DIRECNET MESH NETWORKING STANDARD
WOULD PROVIDE HIGH DATA RATE
COMMUNICATIONS WITHIN HUNDREDS OF MILES.
A group of industry players is proposing an open-standard directional networking system designed to provide 1-gigabit-persecond data communication with anyone in a network on the ground, in the air or at sea, within hundreds of miles.
The planned mobile mesh network, known as DirecNet, will use a networkfriendly data link waveform to provide secure Internet Protocol communications up to 1 gbps for all types of air and surface platforms.
The 1-gbps wideband infrastructure compares with the Common Data Link’s (CDL) current capability of 274 mbps. The higher data rates would easily accommodate bandwidth-hungry data such as video or audio.
DirecNet would use fast-steered directional antennas to substantially boost link power and operating range, and to permit reuse of radio frequency (RF) spectrum. Any DirecNet node can serve as a relay. This will multiply connectivity and extend the range to beyond line of sight.
“We consider this a revolutionary, very high data rate network,” said Gary Nault, vice president for business development and advanced programs for the communications and electronics business unit within Cubic Defense Applications.
Nault said he anticipates that the industry consortium now being formed will release a DirecNet specification in two years.
“DirecNet essentially enables a gigabit data rate at hundreds of miles. That immediately distinguishes it from the other omni-directional networks that are being developed within DoD,” Nault said. Those other networks include the Wideband Networking Waveform, which is at most a 10 mbps data link, according to Nault.
Potential founding members of the DirecNet consortium are Cubic Defense Applications, Raytheon, Rockwell Collins, L-3, Harris Corp., BAE Systems, Boeing, NGC, Herley Industries and ITT Industries. Other interested companies include Innocon and Nova Engineering.
DirecNet meshes with the Global Information Grid (GIG), Joint Tactical Radio System (JTRS) and other networks. Its domain is “the edge of the GIG, or ‘the last mile,’” said Nault. There would be DirecNet terminals on aircraft, land vehicles and ships to close that last mile.
“It’s akin to the WiFi in your laptop,” Nault continued. “Except, instead of communications hundreds of feet, you’re in a network with other nodes that are potentially hundreds of miles away at speeds that are orders of magnitude faster.”
In addition to higher data rates, Direc-Net would substantially improve interoperability, remedying the stovepiped networks that commonly only allow for point-topoint communications, rather than communications among all players.
“While we wouldn’t anticipate that every Humvee within the Army inventory would have a DirecNet terminal, certain ones would, like the ones that need to communicate with UAVs,” said Nault.
LIGHTING BOLTS
DirecNet solves the “lightning bolt” problem, explained Nault. This refers to the lightning bolts that commonly appear on military and contractor presentations and publications showing wireless connections from one node to another.
“Right now it’s not clear, based on the amount of data that needs to go from node to node, what that node looks like,” Nault said. “That’s why we’re so excited about DirecNet. It has the ability to transmit all of the information from mobile nodes in a way that’s not proprietary and that any of us can build to.”
An industry working group is forming a consortium to develop the non-proprietary standard. The vision is to “standardize directional, IP-enabled, highly mobile ad hoc mesh networks through a validated open specification.”
The founding members will sign an agreement with the Open Group, which formalizes the DirecNet consortium. The Open Group, a “neutral party” among the industry competitors, has been instrumental in establishing other consortiums such as the Network Centric Operations Industry Consortium.
The technologies already exist; the consortium’s main objective is to establish an open standard in order to apply those technologies. The consortium’s task is not really to develop technology, but to develop the specification for a standard, so that one company’s terminals can interoperate with another’s.
“Our belief is that we could collaboratively create a specification that was stronger than any one of us could write individually,” said Bud Jewett, manager of Navy business development at Harris. “It would be done in an open environment. So all of us could build to that standard to create interoperable hardware in a competitive marketplace.”
Many CDL waveforms are not interoperable. “The standards have been liberal enough to let us interpret those differently, and make them vendor specific,” added Jewett. “The DirecNet task force will be working to develop an open-standard waveform that allows for mesh networking, wide bandwidth, efficient transmission, beam steering, multiple beams, low data loss and high data compression. That’s the technical challenge that we will work on in a collaborative environment.”
Harris is currently working on several waveforms that respond to many of these requirements. “We on the task force,” observed Jewett, “are optimistic that we can combine the best technology that each of our companies can provide into the standard that would result from our collaborative work. And DoD appreciates that we’re attempting to build in interoperability as part of the challenge of this new mesh network waveform.”
“The other major challenge,” he continued, “is backward compatibility with legacy waveforms. DoD has already invested in many expensive CDL systems and can’t afford to replace them all with improved hardware. As we move to reprogrammable radios, new waveforms may be only a matter of loading new software to enable new networks. That’s a ways off, but is the expectation for the future.”
NODES IN THE MESH
Under DirecNet, manned, unmanned, fixed-wing and rotary aircraft would be equipped with directional antennas. A control local area network (LAN) would interface with those antennas and allocate RF spectrum. The directional antennas would boost link power and range. On the ground, LAN users would transmit data via wideband communications to DirecNet terminals, which would route the data. Narrowband data links could also join the network. In the event the spectrum was saturated on one route, an alternative path automatically would be found.
