DOD develops radio technologies to overcome spectrum limitations

Picture this: A combat fire team fans out on a city block, and each soldier is connected to the rest of the platoon by a radio that provides location data and voice communications. Data and voice push each soldier through a sea of communications traffic, filling every available bit of electromagnetic spectrum. Each soldier in the platoon, company and brigade becomes a node on the network.

From a command vehicle, a radio feeds the troops ground situational awareness data, plugged into a larger ad hoc ground network that pushes video and voice data back to a company command post and directly into the Secure IP Router Network. At a regional command post and the Pentagon, analysts and commanders watch streaming video from a soldier’s view on their computers, provide supporting information, and issue commands to units to provide fire to support forces on the battlefield.

That’s the Defense Department's vision of a fully networked battlefield, made possible by technologies that DOD already is funding or through independent research and development by communications technology providers. Although DOD continues work on application-specific networks, such as the Force XXI Battle Command Brigade and Below (FBCB2) and the Future Combat Systems network, future battlefield communications will depend on getting data from each of those networks to individual soldiers as they need it.

At the frontier of the tactical network, the Defense Advanced Research Projects Agency’s Wireless Network after Next (WNaN) program seeks to fundamentally change the way DOD buys communications hardware and build a network that can support hundreds of connected users. BBN Technologies is largely responsible for developing the program's technologies. The major premise of WNaN is that low-cost nodes — digital communications devices on the network — can be used in combination with an adaptable battlefield ad hoc network to provide communications and data to each soldier, making communications on the battlefield more reliable.

“The way that DOD is currently acquiring communications systems is broken,” said Jason Redi, principal scientist at BBN's Network Technologies business unit. Redi is principal investigator on the Policy-based Information-centric Reliable Ad hoc Network (Pirana), a major part of DARPA’s WNaN program.

“Soldiers in the field — the guys at the furthest edge of the spear — don't have any comms,” Redi said. “The way the guys in fire teams communicate is with shouts and hand signals.” This is partially because providing radio systems to soldiers in combat is expensive with current acquisition methods, he said.

DARPA has recently initiated some staff changes for the program, making Bruce Fette a program manager. Agency officials said Fette would not be available for comment for this article because of his short tenure, and they deferred questions to BBN.

WNaN will not be compatible with the waveforms that DOD is standardizing through the Joint Tactical Radio System program, even though WNaN radios will be software-defined. JTRS’ Soldier Radio Waveform, the standard intended for JTRS-compliant handheld radios, is set low in the frequency spectrum, between 200 MHz and 400 MHz. “There are pluses and minuses for those lower frequencies,” Redi said. “They have much better propagation, but those frequencies are crowded because everyone wants to use them.”

On the other hand, the Pirana radio operates in the 600 MHz to 900 MHz band because it’s based on commercial components, which lowers the cost of the radios so that they’re inexpensive enough for every soldier to carry one, making them nodes on the battlefield network.

Redi compared the cycle for replacing radios to getting a new smart phone. “I get my PDA run over by my car, and I'm happy because I get to get the latest and greatest thing," he said. "But for the most part, the software that runs the network is the same.” If the same model is applied to military communications, soldiers could get radios based on commercial parts, “replaced every 18 months where people are able to get the best in the field at the time, and we're not defining the radios for the next 20 years.” he said.

The current Pirana network has been tested up to a throughput of 400 kilobits/sec among nodes. The modem used in the radio will get an upgrade in August that will boost that throughput to 2 megabits/sec, Redi said.

However, that environment creates new communications problems. Current radio networks can't handle a high density of users operating radios in a specific frequency range. “Going from one out of 30 guys having a radio to everyone having a radio [is] a dramatic increase in network capacity requirements,” Redi said. “You have to ask questions like how many radios can the network handle at once and how many guys can turn on their radios in the same room and have them still work.”

Part of the solution is using dynamic spectrum access (DSA). With Pirana, “we dynamically have your radios decide what frequency they should be using based on what the topological needs are,” Redi said. Those needs include whom the node needs to communicate with, what frequencies neighbors are using, the type of transmissions the user needs to send, and the type and volume of traffic that nearby users are generating.

