Homeland Security’s high-tech gamble

DHS’ low-profile “skunk works” seeks game-changing technologies to thwart tomorrow’s terrorists

Big successes, sparkling failures

Radical ideas can spawn game-changing homeland security technologies. They can also ignite spectacular flameouts. Sometimes, researchers must toil for years to learn their work’s fate. Other times, the answer comes all too quickly.

In 1989, Doug McMakin and a small group of other engineers at the Pacific Northwest National Laboratory theorized that security checkpoint screeners could use millimeter waves — part of the electromagnetic spectrum above microwaves — to reveal both metallic and nonmetallic objects, such as plastic explosives, concealed in clothing. “We believed we could do it, but we weren’t sure,” said McMakin, a staff engineer at the lab.

Steady funding from the Federal Aviation Administration, the Transportation Security Administration and DHS’ Science and Technology Directorate supported his team’s breakthroughs in writing the complex mathematical algorithms and designing the full-body cylindrical scanners required for real-world millimeter-wave security applications.

Since 2003, the operational scanners have been proving their security mettle under extreme conditions: The U.S. military uses them to secure Baghdad’s Green Zone, including the courtroom where Saddam Hussein’s trial was held. Commercial passenger-screening versions help protect Phoenix’s Sky Harbor airport and are being tested in New York and Los Angeles airports.

Now McMakin is turning his attention to a new type of technology for detecting concealed weapons before subjects enter screening areas. He declined to identify the technology, only saying that “it will be different than what we’ve done in the past.”

No future for futures forum
By contrast, the Defense Advanced Research Projects Agency quickly learned the fate of its bold but inflammatory plan to create a futures market to predict the next terrorist attack.

For a short time in 2003, some homeland security officials believed terrorist attacks might be predicted using a futures-trading market modeled after those that forecast prices for commodities such as corn and petroleum.

The monetary rewards possible in a futures-trading system provide an incentive for intelligence officers to voice honest opinions rather than pandering to what their bosses wanted to hear, Robin Hanson, assistant professor of economics at George Mason University and a key contributor to DARPA’s effort, said in a 2004 Federal Computer Week interview.

However, DARPA’s Policy Analysis Market ignited a public-relations firestorm as some senators bristled at the potential for traders to profit from terrorist attacks.

Although DARPA scuttled the project, the idea didn’t die. The University of Iowa Electronic Markets continues to forecast election outcomes. And a commercial trading forum called Intrade (www.intrade.com) lets speculators bid on a variety of political topics, including the date of Osama Bin Laden’s capture or neutralization. The smart money says that won’t happen sooner than next summer.

— Alan Joch

The Web’s dark alleys

Law enforcement officials know that in the past two years the number of Internet pages devoted to terrorism content has increased. They just don’t know how great the increase has been. However, some long-term research related to artificial intelligence offers a clue.
The University of Arizona’s AI Lab uses grants from the National Science Foundation to probe a terrain that lab director Hsinchun Chen labeled “the darker side of the Web.” Chen described his research field as security informatics. For the past five years, he has led a research project to develop a super search engine that can crawl sublayers of the Web and copy terrorist-related text and multimedia files from Web sites, chat rooms, blogs and e-mail messages.

“My projection is that there are about 50,000 Web sites with terrorist content,” Chen said.

A related project at the lab involves an expanding set of automated analysis tools designed to help intelligence and terrorism researchers make sense of the content on those Web sites. Some programs analyze the social networks that link the sites together. Those analyses can scour millions of postings in a few minutes to link communications among forum contributors and identify group leaders.

Other analytics categorize Web sites according to whether their main purpose is for recruiting, training, ideology sharing or communications. Researchers can also rank the violence characteristics of sites based on the language and ideas they display. The software can track an individual’s postings over time and alert authorities when someone is becoming more prone to violence, Chen said.

The objective isn’t simply to identify terrorists and shut down radical sites, Chen said. The real challenge for authorities is how to use the information to impede Web-based recruiting and training activities.

“Al Qaeda is on the Web, and it’s doing an excellent job of recruiting people and inciting the young disgruntled men of the world,” Chen said. “We need to stop the ones who are 13 or 14 years old from changing and becoming radical. We need to fight this war in the cyberspace for their hearts and minds.”

— Alan Joch

Sobering food for thought

One tool in the Homeland Security Department’s Science and Technology Directorate’s efforts to promote leading-edge research and development is the Office of University Programs and related Homeland Security Centers of Excellence, which gather groups of higher-education researchers to tackle tough security topics.

Texas A&M University is the leader in one of those efforts, the National Center for Foreign Animal and Zoonotic Disease (FAZD), which recently received a three-year, $18 million grant in part for protecting livestock from naturally occurring or terrorist-induced outbreaks of foreign animal diseases.

