My understanding is that the physics of imaging satellites is pretty precise, even a few ounces bias on one side can make the whole thing useless. Getting equipment of any sort installed on a very expensive satellite without anybody knowing would be pretty much impossible. I can't even imagine the complexity of designing such a device...it's not like it's a device that can be fired from a gun 300 meters away that sticks to the satellite and some 007 type happens to make the shot while the satellite is in transit to the payload fairing.
It would have to fit perfectly within the power and payload envelope of the satellite and literally not be noticed by any of the engineers or designers working or designing what's essentially a 1-off piece of space hardware. And having it in relative isolation during assembly for months or even years.
It'd be like somebody sticking an extra hard drive or NIC in the computer you built at home and you not noticing it.
Anybody care to speculate what this device was within the bounds of physics or is this just paranoid politics over some specialized American built hardware that France needed to source for the comm system?
Finding backdoor technology two months after signing a contract that neither the UAE experts nor the French engineers had been aware of also seems unlikely, according to both experts.
"The most likely explanation is that a competitor has planted a seed of doubt in an attempt to sabotage the deal," said Charret.
The client (UAE) discovered it, the builder (France) put the tech into the satellite. I don't think anyone is really thinking the USA put these systems in behind their backs. France was in on this the whole time and they were hoping UAE's engineers would not be completely thorough in analysing their designs.
These are older satellites and it sounds like the cameras could move slightly independently so that they may be able to compensate for the entire satellite being off a little bit.
France could be using US components which contain a backdoor and never notice it. Seems perfect plausible.
The other theory that a competitor planted a spy device and nobody noticed is even more troublesome. So my satellite supplier has shaky security procedures and I might end up getting compromised products? Not good.
While I was reading the article one question that kept coming back was who is this unnamed source. It looks much more like a power play than anything else... but given the recent NSA disaster, anything seems possible.
It'd be like somebody sticking an extra hard drive or NIC in the computer you built at home and you not noticing it.
No, it'd be like you buying a NIC and installing it without realizing there was a backdoor in its firmware. Not everything in a satellite is a one-off custom piece, a lot of it is COTS and integrated by the builders.
There's a very strong bias here on HN that everything is software. But Satellite assembly is very much a HW thing as well. Attributing whatever it is to a specific country lends itself to being hardware more than software, probably some parts that have very few or one supplier so the country of origin can be determined.
Why do you assume it's hardware. I imagine the hardware to send the signal is already there, they just need to add some software to proxy it to https://goodies.nsa.gov
If they are building the ground station tech, then there could be nothing at all on the satellite.
If I were a French general and a colleague from the US came and said, ok, we help you sell this dish to the arabs but we add this little thing in here so a bell ring in my office if they plan to lightgun a nearby country.
I'd say yes. I makes everyone happy: French have the money, US has their bell and Arabs have their dish.
>My understanding is that the physics of imaging satellites is pretty precise, even a few ounces bias on one side can make the whole thing useless.
I am having a hard time thinking about what you might mean by this. I have no doubt that the imaging platform itself must be machined and assembled to high tolerance, but this is a matter of dimensions rather than mass. What kind of problems could be created by adding a little bit of mass? It may change the rigid body dynamics but that should be easily corrected in software. Care to elaborate?
My university physics prof was a retired satellite engineer for NASA. I don't really understand the specifics, but he took a class period one semester to go over the difficulties of building devices that work correctly in zero-G, because sometimes effects that are ignored on Earth have major impacts in space. Among those things is a requirement to balance the satellite to within very small tolerances using some pretty precise measuring devices. I don't remember the exact precision, but it was some fraction of a gram.
He gave an example of a program he worked on (an environmental imaging satellite) where the circuit created by the electrical bus was geometrically configured (accidentally) in such a way that a slight magnetic field was generated in part of the spacecraft. This part of the spacecraft would interfere with the Earth's magnetic field and produce some impossibly small force on that part of the spacecraft, sending it into a very slow, unintended, roll. The force produced was impossibly small, he used the analogy of a down feather from a baby parakeet resting on your hand. The result would have been that the spacecraft would have used all the mechanical energy in its flywheels trying to stay stable, and then used all of its maneuvering fuel and become useless in just a few days. It was caught during some sort of balance test and they had to reengineer the circuit to produce a magnetic field in the opposite direction to cancel out the effect.
He gave another example where an instrument package was delivered at the wrong weight due to a change in some insulator. I can't remember if it was too heavy or too light. They had to add small weights to a different part of the spacecraft to maintain balance. IIR he said it took them several weeks to get the weights in the right positions to pass the mass balance tests on various axes.
I have no idea how accurate this is, but I thought it was an interesting and not-intuitive insight into how things in space work differently than on Earth.
He had other interesting stories about thermal management issues and other similar.
You forget that each subcontractor writes the specs for its modules. So Subcontractor X specs theirs as being grossly overweight and overpower. The system integrator bitches them out but budgets for overages. After a reasonable amount of time has elapsed, Subcontractir X reveals a miracle: their wonder engineering team has got the module nearly down to the target budget. Attaboys are dispensed to the worker bees, champagne to the management, and the spy module gets installed.
It would have to fit perfectly within the power and payload envelope of the satellite and literally not be noticed by any of the engineers or designers working or designing what's essentially a 1-off piece of space hardware. And having it in relative isolation during assembly for months or even years.
It'd be like somebody sticking an extra hard drive or NIC in the computer you built at home and you not noticing it.
Anybody care to speculate what this device was within the bounds of physics or is this just paranoid politics over some specialized American built hardware that France needed to source for the comm system?
The stated story smells.
http://www.france24.com/en/20140107-competitor-sabotage-behi...
has a more reasonable analysis
Finding backdoor technology two months after signing a contract that neither the UAE experts nor the French engineers had been aware of also seems unlikely, according to both experts.
"The most likely explanation is that a competitor has planted a seed of doubt in an attempt to sabotage the deal," said Charret.