Reliable radio communications also depend on identifying available frequencies, Redi said. “And using those things, we can try to determine what's the right assignment, and this happens dynamically all the time. Every few seconds we re-evaluate where we are in regards to frequency assignment. And we can therefore make the best choice for the mission as a whole.”

Shifting frequencies

To dynamically shift frequencies and maintain communications, troops need to have multiple transceivers so that they can stay connected, Redi said. If radios operate on only one channel at a time and one radio needs to switch to another frequency, all other radios on the network must follow. “That's clearly an unscalable mess,” Redi said. “Doing DSA really requires that you have multiple [frequency] slices available — at least two.”

The presence of multiple transponders allows radios to shift big data transfers to another, less-crowded network using DSA. That means two nodes sharing a data transmission — for example, a video feed — can shift one of the channels they operate to another, unused frequency to transmit the data without degrading the main communications channel.

The Pirana radio developed by BBN Technologies uses four transceivers. BBN and DARPA are trying to determine the optimal number of transceivers per radio as part of ongoing testing. Redi said they have performed tests with as many as 15 nodes in the lab. “We’re doing a 20-node demo in September, and another in December. Starting in early March [2010], we’ll be delivering 250 nodes to the Army for their evaluation. So we have a very, very rapid run-up.”

The military is already using DSA to deal with communications in congested areas. The Army signed a $12.5 million contract with Raytheon in April for the company’s Enhanced Position Location Reporting System Extended Frequency, an enhancement to the EPLRS radio that will add greater networking capability and DSA technology.

The radios “can automatically and continually monitor the spectrum of ops, detect and classify emitters, identify white space, and opportunistically use those white spaces,” said Tim Strobel, Raytheon's technical director of tactical communications systems.

“The more spectrum you can use, the more bandwidth you can provide to users," Strobel said. "And you can operate more harmoniously with other communication systems — the radios can use spectrum when [others] are not, then stop when they are without pre-allocating.”

Most important for the vehicle-mounted EPLRS, DSA allows radios to operate even if warfighters are using electronic warfare gear to disable remote-controlled improvised explosive devices, Strobel said .

Fewer dropped calls

Network disruption is another common problem for battlefield digital communications. Even during the best conditions, disconnections can occur — as most cell phone users regularly experience. Standard TCP/IP network routing drops packets to destinations that a router can’t reach. Although that’s not a problem with most networks, it creates a lot of problems for wireless ones because repeated transmissions of lost packets can quickly degrade a network’s performance.

“In a world where you're going to get a lot of disconnections, you're going to get a lot of packets dropped on the floor even though that connection might come back up in the next 5 seconds,” Redi said. "Imagine a fire team runs around to the other side of a building, and when they get to the other side, they may get reconnected. But in the meantime, all that data is lost. And that's actually how all of the existing tactical networks in the field operate."

The WNaN system resolves that problem with a technology named disruption-tolerant networking. Using a store-and-forward approach to networking, DTN keeps a cache of data sent in memory. When connections are dropped, a device can retransmit the data when new nodes join the network.

“The individual hops along the path are going to buffer for you to maintain everything even when the connections are flaky,” Redi said.

The Pirana radios have 128M of memory to cache transmissions for short periods. For larger amounts of data, the radios have standard flash memory cards, which have a capacity as large as 32G. That volume of space allows a cache to store information such as situational awareness data — for example, the position of friendly and hostile forces as indicated by the Blue Force Tracking and FBCB2 systems — that can be shared with new nodes joining the network.

Situational awareness is an excellent application for DTN, Redi said. Because the data generally has a specific life cycle, the cache can hold it for a specific amount of time and delete it when the information is out-of-date.

“The way a lot of situational awareness systems work is, when I plop the hostile icon down on my map, it floods the network once or twice, and that’s it,” Redi said. “So if someone joins the network an hour later, they don't ever see that information about where that hostile guy was. With DTN, without making modifications to the app, you can set up the app so the data is stored, and when someone connects, he can query and get all the data displayed on his SA display.”