“DHS asked us to organize the center around the foreign animal diseases, essentially those that are exotic to the U.S. now” said Dr. Neville Clarke, director of FAZD. “And also then to focus on the zoonoses, those diseases that are transmissible between animals and man.”

Topping the list are the highly infectious foot and mouth disease, avian influenza and Rift Valley Fever (RVF), a zoonotic disease native to the Horn of Africa. Although not as deadly as other infections, RVF worries authorities because even an isolated outbreak could quickly spread throughout the United States by long-haul trucks.

FAZD is developing statistical models that track how animals move across the interstate system to quickly determine where a sick animal spent the previous day. The model will also let researchers evaluate the value of different methods of preventing or confronting the disease.

“We basically have been developing a marriage between the epidemiological models and economic models,” Clarke said. “If [an outbreak] happens, then what’s going to be the economic impact not only of the disease under varying levels of infection, but also [in terms of] the expense of developing a vaccine?”

He calls those models a work in progress. “We don’t really know how large the errors are in the estimates that we are making, but we think that we are ball-parking the outcomes in reasonable ways,” he said. “That gives [authorities] more basis for making decisions than they’d have if they were just going by the seat of their pants.”

- Alan Joch

Technology takes a hit

The Homeland Security Department’s total budget has grown in recent years, but money for its Science and Technology Directorate, its primary source of funding for technology research and development, has declined sharply since fiscal 2006.

Total fiscal 2008 DHS budget: $46.4 billion.
Change from previous year: 8.4 percent increase.

Total fiscal 2008 Science and Technology Directorate budget: $800 million.
Change from previous year: 18 percent decrease.

Total fiscal 2007 Science and Technology Directorate budget: $973 million.
Change from previous year: 35 percent decrease.

Source: Homeland Security Department

Hsinchun Chen

Title: Director of the Artificial Intelligence Lab, University of Arizona.

Research: Develop tools to help law enforcement and intelligence analysts track terrorists and extremists online.

Years devoted to the project: Five

What sustains my passion for this work: “Making a positive impact. We are losing the war on terrorism in cyberspace.”

Why game-changing research and development is important
to homeland security:
“Homeland security is replying on outdated information technology while terrorists are using advanced Internet tools.”

What I do when I’m not working: Jogging, reading, playing tennis and basketball, and cooking — mostly on Saturdays and Sundays.

Dr. Neville Clarke

Title: Director of the National Center for Foreign Animal and Zoonotic Disease.

Research: Developing analytical models for food supply safety.

Years devoted to the project: Three years for the Rift Valley Fever models.

What sustains my passion for this work: “The needs of the country are obvious once you get inside the intelligence community and look at what the threat is. The various scientific disciplines that are involved in solving these problems are probably about as exciting as they are ever going to be.”

Why game-changing Research and Development is important to homeland security: “Modern biology offers us the possibility to develop methods for understanding the fundamentals of the disease so that we build response or prevention mechanisms that are not disease specific, but that protect against whole classes of disease. An example would be an anti-viral that jump-starts the immune system into action in a matter of hours, and when you bring that immune system up it’s looking for all the bad critters, not just one.”

What I do when I’m not working: Jogging, playing tennis and backpacking.

Imagine that instead of anthrax-laced letters targeted at members of Congress the next bioterrorist attack to hit Washington is a wide-scale release of a toxin in the transit system. But rather than trusting a haphazard series of stationary air sensors installed at likely release points on platforms and waiting areas, first responders minimize the assault using thousands of mobile biodetectors embedded in a standard tool in every commuter’s arsenal: the cell phone.

According to such a plan, a portion of the phone-toting population would voluntarily use devices that included minuscule bio, radiation or chemical sensors that could detect dangers in real time. If terrorists released a toxin, cell-phone sensors would detect the substances and signal the threat to District of Columbia police via the Global Positioning System network.

Officials would quickly know the type of outbreak they faced and could pinpoint the release points and map how prevailing air currents were spreading the poison.

“Pretty soon you know there’s a botulism release at McPherson Square,” said Rolf Dietrich, deputy director at the Homeland Security Department’s Science and Technology Office of Innovation. “The police could push through the news to anybody in the vicinity, telling them: ‘We have indications of a potential problem. Evacuate to the south, because the wind is blowing this stuff to the north.’ ”

Privacy fears aside, there are some major problems with this idea. Sensors don’t yet exist that can identify biohazards in real time, and today’s technology is not small enough to fit inside compact cell phones. “People won’t want to use [the sensors] if they’re going to drain their battery too fast [or be] the size of a brick,” Dietrich said.

