASCA and a professor in the Electrical and Computer Engineering Department, was the lead author for the cover article in the January issue of Microwave Journal, the industry standard for microwave news (http://www.mwjournal.com/Journal/). The article was entitled “State-of-the-Art Antenna Technology: The 2008 Antenna Applications Symposium.” The Antenna Applications Symposium and its predecessor, the Air Force Antenna Symposium, have for more than 50 years provided a unique forum for exchange of ideas and information about the practical aspects of antenna design, development, and use in systems. The Antenna Applications Symposium is held annually at the Retreat Center in Robert Allerton Park, a century-old Georgian mansion just outside of Monticello, Illinois. The elegance of the facility, a single-track technical program with stimulating presentations, and ample networking opportunities contribute to fruitful group and one-on-one interactions. The other authors were: Jennifer Bernhard, University of Illinois; Robert Mailloux, US Air Force Research Laboratory; W. Devereux Palmer, US Army Research Office. (January 2009)
Muschinski Makes Cover on the Bulletin of the American Meteorological Society
The research of Andreas Muschinski, the first Jerome M. Paros Professor in Measurement Sciences in the Electrical and Computer Engineering Department, was featured on the cover of the prestigious Bulletin of the American Meteorological Society, Volume 89, Issue 11 (November 2008). The article, entitled “Metcrax 2006: Meteorological Experiments in Arizona's Meteor Crater,” details the work that Professor Muschinski and 13 other researchers from four universities and two more organizations are doing in the 1.2-km-diameter Meteor Crater near Winslow, Arizona. METCRAX stands for the Meteor Crater Experiment. Professor Muschinski is especially interested in “atmospheric seiches,” a term he coined for the bathtub-like, sloshing motion of cold-air pools in closed basins, such as the Meteor Crater. A deeper understanding of these seiches will help us improve the forecasting for air-quality and nighttime ground temperatures.
The abstract of the article reads as follows: The Meteor Crater Experiment (METCRAX 2006) was conducted in October 2006 at Arizona's Meteor Crater to investigate stable boundary layer evolution in a topographically uncomplicated basin surrounded by the nearly homogeneous plain of the Colorado Plateau. The two goals of the experiment were 1) to investigate the microscale and mesoscale structure and evolution of the stable boundary layer in the crater and its surroundings and 2) to determine whether atmospheric seiches or standing waves are produced inside the crater. This article provides an overview of the scientific goals of the experiment; summarizes the research measurements, the crater topography, and the synoptic meteorology of the study period; and presents initial analysis results. Analyses show that nighttime temperature inversions form frequently in the crater and that they are often perturbed by internal wave motions. Nighttime cooling produces a shallow (15–30 m deep) surface-based inversion that is surmounted by a horizontally homogeneous near-isothermal layer that extends all the way to the rim, where a second inversion extends above rim level. Seiches are sometimes present on the crater floor. The diurnal propagation of shadows from the crater rim produces important spatial differences in the surface radiation budget and thus the timing of the slope flow transition, and the crater atmosphere is often perturbed during nighttime by a southwesterly mesoscale drainage flow. (January 2009)
It takes a certain amount of nerd courage to don the massive headset recently assembled by a group of UMass seniors. But their augmented reality, "personal head-up display" (www.ecs.umass.edu/ece/sdp/sdp09/wolf/) will make you the go-to guy for directions around campus - even the big city.