Do-It-Yourself Anti-Satellite System

This only appears at the end:

"While it may be true that, when it comes to nuts and bolts, things may not be quite as simple as they sound here, the bare fact remains -- it can be done."

Posted on July 16, 2007 at 6:47 AM • 43 Comments

Comments

Clive RobinsonJuly 16, 2007 7:44 AM

They say that,

"Accuracy and elegance are not issues in carrying out a satellite attack, the researchers say, as long as the projectile hits the satellite"

Err just how accurate do you have to be to hit something the size of a trash can / dustbin that is anything up to 512Km away?

The other line I like is,

"and textbook physics equations for ballistics to help them build a computer model"

So all we do is fire a big bullet up through earths atmosphere...

I wonder if the used a book on meteorology to work out how to deal with atmospheric turbulance.

I beleive Sadam Hussain was spending a very large amount of money on the "super gun" which was only supposed to be acurate enough to deliver a one ton payload to within about 1KM of the target point...

dhasenanJuly 16, 2007 8:19 AM

If it could be done on a terrorist budget, why is the US military spending tens of millions on it? Or do terrorists have a budget comparable to the GDP of middling nations?

DamonJuly 16, 2007 8:28 AM

The difference between Theory and Practice is that, in theory, there is no difference between Theory and Practice, but in practice, there is.

MarcoJuly 16, 2007 8:31 AM

"Err just how accurate do you have to be to hit something the size of a trash can / dustbin that is anything up to 512Km away?"

If you manage to place a shrapnel grenade or a cluster bomb within 500 meters around the target, you're probably done.

AnonymousJuly 16, 2007 8:47 AM

"If it could be done on a terrorist budget, why is the US military spending tens of millions on it? Or do terrorists have a budget comparable to the GDP of middling nations?"

The military has different tactical objectives: they want to destroy a certain object, a motivated terrorist may be less select selective: Booby trapping an orbit with fast moving debris might turn out to be much simpler that aiming at a certain satellite. Obvious Movie Plot Idea (OMPI): a) Supervillain acquires midrange missiles from a Former State of the Soviet Union military, fills warheads with rusty nuts and bolts and shoots them into the GPS satellite trajectories. b) World economy collapses. c) Profit.

MarcoJuly 16, 2007 8:47 AM

"If it could be done on a terrorist budget, why is the US military spending tens of millions on it? Or do terrorists have a budget comparable to the GDP of middling nations?"

The military has different tactical objectives: they want to destroy a certain object, a motivated terrorist may be less select selective: Booby trapping an orbit with fast moving debris might turn out to be much simpler that aiming at a certain satellite. Obvious Movie Plot Idea (OMPI): a) Supervillain acquires midrange missiles from a Former State of the Soviet Union military, fills warheads with rusty nuts and bolts and shoots them into the GPS satellite trajectories. b) World economy collapses. c) Profit.

GuessJuly 16, 2007 8:50 AM

I guess to destroy a satellite, you just let it run into a cloud of shrapnel. No need to be too precise to "place" the cloud. It's different from a hitting an airplane (which can change its direction in a second) or a ground target (where a direct hit is necessary)..

YosiJuly 16, 2007 9:00 AM

In one word: load of crap. Even discussing such BS is insult to one's intelligence.

And no, it can not "be done". To hit satellite in the orbit, you probably should first to be able to deliver _some_ object to an orbit, which itself is extremely complicated task (both from technical and logistic standpoint).

DBHJuly 16, 2007 9:06 AM

"How accurate"...

Remember, this is in three dimensions, so you would need to detonate along the axis of travel at the right time. For something at ballistic speeds, this is non-trivial. In reality, you would need to detonate before you arrived so the shrapnel would continue along the trajectory and hit the target, but not so far ahead that the shrapnel had dispersed too much to have a high likelyhood of impact.

There isn't a way to 'hang' something in space, there are only different orbits which perhaps could intersect. So the shrapnel wouldn't just hang out like a cloud, but would be moving in an orbit, which would probably eventually take it back into the atmosphere for burnup, but could cause other impacts before that...

HellfireJuly 16, 2007 10:09 AM

Rather than physically destroying a satellite, what if you just pointed a powerful directional antenna and blocked out communication with Earth. That would prevent it from being any use, and it would be difficult to determine who was doing it because you were using a directional antenna.

AndrewJuly 16, 2007 10:25 AM

The future of warfare is in space. People who live at the top of a cliff can throw rocks at people below, very cheaply. Shooting back at them with high-powered rifles is possible, but expensive and requires accuracy and skill.

