Beta
×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

Quantum Key Exchange With an Airplane

samzenpus posted about 2 years ago | from the spooky-flight dept.

Encryption 44

submeta writes "Researchers in Munich have successfully performed a quantum key exchange between a moving aircraft and a ground station. Quantum key distribution, which exploits the phenomenon of entanglement, offers theoretically perfect encryption (although it can be vulnerable in practice). This advance is an important step on the way to key exchange with a satellite, which could enable practical usage of the technology."

cancel ×

44 comments

Sorry! There are no comments related to the filter you selected.

This post is relevant to my interests (2, Insightful)

AikonMGB (1013995) | about 2 years ago | (#41355989)

Quantum key exchange via satellite: QEYSSat [qeyssat.com] .

A little unclear on entanglement (2, Interesting)

msobkow (48369) | about 2 years ago | (#41356091)

Doesn't entanglement also imply faster-than-light communications between the two quantum nodes? Or does the speed of light still apply to the entangled systems?

If the former, this would eliminate the lag for satellite communications, which would be a major breakthrough in global communications, even if the satellites could only link up to a few ground stations because of the limited number of entangled "transmitters" per satellite.

Re:A little unclear on entanglement (5, Insightful)

Anonymous Coward | about 2 years ago | (#41356191)

It does imply faster than light propagation of states, but unfortunately it is not possible to use this effect to communicate information from one place to another. Observing the pair at either end breaks the entanglement immediately (faster than light), but there's no way to tell from one end if the entanglement has been broken, so you could imagine that the universe is talking to itself at faster than light speed, but you can't hear what it's saying.

Re:A little unclear on entanglement (0)

Anonymous Coward | about 2 years ago | (#41356841)

That's deep as bro.

Re:A little unclear on entanglement (2)

Brad1138 (590148) | about 2 years ago | (#41357003)

I love science and (when I have time) have kept a skeptical eye on quantum entanglement. I haven't studied it a lot, but have problems with a number of its claims. You are saying it works and is faster than light, as long as no one observes the outcome... Really? It sounds less legitimate every time I hear about it.

Re:A little unclear on entanglement (5, Informative)

Anonymous Coward | about 2 years ago | (#41358127)

The mechanism works regardless of if someone observes it or not. The faster than light aspect is just not observable by a person at one end, it can only be noticed when the two ends talk to each other. In other words, the faster than light aspects can not be used in communication, just shows up in patterns that are observed if you compare results from both ends that must be communicated by traditional means.

I haven't studied it a lot, but have problems with a number of its claims.

Considering how difficult the field of quantum mechanics is, and how many people fail to grasp basic principles even with extensive studying of pop-sci level material, maybe a little more humility would be called for until you have a chance to read more about it.

Re:A little unclear on entanglement (1)

Anonymous Coward | about 2 years ago | (#41360845)

If my understanding of the Dunning-Krueger effect is correct, then humility is not one of the characteristics generally shown by the ignorant.

Re:A little unclear on entanglement (0)

Anonymous Coward | about 2 years ago | (#41361911)

I love science and (when I have time) have kept a skeptical eye on quantum entanglement. I haven't studied it a lot, but have problems with a number of its claims. You are saying it works and is faster than light, as long as no one observes the outcome... Really? It sounds less legitimate every time I hear about it.

It is not about someone "observing" it is the side effect of the process of observation.

In order to observe something requires you to change it in some way in order to detect its existance.

For example you can't see me with your eyes unless photons bounce off me and stick to you.
If they bounce I was (ouch) modified in the process.

If photons are emitted from something directly and absorbed by you they are being modified as they are absorbed by your eyes so you can see. In all cases this process of detection requires modification of some shape or form.

It is this modification which is involved with wavefunction collapse.

The term faster than light implies propogation which aint happenin bro.. just like the word "observation" these words and assumptions they convey exist only to confuse you making a simple concept seem more mystical than it really is.

Re:A little unclear on entanglement (1)

neonKow (1239288) | about 2 years ago | (#41362107)

You are simply misunderstanding it. It's not that it works so long as no one is watching. It's that there's no way to harness it to have faster-than-light communication.

You can read entangled bits and they will have the same value on each end. HOWEVER, interacting with the bits in any way, including reading them, will break the entanglement, so you cannot flip a bit on one end and have it flip faster-than-light on the other end.

