The Pentagon's tactical Internet – a war too early?
The next generation hardware is maybe still on dial-up
The Pentagon is furiously buying up commercial satellite capacity in order to meet the bandwidth needs of a new kind of IT-driven war, reports the Washington Post. But Register sources suggest that the US military has other, rather larger problems in delivering on the digital battlespace vision.
A recent Department of Defense briefing included an instructive illustration of the growth in this hunger for bandwidth, and of what it is that the military intends to do with it. Note that between the first Gulf war and Kosova the requirement grew from 256Kbps to 1.5Mbps, and that the target for 2010 and beyond is 25Gbps in order to achieve "network centric warfare", quite possibly with no soldiers at all needed on the ground. We are currently somewhere beyond the "Web tools" phase, but you'd probably be right if you reckoned that, despite the existence of pioneering units in the military, the Pentagon spinners are a couple of years ahead of themselves when they push the digitised battlespace for this war.
We covered that here. The first unit to be equipped with this technology, the 4th Infantry Division, was originally intended for deployment in Turkey, northern Iraq invasion for the use of, but as you may have heard there were problems with this. Earlier this week it was still in Texas, with its equipment on ships in the Eastern Med, so it will quite possibly miss this one.
The Abrams tank implementation of the digitized battlespace, the M1A2 SEP (System Enhancement Package) is described here, some of the salient points being "improved processors, color and high resolution flat panel displays, increased memory capacity, user friendly Soldier Machine Interface (SMI) and an open operating system that will allow for future growth." The objective of the digitized battlespace is to deliver systems whereby data is gathered by individual units, communicated back to the centre in order to produce a complete picture of the battlefield, and then this picture is sent back to the individual units.
Thus, the commanders (who could conceivably be anywhere in the world) have a clear picture of what's going on, and the troops on the ground also know what's around them, which is friend and which is foe. But how do you handle the comms?
Buying up bandwidth may help the Pentagon deal with some of the big picture, but the current scramble is over bandwidth for less ambitious but more profligate purposes. If you're going to sit on the other side of the world controlling, say, an RQ-1 Predator UAV (Unmanned Aerial Vehicle) that's sending you video, you must expect to be using a hell of a lot of bandwidth. The Pentagon plans many more UAVs, and has lots already deployed in Iraq - they chew up bandwidth, they do not produce a joined up picture. So go figure. Also, note that UAVs were the stars of the 4th's EXFOR tests a few years back - the 'Internet in a tank' didn't figure highly at that point. They're arguably two different concepts, and the one we're not currently excited about is the one that won the wargame.
Aspects of the digitized battlespace clearly need a lot less bandwidth, but this obviously mounts up when you equip a whole army with the technology, and the weak link here is currently SINCGARS. Single Channel Ground Airborne Radio System is the army's standard radio system, and within the digitized battlespace is intended to route data messages via the "Tactical Internet" which itself may be composed of multiple SINCGARS radio nets. Other technologies, eg WIN-T and JTRS (Joint Tactical Radio System) will eventually start to build a functional Tactical Internet (JTRS will begin deployment in 2005-6), but at the moment SINCGARS is what's available.
In operation, this is a pretty narrow pipe. Even taking a fairly conservative view of the comms requirements of the digitized battlespace at client level, you find you need something in the region of a 1200 baud modem connection. The actual data requirements are lower, but with the addition of COMSEC and network management (obviously necessary), that's what you come to. SINCGARS, however, is essentially old technology, a voice system that can't seriously do the tactical Internet. For data it's constrained to a single 25Khz voice channel, and other limitations force it into approximately 3.5KHz available voice bandwidth.
So it's currently possible for units to know where they are via GPS, and to report their position to commanders via their existing comms systems. They can also do combat identification, reducing friendly fire casualties, and the command centres now can at least get all of the data they need to build a pretty accurate picture of where all of the friendly units are; the big picture is possible because it has the available bandwidth, the last mile is a lot trickier. And what it is that the ground units are getting on those colour, high resolution flat panel displays today rather difficult to conceive. Are we, perhaps, a war too early? Maybe best stay in Texas this time around. ®
In our first take on the digitised battlespace, we recklessly claimed that the equipment to do this was already being deployed, and that therefore the gosh-wow stories were not entirely spin. We begin to suspect that this was not the case.