Beauty of Gray

Friday, September 27, 2002

Time for a little R&R

As a note to my regular reader, I am leaving today for a vacation. A two week honeymoon in Italy, to be exact. And if you think I'm going to have the gondolier drop us off at the Internet Cafe Venizia, you're crazy. I'll return to my regular posting (or as regular as it gets) on the 15th or thereabouts. Ciao, and try not to start a war while we're gone.

Link posted by Doug Turnbull at 6:18 AM

More than one way to skin a cat

There was a lot of good information flying around yesterday in the discussion of GPS jamming.One of the more interesting links was to a flight test of the DAMASK system, as a potential upgrade to the JDAM guidance suite. Here's the link to that report (found by Donald Sensing.)

Now, the article doesn't come right out and explain how it works, but this paragraph gives a clue:

DAMASK has also benefited from work performed for the Air Weaponry Technology Program. "Ten years ago in a Category 6.2 program we were looking at template matching as a means for achieving precision accuracy," notes engineer Mike Wirtz. "The target-template concept is an example of an early investment where the payoff doesn't show up for a while, but when it does it's well worth it." In DAMASK's case, the target template can be generated from synthetic aperture radar, photo reconnaissance aircraft, or a number of other sensor/platform options.

So, DAMASK uses what they're calling a template matching. What's that mean in English? Basically, it means that instead of simply flying to a fixed coordinate, the DAMASK seeker is fed a specific target profile--in this case generated either from a photo or from radar information. Then, the seeker looks around in the target area for something matching that profile, and if it sees something, it homes in on it and blows it up.

It's pretty much like something out of a Road Runner cartoon. The Coyote gets his Acme killer robot and shows it a picture of the road runner, and then it flies off looking for the varmint. Only in this case, the Air Force shows their killer robot a picture of a radar dish, or a missile launcher, and off it goes.

With this particular system, you still need to get the bomb (and seeker) into the general target vicinity--it needs to be close enough for the DAMASK system to see the target it's looking for in order for it to work. So the GPS/INS guidance system still has a role to play, getting the bomb inside the seeker radius. But a template based system has tremendous possibilities, and similar technologies are likely to be the next big thing in armament technology (if they're not already the current big thing and we just don't know about it yet.)

Instead of having to have UAVs out looking for targets, or recon teams radioing back coordinates, you just launch out a few hundred drones with instructions to "kill enemy tanks." And you give them each a picture of a tank, and they fly off to go look for something to kill. Add a radio to the hunter-seeker drone and it becomes a simple UAV, sending back target reports whenever it sees something, even if it's not high value enough for the drone to attack.

I'd heard bits of pieces about this sort of technology a few years ago, but I thought it was still years away from realization. But if they actually had a flight test, that means it's gotten to the point where they have working prototypes, which is pretty far along the development cycle. And what's most stunning about the report is that the advertised price for DAMASK is only $12,000 apiece, which is cheap enough to buy and install them in large quantities. At that price, it's not an exotic boutique type weapon, it's an everyday drop it whenever you feel like it weapon.

So if these technologies are that far along, the whole GPS jamming discussion could go the way of the dinosaur much sooner than the 15-20 years I was conservatively estimating in my last post. (Although there will probably still be some applications where you'd like to fly to a specific location. Especially if you're going to be using it close to friendlies. Having killer drones flying around is nice 100 miles behind enemy lines. It's a little less desirable if your own tanks end up in its sights. Which brings us back to the road runner cartoon...)

Link posted by Doug Turnbull at 6:13 AM

Thursday, September 26, 2002
GPS Jamming, Part II

OK, the previous post gave a quick overview of GPS jamming and why the report of the cheap Russian jamming unit isn’t quite as bad as it sounds. But the jamming of GPS signals is still a real concern. Iraq may or may not have the capability to really deny service over a wide area, but other combatants might, and Saddam might be able to do it over a more limited area, for example in Baghdad. Denying precision GPS bomb strikes in the city would make any combat there considerably more bloody, especially for the civilians trapped in the building next to the ones we’re aiming at.

So, how much of a concern is it in Iraq, and what can the military do about it there and in general? As Katzman notes in his article, it would make things a bit more difficult, but in the first Gulf War we fought with relatively few precision guided munitions and came out OK. So we could do it again. Precision guided bombs allow you to minimize collateral damage and also allow you to take out a target with fewer bombs. Jamming the GPS signal would simply mean more collateral damage and a slower degradation of Iraq’s military capabilities, as 3 or 4 airstrikes were needed to take out a bridge rather than just one.