DirecNet needs a fast-steering beam. There are several ways to rapidly steer a beam, but prominent among them are active electronically scanned arrays (AESAs). These are air- and ground-tracking radars currently mounted on F-22 and F/A-18 aircraft. In addition to performingradar functions, they are capable of highspeed data transmission when used with DirecNet terminals.
ITT Electronic Systems and Cubic have carried out three proof-of-concept demonstrations with ITT’s Active Electronic Scanning Antenna array, which enables directional networking. ITT’s fast electronic beam steering antenna provides connectivity with simultaneous multiple independent beams over multiple frequencies at high data rates in both full duplex and half duplex modes (where transmitting and receiving can occur either simultaneously or non-simultaneously).
“Our AESA antenna provides dynamic networking solutions for tactical targeting networking for applications at sea, applications in ground future combat systems and in ad hoc meshnetworks for effective integration into the GIG,” said ITT Senior Business Development Manager Howard Elovitz. “Our AESA enables directional networking to communicate at all times with all elements within the battlespace.”
“The benefits of directional networking are low-probability intercept and covert operations that require high bandwidth and high data rates, while maintaining continuous secure connectivity, achievable with our multi-link antenna system,” Elovitz said, adding that ITT’s AESA has built-in scalability features that make it suitable for network-centric applications.
The ITT-Cubic demonstrations included nodes mounted on helicopters, ground vehicles and ships; four nodes can be communicated with simultaneously. The antennas were used with Cubic’s Tactical Common Data Link (TCDL). In the proofof-concept events, the data rates were the current CDL speed of 274 mbps, but Elovitz said the technology is capable of 1 gbps using half duplex.
The system needs to know where all of the nodes are within the mesh, in order to accurately and rapidly steer the directional beams. So another enabling technology is network management software in order to control the steering. Nault noted that the commercial community already has developed network management software that enables mobile ad-hoc networks (MANETs). But with DirecNet, the requirement of directional antennas adds a level of complexity. The position of all the nodes in a network needs to be known.
CDLs’ method of communication is full duplex. But DirecNet would use half duplex, which simplifies the ad hoc mesh network problem.
“It’s a question of scheduling time,” explained Nault. “There are a certain number of time slots within the network, and each transmitter is given a certain number of those time slots in order to transmit.” This happens extremely quickly, in milliseconds, so to the end user it appears seamless.
Right now the frequencies of interest for DirecNet are Ku and X band, to augment current CDL capability.
Nault explained that the challenges of DirecNet are similar to those of laser communications (lasercom), in which a laser beam is steered to the right nodes. “But with DirecNet, the steering challenges aren’t as great, because a laser beam at hundreds of miles is measured in feet, whereas an RF beam at hundreds of miles is measured in hundreds of feet.”
DirecNet also differs from lasercom in that the latter is typically point to point. Once lasercom links are established with a communications partner, it remains there for a significant period of time. DirecNet, because it is communicating with a variety of nodes, only dwells on a particular node for a short period of time.
There could be anywhere from dozens to 100 nodes in a DirecNet local area network. And there could be a separate DirecNet network operating in the same space as the first.
Adding DirecNet capability to existing CDL terminals would entail upgrading slow-steering dishes, such as in the Fire Scout TCDL, to fast-steering apertures such as AESA, in addition to installing the network management software.
Many vehicles already have the hardware to enable DirecNet. For example, upgrading an F-18 would entail applying a data modem to its AESA in order to form the communications waveform, installing network management software, and inserting an additional two cards to take advantage of the existing AESA, according to Nault.
But Elovitz also pointed out that it still will be a challenge to make current AESAs multifunctional—capable of receiving and transmitting simultaneously.
Another goal for future systems is to bring down the cost of phased array technology needed for multiple agile beam apertures, which is a major thrust at Harris, Jewett explained.
SATELLITE NODES?
The DirecNet working group also is evaluating how to make a satellite another node in the mesh, although this would not happen initially since this issue has its own set of complications.
Regarding satellites, Nault explained, “One of the challenges that you face, when everyone has these time slots, is the speed of light. If you’re communicating with someone that’s a couple of a hundred miles away, you have to account for the fact that it’s going to take a while for your signal to get from your node to theirs. If you add satellites to the mesh, you further complicate it because now you’re not talking about a couple of hundred miles, you’re literally talking about, in some instances, thousands of miles. So the time allocation becomes fascinating.”
DirecNet would not necessarily be limited by line of sight. As long as there is line of sight to at least one other node within the network, there essentially would be communications capability with all nodes. That, incidentally, is another aspect of the network management software challenge: determining how the signal is actually routed when there no longer is line of sight.
In February 2006, Cubic conducted a simple demonstration of DirecNet, although the demonstration was point to point rather than a mesh. The demonstration, under sponsorship of Space and Naval Warfare Command (SPAWAR), consisted of Cubic’s Navy CDL system integrated with Harris’ multibeam Phased Array Antenna.
During the demonstration, Cubic communication elements were placed in two vehicles located about a mile from each other. One vehicle was stationary and the other moved around. The CDL equipment and the antennas were placed in fixed locations. The demonstration included pointing, jammer nulling, beam swapping between the two vehicles, and audience participation. The technology successfully performed dual beam pointing while maintaining links to the two vehicles.
“Right now it’s only a matter of commitment in order to build a terminal,” said Nault. ♦
