Intelligence edge requires sharp eyes on ground and in sky
Geospatial information must be supplemented by local knowledge to ensure force safety
Tech Watch is a new feature that focuses on geospatial intelligence and cloud computing technologies and initiatives. This month’s installment draws on the observations and insights of senior program officers who, having returned from deployment in Afghanistan, came together at a panel at the GEOINT 2010 Symposium. Defense Systems Editor-in-Chief Barry Rosenberg asked them to offer some lessons learned based on their time in theater.
The participants were Army Col. Steven Beckman, assistant director of intelligence at the Joint Staff, who recently served as the CJ2 [intelligence chief] for ISAF’s Regional Command South in Kandahar, Afghanistan; Air Force Col. James Sculerati, chief of the Intelligence Support Division at the Special Operations Command; Air Force Col. James Gear, director of the Air Force’s Remotely Piloted Aircraft Task Force at Air Force Headquarters; and Army Col. Dave Tohn, who served as the National Security Agency’s Afghan Mission Network manager and is a military intelligence officer at Fort Meade, Md.The need for human intelligence as much as geospatial intelligence
Beckman: In , we had an instant where the insurgents drove a truck with a small trailer in the back underneath an overpass on Highway 1, which is the main highway that moves from Kandahar City out toward Helmand and, of course, it’s part of the ring road. That particular day, they actually blew up that overpass. Now, that overpass was a mile and a half away from a Canadian forward operating base that had a Rapid Aerostat Initial Deployment tower. It was under the presence of not one but two blimp versions; one was the Persistent Threat Detection System (PTDS) in Kandahar and another was a Canadian system.
Flying at the same time that that explosion went off was a Canadian Heron unmanned aerial vehicle, and not 20 minutes before it went off, a Canadian and an American patrol both went near that area. But there was no [human intelligence] to indicate what was going to happen. There was no [signals intelligence] data about it; therefore, there’s an explosion. The road gets blown up even though there were assets there that could have seen it.
Fast-forward to this year. About two weeks ago, the commander of Regional Command South, with the governor of Kandahar drove along Highway 1, all the way from Kandahar City out to a place called Band-e-Timor, which is near the Helmand boundary. What is noticeable about that is they didn’t drive in Strykers, they didn’t drive in up-armored vehicles, they drove in SUVs. And in that period, there had been a 14-day period where there had not been a single improvised explosive device strike along Highway 1.
Now, a lot has changed since 2009. I would say that the human intelligence got better. The ratio of both indigenous and International Security Assistance Force troops on the ground got better. You had persistent surveillance from a string of PTDS systems and more unmanned aerial vehicles, but what happened was there was a tying in of that local knowledge connected to the people by the troops on the ground [to] basically assist the technical methods that we have to ensure and prevent things happening.
Geoint to the foxhole
Sculerati: The challenge is when I’m pushing information out to a variety of points not to a single secure node in a planning cell on the end of a secure network. We’re talking about pushing out to a variety of guys. Getting information with [a device] on one hand and [having] a rifle on the other creates new challenges for information assurance. We’re going to have to get away from the traditional answer that we accept no risks and that network security only goes so far because [there is less security in the] last gap. The Army calls it the last tactical mile, but that even is an inadequate definition.
It’s pushing it out to a variety of users in a variety of environments. [In the case of Special Operations Command], it means operating with people we don’t normally share information with, people we don’t normally share technology with. So how do you get the information out to that sharp end? The stream [of info] has got to include a level of detail and can’t compromise itself…not in terms of security but in terms of chopping information off that that operator needs. They’ve built up a trust factor that now allows them to reach out and pull that information. If we start sanitizing it again then they’ll stop trusting it, and we’ll go through this rebuilding process again.
Chat rooms for UAV pilots
Gear: We’ve seen in a practice there’s a lack of standardization in chat. I would expect very shortly we’ll be working in conjunction with our service partners — the Army, Navy and Marine Corps — to develop communications brevity for the chat room. And not only that, you almost need your own personal discovery about where the chat room is that you need to go to accomplish your mission. We have Predator and Reaper crews right now that have 20 chat rooms opened up on multiple monitors. If you go into a ground control station today, there are over 10 monitors for people to observe. So not only do we need to tackle the communications brevity aspect of chat but also how those chat rooms are published and managed and who has control over them so you don’t have interlopers in there, as well. We don’t have it yet but it needs to at least start at that joint level with the other interagencies. We’re fleshing that out right now.
Tohn: [Something you have to] think about is that the [processing, exploitation, dissemination] cycles for the sigint and [imagery intelligence] are different. Clearly, [full-motion video] from an imagery perspective and geolocation from the sigint perspective is near real time and real time, and you can do the cross-cueing, [sharing of data between sensors without human interaction], there if you get it together. But if your sigint collect involves a linguist, it involves doing some transcription and gisting and then getting that back out in order to do a cross-cueing and tipping. You’ve induced lag that can be measured in minutes, tens of minutes or even more.