Building ASAT weapons is not trivial -- nor is it all that difficult, compared to building nuclear weapons. If you can loft a satellite, you have almost all that you need to loft ASATs.

Prior comments about jamming or lasing a satellite are accurate -- with enough power. Think 5 to 10 mW, or the output of good sized power plants. Ever notice that the Chinese and the Russians focused on satellite replacement for their reconsats? We chose a different strategy: Keyhole and it's ilk are powered and can not only shift orbit to optimize position, but "dodge" ASAT attack.

It is much, much easier to physically attack a satellite ground station with standoff weapons than it is to launch an ASAT. Due to orbital mechanics, you can't put all of these within your major nation's defensive perimeter.

A minor power that is stupid (or desperate) enough to attack a major power is going to be sorely tempted to go after its outlying ground stations with infiltrators, long-range diesel-electric submarines, or even a Pearl Harbor style conventional air attack.

Sufficient desperation might cause them to attempt to defeat military security inside the target's home nation. Unfortunately, this is far too easy as most US military bases have a major highway along one side of the perimeter. The rest is left as an exercise for the reader.

Chris SJuly 16, 2007 10:58 AM

Attacks on retrograde geosync were discussed back in the mid-eighties.

http://www.jamesoberg.com/pearl.html

The technique discussed here involves a lunar flyby - which is apparently more fuel efficient than a direct retrograde launch. It also makes the point that although one military may want or need surgical capabilities, another military may get by with a fairly broad brush.

Nobby NutsJuly 16, 2007 11:02 AM

@Andrew: The output of my laser pointer is a few mW, so I guess I'm going to have to try to knock out a few sats. I guess you really meant MW ;-)

Timm MurrayJuly 16, 2007 11:08 AM

@Andrew: "Think 5 to 10 mW, or the output of good sized power plants."

I think you mean 5-10 MW. $10 laser pointers at gas stations can put out 5-10 mW.

One thing to keep in mind in this discussion is that there are plenty of satellites in LEO to hit, which is a far easier goal than geosynchronous orbit. LEO is generally accepted to be 200-2000km.

For comparison, the V-2 rocket was reaching 55km in the 1940s, and it came directly out of the research done by the Verein für Raumschiffahrt, a German amateur rocket group. Current amateur groups have gone past that, and get closer to LEO every year.

NostromoJuly 16, 2007 11:38 AM

This is not BS at all. Rutan's company has shown it's possible to reach 100km without massive expenditure, and that was with a man-rated vehicle, with capability to glide to a safe landing. Instead of a man, a similar vehicle could have carried a second-stage missile. At 100 km it's already above the atmosphere so a missile fired from that altitude can easily reach LEO.
Hitting a satellite? Difficult, but doable. The orbits of target satellites are accurately known. With appropriate sensors, the missile can know where it is. With appropriate control software, it can guide itself to meet the target head-on. The targets are moving at orbital velocities; all it has to do is get accurately in the way of one. It can use radar to home in. The control software is non-trivial but it doesn't take big bucks, it just takes bright people.

Valdis KletnieksJuly 16, 2007 12:48 PM

@wowsers: "how much physics they've studied"?

Probably not much - ISTR that when one of their safe houses got raided, the "nuclear weapons plans" that were found turned out to be articles from the Onion or the Journal of Irreproducible Results, or some such...

cookseyJuly 16, 2007 1:52 PM

I dont know how much you would have to hit. The ballistics behind that kind of a precision shot would never happen on a shoestring.

On the other hand simply lofting and distributing a number of pounds of BB's into a busy orbital path would be quite damaging as space junk is definitely dangerous.

Even if they didn't hit anything initially they would cause a mad scramble and would pollute that orbit for years.

AnonymousJuly 16, 2007 1:56 PM

Seems a lot of the dEtractors have never read up on "Brilliant Pebbles".

Also, you could create a series of large debris clouds (even if the debris exapnds outwards, it will settle into an orbit at some point -- look at the massive anounts of debris that NASA and NORAD track already). Most communications sats are in geo-synch orbits, so that's and easy path to "seed".

John DaviesJuly 16, 2007 2:00 PM

@Nostromo

One can only admire your optimism. You don't work in marketing by any chance!

AnonymousJuly 16, 2007 2:04 PM

@Valdis Kletnieks and @Wowsers
Why do you assume these individuals kow less about these topics than you do? You seem to forget that they come from the very countries that gave us the foundations for much of the physics and astronomy, etc., that is used today. These people aren't stupid just because US media portrays them that way -- but your assumptions that they are gives them al sorts of advanatages.

AnonymousJuly 16, 2007 2:16 PM

Personally, I think it would be easier to just hack a satellite connected to the internet. I'm 110% sure you can control a satellite without spending a dime.