A couple of caveats, though. GPS guided munitions are not the only precision weapons in the US arsenal. Laser guided bombs are also very precise, and they can’t be jammed the way GPS signals can. GPS bombs are preferable because they’re much cheaper and don’t require the pilot or a ground unit to get close enough to “paint” the target with the laser illuminator, but they are still a viable option.

Second, no US bombs or missiles rely solely on GPS guidance. Instead, they use a combination of GPS and inertial guidance---GPS/INS systems, as described here. Inertial systems work by measuring the accelerations that the bomb or missile undergoes. If you know the initial position and also know all the accelerations that you’ve undergone, you can determine the current position. Old ICBM’s used inertial guidance systems to guide them to their targets.

The problem with inertial systems is that they have a drift to them. The longer they stay on, the less accurate they are. When used in combination with a GPS receiver, though, you can have the two systems communicate and have the GPS system’s coordinates fed back into the INS system to compensate for this drift. Now, if the GPS signal is suddenly jammed, the missile or bomb will then begin guiding itself solely by the INS system. It won’t be quite as accurate, but it won’t suddenly veer wildly off course, either. If the INS system is good enough, it could fly quite a while and still hit close to the target coordinates.

This is why spoofing, while orders of magnitude more difficult than simple jamming, is so much more effective if you can pull it off. A jammed missile will still hit near its target. But if you a spoof a missile, you can send it anywhere you want to have it blow up.

Like everything, though, the better the INS, the more expensive it is. So guidance systems involve a tradeoff between jammed performance and cost. Right now, GPS guidance is cheaper than INS guidance, but INS performance is improving all the time. Right now good INS systems use ring laser gyros which are difficult to make and expensive. But continuing advances in micro-mechanical (MEMS) semiconductor-based gyroscopes will eventually make them good enough and cheap enough that they will supercede GPS and the whole jamming issue will no longer be a concern. To put it another way, our vulnerability right now is a window, one which is likely to close in the next 15-20 years.(This is, of course, assuming that pattern recognition software doesn't take over for bomb guidance first.)

You can also design a receiver to be more resistant to jamming. The press report previously linked to was hyping a Lockheed Martin developed jam-resistant receiver. I'm sure there are lots of fancy ways to do this, which even if I knew I couldn’t say anything about. But one simple method is to use an antenna that looks up rather than in all directions. The GPS signal is coming from satellites up in the sky, while the jammers are, presumably, located on the ground. So if you make a directional antenna that receives signals from above more strongly than signals from below, it will be more difficult to jam.

Two final notes, concerning specific comments in Katzman’s article. First, he mentions that the civilian GPS signal is used to track and acquire the military signal. This used to be true, but new GPS receivers can acquire the military signal directly. Second, he was concerned with whether other contractors had jam-resistant antennas like Lockheed, and if not whether they could use the Lockheed design. The answer to the first question is definitely yes. They might not work as well, or they might work better, but every contractor is aware of the problem and is pursuing solutions. But if one design was head and shoulders above the others, the military could probably license that technology and use it in systems for which Lockheed Martin was not the primary contractor. The defense contractors are very competitive, but they also cooperate on a lot of projects, so this sort of subcontracting would not be that unusual.

Final note: As I've said, I'm no expert in this field, so if I got anything wrong or missed any important points, please drop me a line and let me know so I can correct it.

Link posted by Doug Turnbull at 11:42 AM

GPS Jamming, Part I

Joe Katzman (link found via Instapundit) at the Winds of Change posts this article worrying about the potential for jamming of the GPS-guided bombs the US is coming to rely on more and more, degrading our military capabilities significantly. I’m no expert, and in fact know just enough about the issue to realize my limitations and be wary of making overbold pronouncements, but I think I can add a little to the discussion.

First, the military is definitely aware of the potential problems of GPS jamming. There are plenty of smart procurement managers and defense industry engineers working on the problem. This doesn’t guarantee a solution, but whatever capability Saddam (or anyone else) has is unlikely to come as a surprise to military planners.

Katzman links to this old report of a cheap Russian jamming module, with the quoted ability to jam GPS at ranges of up to 125 miles, which sounds pretty bad. He then references this press release from Lockheed Martin about an upgraded, jamming resistant GPS module that they are selling. At the bottom of the article is the following quote:

Russia's Aviaconversia currently markets a 4-watt GPS jammer weighing about 19 pounds but capable of denying GPS reception for about 125 miles, Defense Daily reports. A 1-watt jammer can disrupt GPS signal acquisition for more than 60 miles. A 1-kilowatt jammer can completely disable military GPS receivers for about 50 miles.

An attentive reader will at this point be a bit confused. How can a 4-watt system have a range of 125 miles, but a 1 kilowatt system only has a range of 50 miles? The answer is that the first system’s numbers are for jamming civilian GPS signals, while the second is jamming military systems. The military GPS signal is designed to be more difficult to jam, and hence requires a much higher power to jam. Why is this so? A quick aside now in how jamming works (or can work.)

