Was Quantum Entanglement Experiment Behind “Classified Cryptographic Equipment” Confusion After Antares Crash?

Yesterday evening, an Antares rocket built and operated by Orbital Sciences Corporation exploded shortly after liftoff. The rocket was intended to ferry supplies and equipment to the International Space Station. Orbital and Spacex have taken over some of the duties supplying the space station since the termination of NASA’s shuttle program.

In the early aftermath of the explosion, word came out that the crash site had to be secured because sensitive cryptographic equipment was on board:

The Cygnus mission was non-military, but the company’s Antares program manager, Mike Pinkston, said the craft included “some classified cryptographic equipment, so we do need to maintain the area around the debris in a secure manner”.

That initially struck me as odd. The International Space Station has a large number of cooperating countries, including Russia. It’s hard to imagine that the US would put sensitive equipment into the hands of cosmonauts right now, given the cool state of US-Russian relations. Of course, it would make sense for ISS communications to be encrypted in order to prevent meddling by hackers, but movement all the way to classified (and presumably military or NSA-level) encryption seems to be excessive.

This morning, we are seeing walk-back on the presence of classified equipment:

Shortly after the explosion, CNN quoted a launch director as saying that the spacecraft contained classified “crypto” equipment, but early Wednesday a NASA spokesman said by email that “We didn’t have any classified items on board.”

In trying to make sense of what could have been behind these strange statements, I ran across this interesting announcement of a new cryptographic technology that European scientists have proposed evaluating in an experiment on the space staion:

A team of European researchers have proposed a series of experiments that, if successful, could turn the International Space Station into a key relay for a quantum communications network.

The key basis of physics underlying quantum communications is entanglement. Entangled particles are connected in a way that pretty much defies common sense. If you change the spin of one of the particles, the spin of its entangled counterpart will change – even if they’re miles apart. And that change happens nearly instantaneously – at least four orders of magnitude faster than the speed of light, according to a recent experiment.

Another remarkable aspect of this technology that sounds almost too good to be true is its potential security. After noting that quantum networks are quite fragile, the Forbes article continues:

But why bother with these networks at all if they’re so fragile? The answer is pretty simple – because they’re almost perfectly secure. Here’s how it works. Let’s say that I want to send a message to New York City. My message is going to travel through normal channels, but it will be encrypted with a key. That key is transmitted via the entangled photons – so the changes I make to entangled particles on my end almost instantly show up in the particles in New York. We then compare the measurements of what I changed in my photons to those states in New York City.

Those measurements then comprise an encryption key for our communications. So even if our communications are bugged, nobody can read them without knowing that encryption key. And here’s the important thing: if somebody were to try to eavesdrop on the quantum entanglement, they would alter the spin of the photons. So the measurements I make and the measurements made in New York would be out of sync – thus letting us know that we have an eavesdropper. It also prevents us from creating an encryption key, so we don’t send any communications. Theoretically, a quantum encrypted network is almost perfectly secure. (That said, they’re not perfect, and there are some exploits.)

The announcement from the European group that they wished to carry out the experiment based on what Einstein called “spooky action over a distance” came last April. Then, in June, it was announced that China had carried out a key demonstration of concept experiment back in 2010 but waited four years to publish the result.

With China announcing progress on the technology, one would think that the West would want to accelerate its work in the area, so it would not be at all surprising if equipment for the European experiment was among the items lost when the rocket exploded. Further, one would expect that Orbital would have been told that security for that equipment would be of the very highest level. In discussing the issue of sensitive equipment among the Antares wreckage, PCWorld this morning mentioned the incident of China perhaps examining the wreckage of the US stealth helicopter that was left behind after the mission to kill Osama bin Laden. It could well be that for this crash site, keeping the debris away from prying eyes from China is behind the call for security. Note also that the experiment quite likely would have been coordinated by the European Space Agency on behalf of the European scientists, so NASA’s claim that “We didn’t have any classified items on board” could be parsed as not applying to any classified items that ESA might have had on the rocket.

20 replies
  1. What Constitution? says:

    Can’t digest this. I’m seriously stuck on “four orders of magnitude faster than the speed of light”. Meaning “four orders of magnitude faster than a speed which everybody agrees cannot be exceeded”? But it does sound like there could be interesting applications. And I hope we figure it out before the Romulans do.