SparkyJuly 16, 2007 2:50 PM

You can't just create a "cloud" of debris in a geo-synch orbit, because the velocity of the debris would dictate it's orbit, which means that every object in a given orbit has to be traveling at the same speed, if it is to remain in that orbit for any period of time.

I suppose you can have objects in orbit that move in the opposite direction of the normal geo-synch orbit, which means this debris and a satellite would cross paths every 12 hours.

That way, one could build a missile with small thrusters for maneuvering that would attempt to intercept a satellite, and detonate a charge if it was close enough. If not, you'll get another chance in 12 hours, provided you have enough fuel to maneuver.

AnonymousJuly 16, 2007 2:54 PM

"Military satellites, global positioning systems, weather satellites and even satellite TV systems could all become victims of such an attack."

Am I the only one that thinks this is a rather strange sentence? Military and even civilian satellites can be targeted! One would think military targets would be harder to hit, or at least deemed more important than TV satellites?

wkwillisJuly 16, 2007 4:49 PM

It depends on whether you are going into orbit or just going up 300 miles. The difference is 11KPS vs 3KPS for low Earth oribit with the difficulty roughly square of the velocity.
There is a simpler and cheaper (MUCH simpler and cheaper) solution, but why should I give them ideas...

AnonymousJuly 16, 2007 5:38 PM

> This is not BS at all. Rutan's company has shown it's possible to reach 100km
> without massive expenditure, ... Instead of a man, a similar vehicle could have
> carried a second-stage missile. At 100 km it's already above the atmosphere so a
> missile fired from that altitude can easily reach LEO.

That is pure BS. It's easy and relatively cheap to reach a height of e.g. 100km or 200km by shooting a pretty small rocket straight up. That will cost you less than a million bucks. Inserting something into an orbit of 200km is harder by several magnitudes. The former can be done with a cheap post WW2 technology sounding rocket. The latter requires hundreds of millions, if not billions of development cost. What Rutan did is impressive, but far from what e.g. the Russians did when they launched Sputnik.

To reach even a low orbit you need speed. LOTS of it. That means huge, heavy multi stage rockets. Even the biggest ICMBs can only deliver their payload ballistically, and many of them might not reach even a low orbit when launched without their payload.

Filias CupioJuly 16, 2007 5:58 PM

Following up on the latest Anonymous's point:
Energy required to put 1kg to 100km altitude = mgh = 10^6 Joule (approx)
Energy required to give 1kg sufficient velocity to orbit (about 8 km/s) = 1/2 m v^2 = 32 x 10^6 Joule (approx).
I.e. to a first approximation, it is 30 times harder to orbit something than to just lob it to 100km.

ShadJuly 16, 2007 6:54 PM

A good and cheap shrapnel source may be a bucket of coarse sand and/or fine gravel. At the orbital speeds, a cloud of such particles anywhere in the path of the satellite can be more efficient than a direct hit to the noggin with a sawed off shotgun. The delivery of the payload to the satellite trajectory has to be timed so the cloud does not disperse too much at the rendezvous time, but that is orders of magnitude less precise than the requirements for a direct kill hit.

Maximum altitude achieved by a ballistic missile at the mid-course phase is being quoted to be about 1200 km, Low Earth Orbit altitudes are between 200-2000 km. For a successful collision there is no requirement that the kill vehicle payload stays in orbit for prolonged amount of time; only being there at the same time as the target counts. Accelerating to orbital speed is therefore not necessary. Forming a curtain of micrometeorites that will stay in the path of the satellite for several orbits is a good strategy here.

The accuracy vs payload mass tradeoff depends on the available technology; the party with better electronics will choose more accurate rendezvous setup and smaller amount of shrapnel, while the party with worse computers and bigger rockets will opt for pure brute force and big bags of gravel.

Things that go up there do not have to be complicated. An example of a successful cheap approach is the use of oak wood for ablative thermal shields of some Chinese reentry vehicles.

Timm MurrayJuly 16, 2007 8:12 PM

> That is pure BS. It's easy and relatively
> cheap to reach a height of e.g. 100km or
> 200km by shooting a pretty small rocket
> straight up.

And for targeting another satellite, that's all you'd need to do.

AnonymousJuly 17, 2007 2:43 AM

> A good and cheap shrapnel source may be a bucket of coarse sand and/or fine gravel.

Yeah. A bucket of fine gravel dispersed in in tens of cubic miles of empty space really has a huge chance of successfully killing a satellite. Especially when the sand and the satellite have the same orbit, meaning they roughly travel at the same speed and in roughly the same direction.