Any signal is transmitted at a particular wavelength, or range of wavelengths. And the signal will have a certain robustness—there is a certain signal to noise ratio above which the signal can be read and below which it can’t. (Noise in this sense is just a random background.) A conventional jammer works by just putting out a lot of noise—it raises the floor, attempting to put out so much noise power that it overwhelms the signal power you’re trying to read. Now, a power source gets weaker the farther you are from the antenna (and it goes down as one over the square of the distance), which is why a certain power has a maximum jamming range. The more power your jammer has, the longer its effective jamming range.

A second, more sophisticated form of GPS interference is spoofing. You set up a transmitter that pretends it’s a GPS satellite, and sends out bogus information that will confuse the guidance system into thinking its position is somewhere other than where it is. Regular jamming is kind of like scribbling all over a roadmap so a driver can’t read it. Spoofing would be giving him a roadmap with the wrong roads on it, so where he thinks he’s going isn’t where he really is going.

Anyway, there are a couple of different systemic ways to improve your resistance to regular jamming of the first type. One is simply to have a stronger signal. If your signal is twice as strong, there will need to be twice as much jamming noise to bring it below the lower threshold signal to noise ratio. A second way is to design a signal structure that has a lower threshold signal to noise ratio. Then, the same power signal can be read with more noise present. (I don’t know much about this—you’d have to find an electrical engineer who knows about signal processing.) And a final way is to move your signal around across a range of wavelengths, so that a jammer has to jam a larger portion of the frequency spectrum in order to effectively jam the signal, which requires more power. This can be done passively, automatically. Or it can be done actively—if the signal is suddenly jammed, you jump to another wavelength to try and avoid the jamming.

All of these techniques can raise the effective power needed to jam a signal. I don’t know if military GPS uses all of these, but some combination of them or some other technique I’m not aware of could be used to make the military GPS signal more resistant to jamming. So instead of a 20 pound, 4 watt system that can jam for 125 miles, you end up with a kilowatt system that can only jam military GPS for 50 miles. And it takes pretty good sized equipment to generate a kilowatt. For comparison, the largest radio stations in the US are 50 kilowatts, so 1 kilowatt is on the same order as a small town radio station. If you assume the power/weight ratio holds for scaleup, a 1 kilowatt system would weigh not 19 pounds, but 4750 pounds. It’s a large truck instead of a backpack, and costs a lot more than a few thousand dollars.

Instapundit asks if encryption could make the signal more jam-resistant. I don’t think so, but encryption does prevent spoofing—having a third party send out a false signal.

To be continued in part II: How does the military deal with jamming, and will it matter?

Link posted by Doug Turnbull at 10:57 AM

Maybe we should check the campaign contributors list

If you vote for Republicans in this fall's elections, the terrorists will have won.

Update: Surprise, the blogger archive link seems to be pointing to the wrong place. Just go to the front page and scroll down. It's the post from Wednesday Sep 25 at 11:39 PM

Link posted by Doug Turnbull at 10:25 AM

Monday, September 23, 2002
Great argument from TRB on deterrence

The TRB column in the New Republic had this to say about deterring Iraq:

Above all, it means clearly confronting the most serious critique of the administration's preemption doctrine: that Saddam can be deterred. The Bush administration has not adequately explained that Saddam is prone to recklessly underestimating America's resolve--which is part of the reason he wasn't deterred from invading Kuwait. And it hasn't adequately explained that while deterrence "worked" vis-à-vis the Soviet Union, there's no guarantee it would have continued to work had the USSR endured for another 50 years. (Even during the cold war, after all, there were some very close calls.) The United States relied on deterrence against the Soviet Union not because deterrence was foolproof but because we had no other choice: We could never have preemptively attacked the USSR; the costs would simply have been too great. But the United States can preemptively attack Iraq. Deterrence is no longer our only option, and it isn't our safest one.

Past performance is no guarantee of future results, as they say on the shady radio ads trying to get you to buy gold coins or whatever the investment of the day is. Deterrence worked for a while, but it was not a perfect system, so harkening back to it is not an argument ending slam dunk.

Link posted by Doug Turnbull at 2:51 PM

Sloppy Reasoning on Iraq

I happened to come across two examples today of what I think are very sloppy anti-war arguments, which I thought I’d respond to, since no-one else has. The first comes courtesy of Maxspeak (via a link from the Poorman):

We need to enlarge the aviary to include those who urge the nation to war in an obnoxious fashion, at the same time bearing relatively little personal risk for the outcome, even if they have performed military service.