    • TarheelDem says:

      As I understand it, the phenomenon is a consequence of the math in a theoretical view. This experiment was an (another?) attempt to see whether the math was demonstrable in reality. It is only if it is that the technology based on it can have any reality after long years of applied science development work. If sabotage instead of the usual engineering failure (remember O-rings in cold temperatures), that would be going after some very preliminary information. It does violate one of Einstein’s theoretical assumptions about action at a distance unless it can be explained within a framework in which that assumption is still valid.

    • Rayne says:

      I’m not certain “orders of magnitude” is appropriate for comparative purposes. “Spooky action at a distance,” Einstein’s snarky description of quantum entanglement, theoretically suggests that pairs of atoms synced in the same vibration or spin, will continue to share the same movement no matter the distance between the atoms once they are separated. This can’t be compared to the speed of light, which has a rate. In other words, quantum entanglement is beyond speed — it’s a state of existence. This atom A[1] vibrates at X speed locally, while its twin A[2] on Mars or on planet Alpha Centauri Bb, vibrates there at the same speed. Shift A[1] and A[2] does exactly the same thing, at the same time.

      What I find even more fascinating and potentially useful in communications is the possibility quantum entanglement’s effects may remain even if the pair once in sync is de-synced. There might be shadowy quantum fingerprints left behind if QE-encrypted communications were broken. This might be used to assure the continuity of a message, not merely ensure detection of broken encryption.

  2. TarheelDem says:

    And then, since the vendors are private, “cryptographic” might be a euphemism for “proprietary”. Wouldn’t want competitors knowing how you engineered your proprietary equipment, would you. As this privatized space program takes off, we are likely to see even more efforts as secrecy that will feed back into NASA.

    It is both nations and corporations jockeying to exploit the “final frontier”. And all of them hyped up on three-quarters of a century of science fiction.

  3. Rayne says:

    * laughing * Wow, the size of the truck fleet one could drive through this statement.

    “We didn’t have any classified items on board.”

    “We” — NASA made the statement. Not the NSA, not USAF, though. And not GCHQ or any other FiveEyes intelligence agency.

    “classified items” — again, by which entity’s classification? Or is it so very deep that it’s covert and disclosing its existence isn’t possible? Or is it something so ubiquitous, like the backdoors on all NIST-standard electronics used in the US that it’s right there but not discussed?

    “on board” — as mission equipment? or as cargo? What did this mean?

    The public knows something wasn’t quite right because of the delayed response — not the 5-20 seconds of mute, gasping cognitive dissonance we all shared watching the rocket launch and explode, but the time following the failure before formal statements were made. The backtracking about the classified or crypto materials is itself a clue something wasn’t completely transparent about this launch.

  4. seedeevee says:

    Space Systems/Loral (Bernard Scwartz – big Clinton donor) had a fling with Chinese launchers in the 1990’s … When a launch went bad the Loral engineering team spent a lot of time searching the nearby destroyed Chinese village looking for proprietary pieces.

    Of course they did not find them all.

  5. edge says:

    It prevents eavesdropping, but it can also be used to prevent communication. If I constantly eavesdrop on all your quantum lines, you’ll never be able to communicate and will have to all back to traditional communication systems.

  6. Adam Colligan says:

    One thing that needs to be made clear here is that quantum entanglement experiments do *not* violate the speed-of-light barrier in any practical sense. Relativity dictrates that no *information* — no causal influence — can travel across space faster than the speed of light.
    The nature of quantum cryptography always requires that a traditional signal be sent and received through a speed-limited channel in order to interpret the meaning of whatever has been locally measured. So you can’t know how, or even if, your distant partner has manipulated / measured the quantum state of her entangled particles until you compare notes the old fashioned way.
    Entanglement provides very interesting benefits for message security but not for message speed.

    • Rayne says:

      I’ll disagree with this premise, provided our ~current~ expectations of message are binary — a simple yes/no, on/off, changed/unchanged, 1 or 0, a single bit of information.

      It’s extremely important to understand a single bit is all it takes to set a transaction in motion, like a trade thousands of miles away upon announcement of a policy decision. It may not be entirely secure, but if no one is aware of the communications system or possesses the means to disturb it, it’s viable until opponents catch up.

      Or until the system’s disrupted by unfortunate, untimely explosions or malware intrusions…

      • Adam Colligan says:

        I know it’s counter-intuitive, and it’s understandable when anyone new to the concept has this misconception. But despite the entanglement being expressed in some ways “instantaneously”, not even a single bit of actual data can be transmittted faster than light. Causality remains in tact. There are a lot of good lay resources on the topic at varying levels of detail; it might be worth starting with Wikipedia at No-communication theorum.