What you CAN do with this weird property is to use it for secret key exchange securely. You can generate a bunch of random bits and both ends will have the same bits on each end.

This article is about doing that key exchange from a moving platform with a stationary platform, so it's about really accurate laser targeting.

And I agree with other commenter: a little humility on your part would be in order. This is hard science at work, not voodoo.

Re:A little unclear on entanglement (1)

neonKow (1239288) | about 2 years ago | (#41362337)

I made a mistake. This article might not have anything to do with quantum entanglement. Everything else applies about how entanglement is supposed to work and how it's not "faster than light."

Re:A little unclear on entanglement (2)

DriedClexler (814907) | about 2 years ago | (#41359093)

I always have to wonder about interpretations of quantum mechanics that are able to believe those ideas at the same time:

"Yes, the information propagates faster than light, but in just the right way that we could never use it for communication" --> isn't that the same thing as "The information doesn't propagate faster than light"?

Re:A little unclear on entanglement (0)

Anonymous Coward | about 2 years ago | (#41359299)

"Yes, the information propagates faster than light, but in just the right way that we could never use it for communication" --> isn't that the same thing as "The information doesn't propagate faster than light"?

The idea of information being "propogated" is incorrect. Nothing is being propogated hence distance and speed are meaningless.

The system is in fact indeterminate until its wavefunction collapses. This is to say the answer is not actually realizable until you ask for it.

Re:A little unclear on entanglement (3, Informative)

Skillet5151 (972916) | about 2 years ago | (#41356207)

No entanglement phenomenon has ever been shown (or really believed) to be capable of transmitting information faster than light. I promise you'll see it right there in the Slashdot headline if a decent experiment ever seems to show any kind of FTL information transfer.

Re:A little unclear on entanglement (5, Informative)

kasperd (592156) | about 2 years ago | (#41356243)

Doesn't entanglement also imply faster-than-light communications between the two quantum nodes?

No. It's true that you can have entangled particles far apart and measure the state of both of them at the same time. That does mean you instantaneously know what the other end would read. But that does not imply communication, since you don't have any control over what value comes out. And you can't even find out if the other end was measured or not. In such a situation it doesn't really matter which of the two particles is measured first. Measure particle one first, then the other collapses to a state consistent with the measurement of the first. Measure particle two first, then the first collapses to a state consistent with the measurement of the second. Which of the two happened depends on the viewpoint of the observer.

And actually quantum key exchange does not need entangled particles at all. There are certain optimizations, that could make use of entangled particles. But in plain quantum key exchange, you send a stream of independent particles where the sender knows the state of each particle he sends. The receiver doesn't know the state of the particles as they are received, but may learn something about the state, depending on how they are measured.

I have read of some implementations, that produce a particle in a known state by first producing an entangled pair of particles, and then measuring one of them. By the time the other particle leaves the sender it is no longer part of any entanglement. And the fact that it ever was part of an entanglement is just a minor implementation detail, that doesn't actually impact the protocol.

Re:A little unclear on entanglement (1)

someone1234 (830754) | about 2 years ago | (#41359549)

But, this is the same as having a sheet of paper with the same word written on it.

Re:A little unclear on entanglement (1)

emt377 (610337) | about 2 years ago | (#41367315)

But, this is the same as having a sheet of paper with the same word written on it.

That's what key exchange is all about.

Re:A little unclear on entanglement (2)

sFurbo (1361249) | about 2 years ago | (#41359551)

I think there exists two kinds of quantum cryptography. The one you describe is the one implemented now. It doesn't require entangled particles. IIRC, the other goes something like this:

1. Alice creates two sets of entangled particle, and sends one to Bob.
2. Alice encodes her message into a from that can interact with the entangled particle.
3. Alice lets her message interact with the entangled particles, creating the coded message. It is now no longer readable.
4. Alice sends the coded message to Bob. It is not readable, so it can be done over an open channel.
5. Bob lets the coded message interact with his set of particles, retrieving the original message.

AFAICT, entanglement creates a one time pad that is guaranteed not to be known by anyone else. The method implemented now creates random, shared information.

I am not sure whether the coded message created in step 3. is entangled with Bobs particles, or whether step 3. breaks the entanglement.