> And for targeting another satellite,
that's all you'd need to do.

Totally dude. Hitting a tiny object travelling at 28000km/h with a rocket that has a velocity of roughly zero at its apogee. The rocket has to be there EXACTLY on time. It must be VERY close to the targeted satellite. And then it must point and detonate its (directional!) warhead at exactly the right moment to create a high enough spatial distribution of shrapnel of sufficient mass to kill the satellite. Pretty simple, no doubt

I wonder why the Yanks spend dozens of billions of dollars to develop their missile shield since according to T.M. it is so simple to hit a satellite/warhead/whatever by simply shooting a rocket in its path...

another bruceJuly 17, 2007 3:20 AM

i'm no expert at this, but wouldn't it be easier to hit a satellite with a laser or a directional electromagnetic pulse than it would be to hit it with a rocket?

gregJuly 17, 2007 5:51 AM

The concerns are the large cost asymmetries. Its much cheaper to deny space to your opponent that it is to get access to space.

Suborbital antisat would not be hard for well funded groups. Were not talking a bunch of disgruntled teenagers. Were talking small government level. Orbits are easy to predict very accurately (or give your MSc in physics back). Remeber the difference between suborbital and orbit is about 7000m/s. There really is no comparason.

On the other hand. If you can shot something up to 300km. You can also shot horizontally something >300km (~600km for simple ballistics). In the middle east this would pose a much greater problem for example.

Even more interesting is blinding. Using a laser its possible to blind or even damage instruments on spy satellites. Most serious military would have thought of this however...I would think.

AnonymousJuly 17, 2007 1:00 PM

"You can't just create a "cloud" of debris in a geo-synch orbit, because the velocity of the debris would dictate it's orbit, which means that every object in a given orbit has to be traveling at the same speed, if it is to remain in that orbit for any period of time."

You're confusing yourself.
For an object to stay in geosync orbit, it has to be a certain distance, which depends on the mass of the object. Any other object in the same orbit, also in geosync, has to be nearly the same mass.

Debris particles, having much less mass, can travel much slower than a comparatively huge satellite. Of course, the particles of debris will begin to drift into different orbits as soon as they're released, but as long as you're relatively close to the correct speed, they should drift in the same orbit until the satellite hits it.

BennyJuly 17, 2007 3:06 PM

Anonymous at July 17, 2007 01:00 PM,

I believe Sparky was right, and you are wrong. It is not true that an object's orbit depends on its mass, or that lighter objects can travel slower than heavier objects in the same orbit. Please see equation 3.6 on this page:

http://www.braeunig.us/space/orbmech.htm

This equation demonstrates that given desired orbital distance r, gravitational constant G, and Earth's mass M, one can solve for required velocity v. The object's own mass plays no part. As Sparky states, objects in the same orbit must have the same velocity, their mass is irrelevant.

X the UnknownJuly 17, 2007 6:32 PM

In the same vein, it should be trivial to "hit" the largest object in our Solar System - the Sun. After all, it's huge, and gravity is all working your way! Let's just dump all of our radioactive waste into the Sun!

Except, it ain't necessarily so...

Turns out, with the pinnacle of modern technology, the only way we even have a prayer of dumping something into the Sun is to first send it on a multi-year, computer-controlled (and even then, with continual flight-path & program adjustments) trip out to Jupiter, where a very intricate orbital slingshot maneuver could then be used to (maybe) get enough velocity-change to actually "drop into the Sun".

Note, also that the "mathematical precision" of a ballistic orbit is fallacious, too. That ONLY holds true in the imaginary case where there are only two gravitationally-interacting objects in the whole universe. Once you have three or more objects, you get the infamous "three body problem" - which has been proven to be mathematically unsolvable. So, what we have in reality is a mathematical prediction based on the known-to-be-false assumption that only the Earth's gravity is interacting with the satellite. In practice, this tends to be "close enough" that we can probably find the actual location by looking around near where the prediction says, then resetting our initial conditions to what we actually measured before running more predictions. Heck, even the (in this case, extremely relevant) rotation-speed of the Earth is known not to be constant or totally predictable!

-----------------------------------

Different orbital = different energy-levels. Moving from one to the other (e.g. from Earth-Surface to Stable Orbital) requires expending a suitable (generally very large) amount of energy.

If you do it with a computer-control rocket, with billions of dollars worth of avionics and control-systems, you can make adjustments as you travel, giving you some reasonable chance of actually arriving near your target - with a payload that is some fraction of a percent of the mass of your incredibly-complex delivery-vehicle.