        • Rayne says:

          Thanks for the effort, but I’ve sunk quite a bit of research into quantum entanglement over the last 15 years in relation to consulting I’ve done on the future of work and supporting technology.

          I’ll point to one of the works cited by the Wikipedia link you provided. The authors Peacock and Hepburn said,

          “…The notion of nonlocal energy is, admittedly, difficult to grasp. One might be inclined to think that according to a causal interpretation, there must be some sort of superluminal transmission of a localized pulse of mass-energy between the remote particles. However, if we ask whether energy is being shuttled superluminally between A and B, by tachyons perhaps, we miss the point. Some such description might be useful in some contexts. However, the real point is that mass-energy is nonlocal, a global property of a multi-particle system. The multiparticle system as a whole will have a spectrum of possible energy states, and the energy is not any place in particular at all; it is just a general property of the system, that may make itself manifest in a variety of ways. It is probably safe to say that this is analogous to the way in which the energy of an electron orbital in an atom is a global property of the orbital as a whole. Localization of mass-energy is a process that happens in certain specific circumstances that we do not fully understand as yet.

          Our remarks here can only serve to indicate some enticing possibilities. The important point to note is that there are numerous indications within quantum physics that the dynamics of multiparticle systems are, in general, nonlocal. It seems to be largely philosophical prejudice against nonlocality that has prevented us, so far, from following upon these leads—a philosophical prejudice against which Bohm and Hiley argue persuasively([1]), but to which they appear to have fallen victim themselves. …

          [boldface mine; source, Begging the Signalling Question: Quantum Signalling and the Dynamics of Multiparticle Systems, c. 1999]

          Hardly sounds like a strong refutation of nonlocality and quantum entanglement’s apparent violation of causality.

          An enormous amount of work remains investigating properties of multiparticle systems, including the nature of singlet states and other as-yet unidentified forces acting on multiparticle systems (example). “[t]here is new physics to be uncovered,” as Peacock and Hepburn suggest, and other researchers support (see Peres & Terno, c. 2003, Open problems, p. 27, PDF) — which is the point of this blog post: NASA may have had research on board the Antares to this end.

          An easy explanation for NASA’s odd initial reply and subsequent backtracking: NASA may not have been classified material being conveyed since test equipment might well be found in unclassified lab environments. It would, however, be the study itself (example), or the order/department directing the study, which would be classified, and it may not be on board but in an earth-bound environ.

          Given the status of the Chinese’ work on quantum entanglement in tandem with PRC’s rampant cyberwarfare, and the status of our own quantum crypto work, I’d be damned surprised if someone weren’t pushing hard for (and/or already conducting) classified quantum entanglement studies in space including A/B switch-type communications.

          Or perhaps even using technology based on research not yet declassified…

        • Adam Colligan says:

          There are really four distinct concepts that I think are being conflated here. (1) Nonlocality per se, (2) speculation about ‘nonlocal’ communication, (3) superluminal communication, and (4) actual quantum cryptography experiments.
          (1) Experiments with quantum entanglement (as well as with plenty of other related phemomena, like delayed choice) are at odds with our instinctive grasp of causality and locality. I was certainly never meaning to imply that they make “sense” in the traditional manner. They’re spooky, they’re at a distance, and you can argue about the ‘action’ part.
          (2) Relativity generally limits motion through space rather than the stretching of space, and we’re familiar with the wormhole-as-shortcut idea. But there are also problems and paradoxes casting very serious doubt on the idea that even wormholes (or other tricks like Alcubierre drive) could ever really be used to violate causality.
          Some interpretations of quantum entanglement could be called dimensional: the idea that the particles are not as ‘separated’ as they appear to us. A version of that sentiment is what you have quoted. But that is not the same thing as seriously opening the door to causal influence between two points in spacetime that are outside each others’ light cones. Similarly, time itself remains a mysterious beast, sand there are lively speculations, both credible and fringe, about what should and should not be ‘allowed’.
          (3) However, all that stuff generally falls into the category of “we could possibly be wrong about some basic, well-founded principles in physics”. That is different than being in the category of “our basic, well-founded principles in physics appear to credibly allow for this shortcut”. The fact remains that there is no real evidence either of tachyons or of any ways to leverage quantum ‘spooky-ness’ for faster-than-light communication. The experiments that put bounds on how fast a tachyon would have to have been going (e.g., at least four times the speed of light) are all retrospective constructions. As far as I’m aware, there is not even a tabletop setup or credible design for one that would actually allow the extraction and use of any information without the aid of speed-limited signalling. You can play in the lab with lots of crazy things that are “faster than light”, like the phase velocity of photons, but the idea of transmitting information with them is, in almost all frameworks and all real experiments, completely barred.
          (4) That brings us to quantum cryptography experiments that are actually being done. Even if you subscribe to the view that some form of superluminal communication is actually possible, quantum cryptography is not actually leveraging this in any way. Cryptographic entanglement setups rely on well-undertstood processes of signalling that do not violate causality or attempt any shortcuts around the no-communication theorum. Even if you think the entanglement itself might contain some secrets of practical nonlocal influence, the manner in which it is being employed does not.
          Say I had a magic baseball that some people credibly theorize could communicate with the ghost of George Washington. I do an experiment where I throw the baseball from a pitchers’ mound to home plate: my hypothesis is that I can throw a strike. It’s silly to write about my experiment as if it were an one that in any meaningful way involves talking to George Washington.