I do not make the foolish claim that rural folks are pro-war and urban anti-war. Obviously there are plenty of each in both places. I do maintain the risks differ by geography, and those in relatively safer locations with a yen to support military adventures ought to be a little solicitious of those in relatively less save ones. The front lines, after all, are different now.

However, this argument could equally well go the other way. I work in DC, and that’s a reason why I support action against Iraq, and (at least in theory) support the idea of other preventive strikes. This argument assumes the real issue of dispute, which is what course of action by the US will minimize the risk of a WMD attack against us. It might not be true in all cases, but for most of those who support a war, the primary reason why is the perceived threat of Iraqi WMDs. Despite the claims that war is all about oil, it’s not.

Those who support a war do so because they believe it will minimize the chance of a WMD strike on New York or DC, not because they don’t care if such an attack occurs. And if you believe this, then suddenly those who oppose a war and don’t live “on the front lines,” as Max has it, are the ones being cavalier about the risks involved with their favored course of action.

The second flawed argument is both less interesting and more common, and it comes to us courtesy of Hesiod at Counterspin Central:

And the Administration is, perhaps unconsciously, ADMITTING that it goes against our values by the very lies and distortions it is telling about the threat Saddam poses, and its hyposcrisy with respect to North Korea and Syria, to take two examples. Obviously it cannot be about nuclear weapons and threats to our allies, and It can't be that we are liberating a people from a horrible Arab regime that supports terrorism, because that is true for those two countries, and we aren't threatening them with invasion.

There are two glaring flaws with this argument. The first become clearer if you abstract a bit. Set “nuclear weapons” as condition A, and “horrible Arab regime that supports terrorism” as condition B. Hesiod is arguing that attacking Iraq because it is A and B is inconsistent because North Korea is A and Syria is B, and we’re not getting ready to attack them. But as anyone who ever had to do Boolean algebra and those little Venn diagrams in junior high can tell you, A and B is quite different from A or B.

This is a variation on a distressingly common fallacy in historical arguments. Often, for a given historical event, there will be multiple competing theories about why it happened. Usually, the real answer is that it’s a combination of the listed factors, but a normal argument is to try and debunk each individually. The answer can’t be factor 1, since it was present at other times and nothing happened. Same with factor 2, etc. While the revisionist, iconoclast historian ignores the fact that 1, 2, and 3 had never been present before simultaneously, so maybe that had something to do with it.

Basically, it’s ignoring the idea that each listed condition might have been necessary, without any one being sufficient. More deeply, it’s a sign of simplistic thinking, believing that everything must have 1 cause, which would always produce the given result. Stated that way, it seems ludicrous, yet you’ll regularly run across this line of argument in all sorts of contexts.

Anyway, Hesiod’s argument is doubly wrong in this case, since even if you grant that a good parallel exists for Iraq which actually satisfies both the conditions he listed—which it does in Iran [update--before anyone else corrects me, I'll note that of course Iran isn't Arab, so the parallel isn't perfect. But it's close, since they are a supporter of Middle Eastern terrorism.]—that this somehow delegitimizes any action against Iraq. But this is nonsense. Because even if you think that A and B together are grounds for a preventative overthrowing of the government, there still has to be some order in tackling opponents. You can’t do everything at once, but that isn’t an argument for doing nothing.

It also ignores what other special considerations might play into the choice of action, and what other actions are being taken with regard to North Korea, Syria and Iran. Iraq is a good first step because of the severity of its threat, the fact that it’s been in violation of UN Security Council resolutions and so international support is more likely, Iran is facing domestic troubles that might solve the problem for us, and so on. Hesiod’s argument is equivalent to saying that the US was hypocritical for invading Guadalcanal during WWII because the Phillipines were also Pacific Islands under Japanese control, and we weren't invading them at the same time.

While this bit of poor reasoning ("Why aren't we invading Pakistan and China?") is common among anti-war commentators, it’s actually an argument against one of their other underlying concerns, which is that the US is a loose cannon, governed by a wild cowboy, intent on invading and conquering the world to impose our hegemony through the barrels of our M-16’s. The fact that we’re not pursuing a military strategy with regard to North Korea, Syria, Iran, and other nations that might seem to fit the profile is strong evidence, in my opinion, that their view of the US and its foreign policy is wildly inaccurate. Military action is seen as the last option, one to be taken only when other courses of action have been tried and have failed.

Link posted by Doug Turnbull at 1:53 PM

A Renaissance blog: Politics, sports, literature, history, and whatever else strikes my fancy.


Sneaking Suspicions
Crooked Timber
Unqualified Offerings
Asymmetrical Information
Cranky Professor
The Poorman
newsrack blog

Powered by Blogger