Paul Benda must not sleep well. As director of the Homeland Security Advanced Research Projects Agency, his job is to wring everything he can from a science and technology budget that took a 50 percent whacking in 2012. Benda is hopeful his budget will bounce back to $485 million in 2013. Even if it does, his bottom line remains this: U.S. spending on homeland security technologies is inadequate for the long list of terrible ways the country could be targeted. The list includes biological attacks; agricultural sabotage; loose nukes; cyber attacks on banks or other infrastructure. Benda discussed his strategies for making progress despite a tight budget. Excerpts>>
Director Paul Benda, Homeland Security Advanced Research Projects Agency
Why do we need HSARPA?>> It’s statutory law as part of the Homeland Security Act. There’s an answer for you. It turns out that the challenges that DoD and the intel community face are completely and utterly different than the homeland security challenges we’ve got. If you look at counter IED tools that JIEDDO (the Joint Improvised Explosive Device Defeat Organization) develops, and that DARPA develops, they work great in theater. They can blank out any radio waves as they roll down the street. If I blanked out the rf frequencies rolling down Pennsylvania Avenue, do you think that would go over very well? No it wouldn’t. What we want to do at HSARPA is leverage those investments that DoD and the intel community make, and translate them to a homeland Security application.
Out of sync>> I think our investments are tied to (intel analysis). I don’t think our budget numbers are tied to that.
Competing interests>> Is the $485 million for S and T sufficient? I have to detect submersible, semi-submersible vehicles; I have to do bio terrorism; I have to do cyber security; I have to do search and rescue missions; I have to do FEMA disaster response; I have to support U.S. Secret Service operations; I have to do data analytics for ICE (Immigration and Customs Enforcement); do counter-proliferation missions. My job really is trying to leverage what is out there either with our DoD partners or with our national lab partners and trying to make them cheaper and accessible to my homeland security operators.
‘Existential threat’ >> BioWatch was a program that was stood up back in 2002, 2003. We still fundamentally use the same technologies that we did 10 years ago. Even Gen 3 (sensor technology) that’s coming out, while it’ll reduce the time it takes for us to detect an attack, it still detects it after the fact. If you’ve got a terrorist who can produce three kilograms of anthrax and can attack one city at a time, what technologies exist today that allow us to detect that attack in real time? There are none. If I can’t detect it in real time, my odds of interdicting them and stopping them from doing future attacks are pretty low. Our investment in that area, while not insubstantial, isn’t where it needs to be…that (bio attack scenario) is an existential threat to the nation, quite frankly.
Cyber insurance >> If you look at a significant cyber attack that could take down part of the financial industry, the economic impact of that attack could easily reach into the billions. So, do you think $60 million dollars as an insurance policy for that makes sense?
2012 budget cut>> I’m hopeful in the discussions that we’ve had with Hill staff and with others that they view that cut as a mistake. They felt that that was too draconian of a cut for S and T. Where they want to take the department in terms of efficiencies, they recognize the need to have investments in technologies to get there.
Tough calls>> All of our R and D investments go through a portfolio review process where we have set up this assessment framework. We bring in outside experts as well as our internal customers, and we grade the programs after a briefing by our program managers. How well is that program aligned with this assessment framework? How well is it meeting goals? When we received a cut like that (50 percent cut in 2012), having this portfolio review process allowed us to really scale back to those highest performing programs. But in the end, we ended up cutting good work as well, because when you cut 50 percent you’re getting down to the bone at that point.
Re-starting programs>> In terms of our border security programs, we really cut back on tunnel detection and activity monitoring. Even though those were good programs and necessary, DoD has a fairly significant investment in that. We were trying to maintain forward progress with zero money by leveraging DoD’s work in that area. That’s one where we’ll probably put some money back in so we can pilot some of those technologies specific to DHS missions and DHS needs.
Tunnel challenge>> You’ve got a border town on the U.S. side and the Mexican side, and they share storm sewers. People always wonder, ‘Well why don’t you just put bars on the front ends of them?’ Well, if someone gets in that tunnel, and gets trapped and the water’s coming, they drown. Even if you put bars on them, they can cut them. So, how do we measure if there’s people in those tunnels or not? And then how do we give (border patrol officers) situational awareness and maintain officer safety if someone goes in there.
Secure Border Initiative lessons>> I was part of the team that did the follow-on analysis of alternatives that resulted in the recommendation to the secretary of killing that program (SBInet). Lesson one is to make sure we don’t presuppose a technology solution. People thought these integrated fixed towers with the radars and the cameras would work across the entire border. I think that CBP (Customs and Border Protection) vastly underestimated the amount of effort required to create a fully integrated system that could do that in terms of the comms challenges.
HSARPA’s role on SBInet>> It didn’t actually (go through HSARPA). The Science and Technology Directorate has the statutory authority to coordinate all R and D in the department. Would you call SBInet an R and D program, or would you call it a technology integration program? It turned out that there was a lot more R and D needed than anyone expected, but I think they (CBP) originally thought it was a commodity buy kind of thing – at least Boeing certainly sold it that way.
New approach>> Look at border security between points of entry – the SBInet problem. How do border patrol agents do their work today? Well, they’ve got patrols, they’ve got dispatch, they might have underground sensors. We map that out in the operational context: ‘This is how you do your work today. How can technology help you do your job better?’ It’s a subtle shift but we’re not asking the border patrol guys, ‘Hey do you guys need a better underground sensor or camera?’ We’re actually saying, ‘So this is how you do your work today, right?’ Hey, I was talking with my DARPA friends over there, they got this high tech sensor. What if we replaced this sensor with that sensor?’ We go through that kind of systems analysis.
Tensions with component agencies>> The story’s much different than it was six years ago. The components (Customs and Border Protection; Immigration and Customs Enforcement; Transportation Security Administration, etc.) are very interested in working with us in terms of actually providing us funding to help us do our jobs. Whereas in the past, they were okay going off on their own.
New ‘R and D strategies’>> We’re interviewing (component managers) one level below the component head. We want to do the systems analysis that says, ‘How do you do your work?’ and then pick where we can have the most impact with the littlest R and D investment. Then we propose technologies we think can help them solve those priorities. We’re doing this for every component within DHS.
Smartphones>> Our communications infrastructure can be spotty, especially on the border. Border patrol officers will use radios, and where they don’t have radio comms, they’ll use a personal cell phone. You can’t let the lack of comms completely degrade your mission effectiveness. We have to be aware of where to use smartphones. It would be great for them to have a smartphone that knows where the sensors are and they can have better situational awareness. You’d love to see as these integrated fixed towers go up so that the officers have access to that video feed. If we ever get to the point of having tactical UAVs flying along the border, you’d want these guys to have access.