Re:A little unclear on entanglement (2)

kasperd (592156) | about 2 years ago | (#41359891)

I am not sure whether the coded message created in step 3. is entangled with Bobs particles, or whether step 3. breaks the entanglement.

It depends on how you do it. It could come out either way. However the only point in using the second method you describe is, that you can achieve entanglement at that point. I just don't know of any application of that, since usually the data you eventually want to transfer securely, are classical bits.

If you just want to protect classical bits, then the more complicated and more fragile method you describe doesn't offer any benefits. An ordinary quantum key exchange will in the end produce the same key at both ends, and that key will consist of classical bits, that can easily be stored by the two parties, until they need to communicate.

Re:A little unclear on entanglement (1)

sFurbo (1361249) | about 2 years ago | (#41360523)

I see. While writing my reply, I got the feeling that it had to be something like what you wrote, but I didn't really get a firm grasp of it until your reply. Thank you for that :-)

I think I have the method from some very popular scientific descriptions at least a decade ago, so they probably didn't go into too much detail about entangled data.

Re:A little unclear on entanglement (1)

emt377 (610337) | about 2 years ago | (#41367299)

And you can't even find out if the other end was measured or not. In such a situation it doesn't really matter which of the two particles is measured first.

Are you sure about this? Once collapsed the particle will no longer interfere with itself if you pass it through e.g. a double slit. And I always understood that entanglement meant the two particles shared a common wave function, so would collapse together - basically when you measure one the other will take one a single state as well. Otherwise you'd be claiming the entangled particle is not yet collapsed but you'd know what it will collapse to, which is an impossibility (if you know what it will collapse to, then it's collapsed.)

Re:A little unclear on entanglement (1)

kasperd (592156) | about 2 years ago | (#41367983)

Are you sure about this?

Yes.

Once collapsed the particle will no longer interfere with itself if you pass it through e.g. a double slit.

That phenomena is not dependent on entanglement. The self interference requires only a single particle. It doesn't matter if it is or ever was entangled with another particle. AFAIK the self interference requires the particle to be in a superposition of passing through both slits at the same time.

Entanglement is more complicated than just a particle being in a superposition. In fact entanglement is also more complicated than having two particles, which are both in a superposition. If you have two particles each of which are in a superposition, then they can collapse independently to any combination of single states.

Entanglement means the probability distribution of the collapsed states is one that cannot be expressed as just the product of the distributions of each of the particles. The two frequently given examples are either that they are guaranteed to collapse to identical states or that they are guaranteed to collapse to opposite states. But there are other ways they could be entangled.

If we just take the guaranteed to collapse to identical states example, then particle one may have 50% probability of collapsing to state A and 50% for state B. Similarly particle two has 50/50% probability. If you simply multiply, you get 25% for each of the four outcomes. That would be true if they were independently in superpositions. But since the actual probability distribution across those four outcomes is not 25% for each, then they are not entangled. The actual probability distribution would be 50% for A,A 0% for A,B 0% for B,A 50% for B,B.

Entanglement is more than just a probability distribution. But the details about such entanglement beyond the probability distributions is more complicated than what I am able to explain, and some of it, I don't even remember.

Otherwise you'd be claiming the entangled particle is not yet collapsed but you'd know what it will collapse to, which is an impossibility

It is correct that it is impossible to know what the state will collapse to, before it collapses. And that's not what I was saying. The tricky part is how the quantummechanics interacts with relativity.

As either end measures their particle from the pair, the wave function will collapse. But if both do it "at the same time" then it is not well defined, who measured it first. Who was first depends on the inertial system of the observer. One observer will see A measure first causing a collapse and B measuring an already collapsed particle. Another observer will see B measuring first causing a collapse and A measuring an already collapsed particle. But all observers will see A and B performing consistent measurements.

So either party can measure their particle and know that after they have measured, the collapse has happened. But if they don't measure, there is no way of knowing, if the collapse has happened.

If there was a way to find out if the collapse had happened, it would break causality. Instead of both measuring "at the same time" they could do something else "at the same time". A could either measure or not measure, and B could try to find out if A already did measure. But from one observers viewpoint, B would do that before A measured, so B would definitely find that the particle had not yet been measured. But from another observers viewpoint, A would measure before B tried to find out, if the particle had been measured.