  7. P J Evans says:

    They’re thinking that what happened – and I’ll believe it, having seen the video – is that one of the Russian rocket engines they were using blew up. Said engines are from the 60s, stored who-knows-how-carefully, and have been problematic for several years.

  8. Evangelista says:

    OK, I have a hot tip for Empty Wheel (EW). I want to send from SoCal to Wis without MIM (Man in Middle) in the Mountains of Utah (MOU). So I encrypt and send crypt key via Quantum Tangle Technology (QT Tech). EW acknowledges receipt of a key. I send the hot tip. EW replies that key did not decrypt. So I know that MIM in MOU intercepted the key. What about the message? Did MIM in MOU decrypt using the single-tangled key she intercepted and then resent, causing it (the key) to double-tangle and so not work?
    And then, these quantum things being fragile, what if it was a lightning storm in Utah, not MIM in MOU, freeing up a stray ion, who, having nothing to do, hooks to my key, notices it is a mess and so straightens it out, so we all (except the MIM in MOU, who is living at light-speed and so doesn’t know anything about it) only think the key got intercepted?
    note that all of this works in quantum mechanics. It’s only in reality it all, or some, might not…

    • Adam Colligan says:

      There are clever experiments that have been performed already regarding attempts to hide man-in-the-middle attacks against quantum cryptography systems. (These inevitably target implementation flaws — like you said, the weaknesses are in the practical setups rather than the basic concepts).
      The MIM should not have been able to decrypt your initial key message as long as the system was at all robust. Even outside the quantum realm, modern key exchange is performed in such a way that should prevent access for a MIM even if the MIM can see all the data transmitted during the exchange, all the way back to the start. The Wikipedia entry for Diffe-Hellman key exchange is quite lay-accessible in explaining it, with minimal math. Way oversimplified: there are calculations performed on each end that create secrets for each user that have a public part and a private part. You can take any arbitrary data and encrypt it to the other person’s public key, and it can only be read by someone who has their private key (even though you never see their private key). And you can take any arbitrary data and sign it with your private key, and anyone who reviews your public key can be certain that it must have been sent by you — again, even though they can never see your private key.
      The main weakness exploited by MIMs today is that the key exchange depends on both of the parties knowing which public key to use for each other in the first place. The public keys are esoteric: they’re big chunks of random-looking data that turn over all the time, and on the end of an ordinary web user, they get generated constantly for new connections. So if you want to send a message to your bank that only your bank can read, you need to, as it were, look up their key in the phone book. The real problem is: how do you know that a thief didn’t take the phone book off of your porch and replace it with a copy that listed wrong address/key information for banks in general or just your bank? There is a public key infrastructure that tries to deal with this problem. Essentially, in common practice, you have to trust the security at certain private authorities that hold onto, and broadcast, lists that verify which keys really are associated with which companies. And you have to trust your browser or other software vendors to listen to good authorities and ignore bad ones. And so you also have to trust that you’ve not downloaded a doctored copy of you browser, etc., ad infinitum.
      Some MIM attacks are actually against flaws in the implementation of the key exchange itself. But in general, MIM only works by compromising a certificate authority or list of certificate authorities in order to trick you into doing a key exchange with the wrong key for the person you’re trying to talk to. Provided you’re doing the key exchange using the real key for your partner, it doesn’t matter who is listening. This won’t change with the advent of quantum cryptography. The main point of quantum cryptography is to be secure against mathematical attacks on the keys or the transmission of chunks of the keys.
      End detour
      To the fragility question: yes, lots of things can cause the coherent quantum system to decohere or however you want to describe it. A lot of the engineering work being done now is about how to try to make it more robust — how you can be less dependent on individual particles’ states and/or how you can shield them better. If you’re getting a high (or total) loss rate, it could always be some logistical problem rather than a real adversary. But the point is that if you are getting through, you can be confident you’re secure. If you’re not getting through, it doesn’t tell you how confident you should feel that you’re not secure. The system is designed to work only when it’s known to be secure (so it fails safe, where ‘safe’ is no meaningful transmission).
      Yes, that means that you could simply disrupt all the traffic on the line and force the users to not talk to each other or to revert to some other form of communication. But the special “fragility” of quantum systems isn’t necessarily a game changer in this respect. In fact, many regimes already selectively block SSL or other forms of traffic for this very reason. That’s why Tor now has ‘pluggable transport’ layers that try to hide the encrypted traffic as plain-text traffic. But in most situations, an adversary doesn’t really have the practical option of shutting down a whole line just because they’re frustrated they can’t eavesdrop on a specific conversation, instead deciding just to make sure the conversation can’t happen.

