Fail-safe field communications nearly ready
DARPA expects to deploy Disruption Tolerant Networking to strengthen battlefield communications by next year
- By Brian Robinson
- Aug 03, 2009
The Defense Advanced Research Projects Agency is close to deploying a program that has been in the works for five years and aims to give troops in the field network communications that won’t fail regardless of the operating environment.
DARPA is completing the third and final phase of the Disruption Tolerant Networking (DTN) program, with every expectation that the technology will quickly become a necessity for military communications.
“Within three years, you probably won’t see a procurement that doesn’t have DTN as one of its requirements,” said Jason Redi, a principal scientist at BBN Technologies, the program's lead developer.
DARPA officials said they expect the military services to deploy the technology in 2010, if not earlier.
Several other DARPA programs — such as Wireless Network after Next, Mobile Ad Hoc Interoperability Network Gateway and Optical RF Communication Adjunct — are already using DTN technologies to provide reliable communications on the battlefield.
BBN received the $9 million final-phase award in August 2008 to integrate DTN into military networks and investigate ways of building secure large scale, self-organizing networks. Those activities are due to be completed by the end of the first quarter of 2010.
The first two phases developed technology for reliable on-the-move and over-the-horizon communications.
Unlike traditional TCP/IP networks, which send data in packets via a secure pathway, DTN organizes the information flow as bundles that can be forwarded to a recipient through intelligent custody routers using whatever means are available — terrestrial, satellite or even network nodes that are only occasionally available, such as those on unmanned aerial vehicles.
That approach sidesteps the terrain, weather and movement obstacles that often disrupt wireless communications, particularly on the strategic edge of the network.
Furthermore, DTN does not need to send messages to a specific numbered address, as a TCP/IP network does, Redi said. The message bundles can be addressed simply — for example, to all company commanders within a kilometer of the originating system.
“You do need a good naming structure for [the DTN system] to work, but the [Defense Department] is one organization that already has a detailed structure for how naming is done,” he said.
DTN does for tactical and disrupted networks what TCP did for wired networks, DARPA officials said. Both take the burden of ensuring reliability off the application and its designer and put it onto the network.
“This is extraordinarily powerful,” the officials said in a statement. “It simplifies applications, it results in more robust applications, and it eliminates the need to build stovepiped reliability into the application, generally not a good idea in any case because applications wind up being used in ways and in environments that were not anticipated by the designers.”
DTN can be applied as an overlay to existing TCP/IP networks, where it would run natively in the application layer. Alternatively, it can run underneath regular networks where IP packets are translated into DTN bundles, meaning applications don’t have to be modified to work with DTN.
The drawback to the latter approach is that DTN users are limited to the network’s IP address system, which means they can’t be as creative as they could be with native DTN applications, Redi said.
The biggest hurdle to getting the technology into production systems is figuring out where to put DTN in the overall networking infrastructure to get the best results, he added.
Beyond that, the barriers to implementing DTN are small, he said. The additional impact on server/router CPUs is fairly minor, while the demand on other resources, such as storage, is similarly low.
“Even with limited resources, users will see an immediate increase in the benefits of using DTN,” Redi said.
DARPA officials said the Army and Marine Corps are performing field experiments to evaluate and validate DTN’s performance benefits.
No follow-on programs have been approved so far, but DARPA officials said there is considerable interest in content-based networking — a mechanism for accessing information and military units without knowing in advance where they are — and security technologies developed under the DTN program.
Brian Robinson is a special contributor to Defense Systems.