But in his S&T Office of Innovation role and related duties as director of HomeWorks, DHS’ relatively low-funded projects or “skunk works” in the battle against terrorism, Dietrich spends his days contemplating what-if solutions that may or may not ever overcome their steep technical hurdles.

Projects like the one described, dubbed Cell-All, are risky beyond the obvious risk of resource investments that might never pay off. They are the efforts of DHS’ S&T Directorate, which has been the target of sometimes withering criticism from Congress and others.

A report released last January by the conservative think tank the Heritage Foundation states that five years after its creation, the directorate’s “track record is clear enough, and it is not good.” The report adds that the directorate’s initial strategy of creating a research and development arm that serves the needs of the entire homeland security community has failed. “Its over-reaching mission has created a sprawling plethora of activities that does nothing well. The S&T Directorate requires a more focused mission, new thinking and a more streamlined organization,” it concludes.

James Jay Carafano, senior fellow at the foundation and one of the report’s authors, has since tempered his criticism, saying he is encouraged but still skeptical that Jay Cohen, former chief of naval research who became the directorate’s undersecretary in 2006, can correct those shortcomings.

“Anytime Jay Cohen talks, you hear a lot of things that sound like [the directorate is] moving in the right direction,” Carafano said. “But the proof really is in the pudding. I’m still on the sidelines as to whether this is just a kind of happy talk or whether this is a substantive movement forward for making S&T more of an integral part of the department and getting more value per dollar out of what they do.”
Over-reliance on the types of technologies being promoted by the directorate may not make the country safer, no matter how innovative it is, others have warned.

& amp; amp; amp; amp; amp; amp; amp; amp; amp; amp; ldquo;It’s important to look at how technology i mpacts your whole system,” said Henry Willis, a policy researcher at Rand, another think tank. “For example, port security can’t focus only on the ability to detect illicit substances and goods. It also has to look at how the technology used to do that affects the flow of goods. We can keep bad stuff out 100 percent of the time if we just don’t allow anything into the country. But clearly, no one wants to do that.”

Others remain critical of spending priorities. “We’re not investing enough in R&D for cybersecurity in the S&T Directorate,” said Rep. Jim Langevin (D-R.I.), chairman of the Homeland Security Committee’s Emerging Threats, Cyber Security and Science and Technology Subcommittee. “It’s a glaring vulnerability given that we’ve had a number of penetrations across our networks,” including at DHS. Investments in long lead-time projects, such as those sponsored by HomeWorks, are valid as long as the federal government also increases spending for cybersecurity basics, such as firewalls and intrusion-detection devices, he said.

Scarce R&D dollars
Congressional spending priorities have meant that R&D dollars are getting harder to attain. Although DHS’ budget grew 8.4 percent from fiscal 2007 to fiscal 2008, the directorate’s share — about $800 million or 2 percent of the total — dropped 18 percent.

Dietrich’s staff, which includes the Homeland Security Advanced Research Projects Agency (HSARPA), patterned after the Defense Department’s Defense Advanced Research Projects Agency R&D agency, parcels out those funds for projects into two broad categories, defined by their expected lead times.

The Homeland Innovative Prototypical Solutions (HIPS) category targets innovations that produce prototype demonstrations of game-changing technologies in two to five years, Dietrich said. Examples include Project Chloe, which would involve high-flying drones with missile sensors and lasers designed to protect civilian aircraft from shoulder-fired rockets. Sensit, another HIPS initiative, could result in a variety of sensors for detecting dangerous liquids packed in tourist luggage.

But it is in the category of High Impact Technology Solutions (HITS), including Cell-All, where DHS rolls the R&D dice. Those projects may require three years to produce a basic proof of concept. “We recognize [these projects] could have a considerable risk of failure,” Dietrich said. “But if they succeed — wow.” 

Thinking in true skunk works fashion, Dietrich doesn’t expect all HITS projects to reach fruition. He estimates that less than half of the eight initiatives now under way will achieve their original objectives. Nevertheless, like Apollo Program R&D that serendipitously spawned spinoffs such as CAT scans and MRI medical imaging equipment, Dietrich said, homeland security research efforts may benefit in unforeseen ways even from HITS failures.

HITS projects receive only 1 percent, or about $8 million, of the S&T Directorate’s budget. “I get to have the really exciting stuff, but most of the undersecretary’s money is not mine,” Dietrich added. His staff doesn’t perform research, but instead its members act as project managers to contractors from industry, academia and government-backed national laboratories. HITS projects provide funds to one university, two labs and five companies. Because HITS started less than a year ago, it’s too soon to tell where successes may come from. “I don’t have anything [I can hold] in my hand that is a product of HITS,” Dietrich said.