  9. Cujo359 says:

    If there were any bulk encryptors on board, they may have decided to make sure they weren’t keyed before allowing anyone in the area. That doesn’t mean that they were using NSA-level encryption,just that they wanted to keep those communications channels secure from any people who might have shown up in the area.

    • Adam Colligan says:

      It would seem unlikely to me that even an NSA-sponsored experimental box on the ISS would bother using any special, operational cryptographic algorithms, much less operational keys. This technology is in its infancy, and it would seem idiotic to put any real sensitive information through an untested device on a joint-Russian platform. Also, more than likely, the normal operational NSA algorithms are just the public ones with slight modifications kept secreat as extra insurance or extra intimidation. I’m not at all inclined to believe that there would be any need for the NSA to say, “but would the box work with *our* setup rather than normal RSA or what have you?”
      Now, there certainly is interesting research being done in terms of how you actually strructure the box or write computert instructions for it. This would be especially true if the box were trying to do any actual quantum computation, where there are all these prtoblems with efficiency and error checking and whatnot. So it’s plausible that there could be something classified about it.
      But the main purpose of sticking a box like this on the ISS is not because it’s some great operational relay (after all, the DoD has its own long-endurance shuttle now anyway). The point is to do experiments with transmitting entangled particles over distrances of hundreds of miles though potential intterference like you find at various layers of the atmosphere at different times of day, different times of year, etc. You could use any arbitrary cryptographic system to “key” these experiments and get the test data you came for, so there’s no real reason to use a sensitive, classified one.
      None of this means that the NSA (or whoever) would be happy to let anyone look at any of their equipment. After all, they don’t even want you to look at their legal authorities. But that doesn’t mean thst if you did look at the box, it would help you break any codes — or even be that interesting. It just means intelligence and defense research agencies are secretive, like the sky is blue.

  10. What Constitution? says:

    Thank you to all who have contributed to the effort to understand what this particle thingie is all about. A monument to Things Which Cannot Be Explained in 144 Characters Or Less. I knew Wikipedia served a higher purpose.

    Still not clear on the concept, though I do find “nonlocality” to be a fascinating concept which, it appears, may come closer to describing whatever this is than did the phrase used in the article Jim White had to start with in the post, which had elected to say “four orders of magnitude faster than the speed of light”. I get it, that’s not really what is meant.

    So I think I’ve got it: what this really is, is “magic”. And, since that’s another way of understanding Mr. Einstein’s initial moniker of “spooky action”, I think I’ll go with that as my layman’s grasp of what’s being postulated.

    To put it in context of movie quotes, it’s apparently covered in the first “The Santa Clause” movie, when Tim Allen objects that his ex-wife’s new husband has told the child there’s no Santa Claus and this exchange ensues: “Laura: All Neil told him was that Santa was more of a feeling. More of a state of mind than an actual person. Scott: Kind of like Neil.” That’s what this phenomenon is: it’s kind of like Neil — it seems to be there, but nobody’s really sure. And it definitely doesn’t exceed the speed of light, it just sort of happens. Got it. Magic. I’m going to have to stick with Newtonian physics for a while longer.

Comments are closed.