If you do it with a purely ballistic approach, your delivery-vehicle can be almost all payload - many more ball-bearings and explosives with which to hassle the target. Unfortunately, the accuracy of your "launch" probably has to be beyond the mechanical capabilities of anything currently manufacturable. Even disregarding the (mentioned by others) effects of atmospheric turbulence (which are enough to affect the focusing of light, by the way), the microscopic imperfections in your gun-barrel and the tiny imbalances and unpredictabilities of the burn-rate of your propellant are going to introduce enough deviation that you'd be lucky to get within a kilometer, IMHO.

Of course, if you use a high-tech propellant, instead of a chemical one, such as an electromagnetic launcher, you can get around a lot of the physical-imperfection problems. A nice, computerized, high-powered, sequenced-pulse, multi-megawatt electromagnetic launcher might just do it. If you could get a high-enough cycle-rate, you could just spew out a virtually-unlimited number of shots, and "blanket" the target-area. Sort of "Aegis" for satellites. But wait - isn't that a rail-gun? Hasn't the U.S. military been spending untold amounts of secret-budget money trying to develop this for the last few decades, and the best they can do is a warehouse-sized contraption that can fire a single shot, then has to have its core components completely replaced, because they weld together as part of getting the flux-levels required? Hmm...maybe not a home-basement job, after all.

ZytheranJuly 17, 2007 8:34 PM

Dropping junk into sun..
Thinking about this I cant see the technical problem?
Step 1: Go into orbit around the earth with the plane of orbit at right angles to the the earth-sun vector. 23 degrees off a polar orbit.
Step 2: Accelerate back along earth's orbit until the radial velocity around the sun is zero. (11.2 km/s can be used to escape earth's orbit anyway and another 18.6 km/s to reduce orbital velocity around sun to zero)
Step 3: Fall into sun and time it to miss Venus and Mercury?
Step 4: Write movie plot and profit.

wkwillisJuly 17, 2007 9:09 PM

You are a satellite moving at about 7000 meters per second in a 300 Km low Earth orbit. There is a 1 gram grain of sand in your path, sufficient to jar loose connections in your electronics package and kill you.
The 1 gram grain of sand is part of the payload of a ten tonne hypergolic pressure fed rocket, basically a large reinforced plastic chemical storage tank with a nozzle on one end and two valves you open to launch.
Assume the rocket is ballistic with a 1% error rate (about 3Km circle) for longitude, lattitude, and altitude, or about six square kilometers area. It's area, not volume, because you are traversing the volume.
One tonne is 1,000,000 grams, so there is one grain of sand per 6 square meters.
If I were them, I would budget for 100 rockets and guarantee a kill.

AlexJuly 18, 2007 3:53 AM

Well, yes, wkwillis. Dunno about "terrorists", but there are no shortage of *states* with old Scuds or similar. More of those were fired in Afghanistan's civil war than in the Gulf War.

gregJuly 18, 2007 5:58 AM

@X the Unknown

Its not that hard to predict LEO orbits. you need to use at least a 2nd order model of the earth (aka its oblate) and know some friction data which can be derived from observation's. Also take into account the thermoshear weather. you probably don't even need to worry about the moon, and even if you did its no problem. Its not that hard and it is accurate. Dam look at the photos of the space shuttle made by a guy with mead telescope. 100% predictive tracking!

As for guidance system. They where hard 10-20 years ago. but now a few thousand dollars would be more than enough. You can even get them for RC planes etc... Basic Gov level funding would be more that enough... The Rocket would still be the hard bit.

X the UnknownJuly 18, 2007 1:09 PM

"Step 2: Accelerate back along earth's orbit until the radial velocity around the sun is zero. (11.2 km/s can be used to escape earth's orbit anyway and another 18.6 km/s to reduce orbital velocity around sun to zero)"

That's the rub. We currently don't have any rocket-technology capable of achieving this kind of delta-V, at least from an Earth-lauched vehicle.

I guess you could build one in orbit (like the space-station), and after several years of tricky, incredibly-expensive construction and fuel-caching, you'd get one shot at it - hope the totally-untested thing, built in a completely unique environment, works the first time.

ShadJuly 18, 2007 2:36 PM

X the Unknown: The problem here is to dissipate the kinetic energy without using too much fuel, so conventional rockets are out. An ion engine may be helpful here, however. Another possibility is to employ e.g. a solar sail in a way that would provide the required braking.

Leave a comment

Allowed HTML: <a href="URL"> • <em> <cite> <i> • <strong> <b> • <sub> <sup> • <ul> <ol> <li> • <blockquote> <pre>

Photo of Bruce Schneier by Per Ervland.

Schneier on Security is a personal website. Opinions expressed are not necessarily those of IBM Resilient.