Disaster Response

Two research teams are working to get no-hands data and phone service to rescuers and managers in the field

Two months after the Sept. 11 attacks, Virginia Governor Jim Gilmore convened a meeting of the Virginia Preparedness and Security Panel to assess the latest tools in the fight against terrorism. One of the technologies on the demonstrations list was a wireless disaster communications system developed at Virginia Tech.

Establishing a hub in the campus library, the VA Tech research team was able to show how wireless equipment could provide access to e-mail, phone service, Internet browsing and video-conferencing a building away, said co-principal investigator Scott Midkiff, a professor of computer science and electrical engineering at Virginia Tech.

There were lots of oohs and aahs, but the demonstration seemed to make a particularly strong impression on Chief Edward Plaugher of Arlington County Fire, a first responder to the Pentagon disaster. After the defense nerve center was rammed by a hijacked airliner, it took Plaugher and his team a week to establish an Internet connection at their command post, he said.

"The ability to make decisions from an informed standpoint rather than from your gut dramatically changes the outcome of the incident," Plaugher told the Roanoke Times.

One of the big problems in a disaster is that phone and data lines immediately go down, hurtling emergency workers back to the communications Stone Age of runners and written communiques. From a group of firefighters who worked a deadly fire in New Bedford, Mass., Midkiff's research team learned that cell phones often are no help, he said.

"Every day at 11, the press, TV people especially, grab all the available cell circuits, because they go on the air at noon," Midkiff recounted. "From about 11 to 1:30 or so you're without communication, and other times are busy too."

Even if you have phone lines, you can't necessarily use them to transmit data-rich but vital information, such as video, mapping, connections to Internet files and other geographic services, Midkiff said.

For a solution to the problem, Virginia Tech turned to a section of radio bandwidth known as Local Multipoint Distribution Services (LMDS), which commercial concerns have been trying to market, thus far with little success. Virginia Tech holds the license to the frequencies across a swath of Virginia, for economic development and research purposes, Midkiff said. Long before Sept. 11, Virginia Tech began talking to SAIC, a government IT contractor that works with the Federal Emergency Management Agency and the National Response Center, about using them for emergency services.

LMDS require line-of-sight to operate properly, which had discouraged their early use. One of Virginia Tech's innovations was to add a channel sounder to the equipment, to help workers pick the best transmitting location. The sounder has a second critical function: it constantly tests how good the connection is, and how much data can be sent under current conditions.

"You may have a good connection but someone drives a truck in and causes an obstruction and so you can move the equipment, or bounce the signal off the truck," Midkiff says. "Or perhaps you're in a situation where you can't get an optimal connection but a good one, that still allows data to be sent at a lower rate."

The researchers are working with a commercial product called E-Team, which is slated for use at the upcoming Winter Olympics in Salt Lake City. The product lets workers share information, integrate data and receive progress reports.

The research is far from over. The current range of the Virginia Tech system is two miles; Midkiff hopes to take it to 10 miles. And he wants to increase the data rate tenfold. While reaching for high capacity, it's also important to remember not to overload emergency workers with information, he said.

"The guy going into the building to rescue people, all he needs is a cell phone probably," Midkiff says. "There are very clear procedures, chains of command, actions taken that have been developed over time, that apply to a lot of disasters and they're very successful. We're trying to make that process work better."

In another test of mobile communications, a research team from the University of California, Santa Barbara and Iowa State University is investigating transmission of Geographic Information Systems or GIS in the field. The way maps, weather information and other geographic data are layered and stored in a GIS opens them to complex analysis, but restricted bandwidth, limited computer capacity and cramped displays have stymied efforts to use the voluminous data files out of an office or laboratory setting.

Principal investigator Sarah Nusser, a professor in the Department of Statistics and the Statistical Laboratory at Iowa State, is trying to put together a flexible architecture -- including a portable computer, Global Positioning System (GPS) receiver, cell-phone with Web access and display -- to do just that. The GPS links up to 24-space satellites to pinpoint geographic locations.

"The wearable computer sits in a side pocket of a fishing vest, the GPS is on the shoulder, and the display is clipped on to glasses, viewable in the peripheral vision," says Nusser, statistics/ ecology and environmental biology professor at Iowa State. "If you need to have your hands free, and you want to reference information at a moment's glance, that would be one way to deliver."

The GPS could be adapted to place a floating glyph or other easily read graphic symbol leading you to your destination, even outlining the building name and office number in bold, she says.

Nusser also looked into the use of the Twiddler 12-key (one-handed) keypad plus mouse for typing commands, in the field, but found it was hard to use without significant training. She's now considering voice technologies for issuing commands.

The biggest stumbling block is how and where to conduct the data-heavy computing to allow requests from field workers.

"What we've done is set up a handheld computer with a wireless connection to a laptop that can get on the network, put the request in a data warehouse, manipulate and send the answer back to the client," Nusser says. "You could have the laptop in your truck, and be out walking on terrain with a handheld to send commands back to the laptop."

Nusser has been working with government field surveys, but says her research could easily translate to disaster response.

"If you have a live disaster and injuries, you might be able to access somebody's medical information, pull down a map of hospital distribution through a city and select the most appropriate facility and how to get there. Or you could go and get the latest satellite view of the disaster area, find out you can't get to this hospital, but you could get somewhere else with a physician of the same specialty."