Can you ID me now?
For Cell-All to succe ssful y re ach c mmut ers&r quo; phon s, r esear hers must develop real-time biosensors, a significant advancement from today’s technologies that require days of lab work to confirm the identity of hazards.

“I’m looking for so m thing that within a matter of minutes can detect, analyze and transmit a signal that says this is anthrax, botulism or whatever,” Dietrich said. “A lot of people have told me that is impossible; that the state of the art isn’t even close to that.”

One scientist who said she doesn’t believe real-time biosensors are impossible is Rita Colwell, whose tenure as National Science Foundation director spanned the 2001 terrorist attacks and subsequent anthrax poisonings. Now a professor at the University of Maryland and the Johns Hopkins School of Public Health, Colwell said those events convinced her that homeland security efforts require modern sensors that can quickly identify pathogens.

“If you know what the organism is and you know its genetic composition — whether or not it’s drug-resistant, whether it’s been genetically engineered — then you have the power to successfully mobilize the response to that threat,” she said. “Right now, [we rely on] a kind of experiential guessing.”
At best, today’s identification techniques, which attempt to match bits of DNA from known pathogens, require a day or two of work. “With anthrax, it took us a week to find out what was going on, and we still have not determined the culprit,” Colwell said. “I want to develop tomorrow’s technology.”

Colwell’s research focuses on probing the DNA and RNA of samples using a technique known as direct sequencing and having the tools to conduct those analyses in an hour or less.

“What I want to do is encode the DNA and make a genome sequence match,” Colwell said. “That would provide us with as accurate an identification as we can perform.” DHS funds her work, but she declined to specify the amount.

Adding to the technical challenge is Colwell’s goal of creating a handheld sequencing tool that first responders can easily bring to an emergency site. Samples could come from the air, water or blood, she said. However, her prototypes, now the size of a small refrigerator, are “a little, how shall I say it, clunky,” she said.

Nevertheless, Colwell said, she is encouraged by the interdisciplinary nature of modern science, where engineers, biologists, mathematicians, chemists and sociologists might all collaborate on security solutions.

Twenty-first-century science “is driven by a holistic systems approach, and that’s the approach that we feel is the most powerful to get this job done,” Colwell said. “We are looking at three to five years as a reasonable time to get prototypes and to achieve [our goals of] accuracy, precision, rapidity and reliability.”

Tunnel vision
In addition to Colwell’s work, HomeWorks includes another project that addresses the homeland security challenge of securing borders against terrorists or smugglers who may dig tunnels to bypass border agents. “In many places, there’s an American town on one side of the border and directly opposite is another town in Mexico,” Dietrich said. “Smugglers will buy a house on both sides and dig a tunnel between them. Through that tunnel, you can smuggle guns and people.”

He said the tunnels uncovered so far have all been found either by luck or intelligence efforts. “There is no technology that exists now to find those homes,” Dietrich said. So HomeWorks is offering contract opportunities to researchers who can provide viable ideas for detecting tunnels perhaps through detecting gradient changes in gravity caused by the lower mass of a tunne com ared ith he surrou ding area

&l quo;There are several technologies that have been proposed, and we are evaluating them to see which ones we are going to pursue,” Dietrich said.

If researchers discover a solution, they may claim inspiration from an unlikely source — a loose collection of science-fiction writers that belong to a pro bono o ernment advisory group known as Sigma. Border security was one of the discussions that writers such as Greg Bear, author of 30 science fiction books, including the best seller “Darwin’s Radio,” and Jerry Pournelle, co-author of “Footfall and Lucifer’s Hammer,” joined at a meeting with HomeWorks officials in May.

The Sigma group’s ideas ranged beyond magnetic fields to include a radical plan to drop 10-foot probes from planes flying along the border. The probes would be designed to stabilize themselves in the air and perhaps be propelled by onboard rockets to hit the earth with enough force to bury themselves in the ground. Once embedded, the probes would monitor underground sounds and vibrations that reveal any tunneling work.

For Sigma founder Arlan Andrews, a writer of sci-fi short stories and acting environmental program director at the Naval Air Station in Corpus Christi, Texas, there’s a natural affinity between HomeWorks and the futurists in his group. “Science-fiction writers have always lived in the future,” he said. “We’ve been there, explored it and can come back and share it for the national
interest.”

Joch is a business and technology writer based in New England. He can be reached at ajoch@worldpath.com

2014 Rising Star Awards

Help us find the next generation of leaders in federal IT.

Reader comments

Please post your comments here. Comments are moderated, so they may not appear immediately after submitting. We will not post comments that we consider abusive or off-topic.

Please type the letters/numbers you see above