The only reason that doesn't violate causality is because you cannot find out if the particle was measured. A having measured knows the collapse has happened, and what B would get if B actually did measure. But A doesn't know if B measured. B has a particle and having not measured doesn't know what the outcome would be. By the time A could communicate the outcome to B, the collapse would definitely have happened.

Re:A little unclear on entanglement (1)

Aphoxema (1088507) | about 2 years ago | (#41356245)

The answer of if it's faster than light is "maybe", but it's reasonably certain that information can't be exchanged faster than light. All entanglement can be used for is to verify something that is already known and you only get on chance to test the condition. If you get it wrong, you waste the entanglement by adding new information to it. It's essentially a one-dial etch-a-sketch.

Re:A little unclear on entanglement (1)

tehcyder (746570) | about 2 years ago | (#41363353)

It's essentially a one-dial etch-a-sketch.

Now THAT's what I call an analogy.

Re:A little unclear on entanglement (2)

gweihir (88907) | about 2 years ago | (#41356405)

No. It could basically be simulated by both particles agreeing beforehand on a state and then reveal that simultaneously. Cannot be used for FTL communication.

There are some quantum-physical effects that indicate that this is not what happens and the two particles really sort-of synchronize in an FTL way in real-time, but my impression of the reasoning was that it requires a bit more assumptions than I am comfortable with, i.e. I think it may turn out to be wrong.

Re:A little unclear on entanglement (1)

_8553454222834292266 (2576047) | about 2 years ago | (#41358137)

Wouldn't your first paragraph violate bell's theorem?

Re:A little unclear on entanglement (1)

gweihir (88907) | about 2 years ago | (#41360345)

Possibly. I am merely talking about the use for key-exchange.

The thing is that quantum key-exchange is completely irrelevant and useless, as both particles need to be at the same point at some time in the past to get entangled. Instead you could just generate a key at that time and transport it to both locations. Security should be about as difficult, but transport will be far easier, as there is no need to keep particles in an entangled state. For the key-exchange scenario, it is also completely irrelevant when the key was generated, it just needs to be secret and known on both sides.

That is why this "usage scenario" is completely bogus with regard to practical applications. It has merit as a scientific demonstration only.

Side note: In modern cryptography, you would actually generate two secret/public key pairs. Even more secure, and that quantum stuff cannot do anything like it AFAIK.

Re:A little unclear on entanglement (0)

Anonymous Coward | about 2 years ago | (#41364539)

With quantum entanglement you can securely transmit a one-time-pad (secure because you'll know if there is a man in the middle intercepting it), and you absolutely do not get more secure than using a one-time-pad. The advantage of transporting a key with quantum entanglement over a traditional method is the you know if there is a man in the middle, with traditional key transmission you can't necessarily be sure. I'm not saying that it is a practical method to use, but theoretically it can't be beat.

Re:A little unclear on entanglement (1)

WaffleMonster (969671) | about 2 years ago | (#41365191)

With quantum entanglement you can securely transmit a one-time-pad (secure because you'll know if there is a man in the middle intercepting it), and you absolutely do not get more secure than using a one-time-pad.

Does entanglement include some kind of cosmic identity filter to validate who you are taking to? ...so really it is capable of detecting no such thing.

All this does is allow people who already have a preexisting trust relationship via classic means to accumulate OTP.

The problem with lack of effective binding between classic and quantum channel leaves me wondering what the practically useful aspects of the exercise is.

If something is that sensitive you can't use crypto then would not prearranging a few TB of OTP be more secure than worrying about binding between classic and quantum channels?

A single SD card filled with random noise is enough for years of secure talk time.

If you could afford to be slightly less paranoid one could choose to use cryptographic means to stretch out your OTP pool as necessary based on the sensitivity of data. Would that really be any less secure than risks associated with binding classic and quantum channels with assumption quantum channel actually works 100% as advertised?

Re:A little unclear on entanglement (1)

gweihir (88907) | about 2 years ago | (#41370511)

You miss that you have to transport the entangled particles. That is about as insecure as transporting the pad in the first place. Maybe you are referring to some other form of "quantum crypto" that uses polarization, but not entanglement. In that system, single photons are sent. It also has been broken several times by now.

Re:A little unclear on entanglement (4, Insightful)

camperdave (969942) | about 2 years ago | (#41356539)

No. It's kind of like flipping a coin. If you know one side is heads, you automatically know that the other side is tails no matter how far apart the sides of the coin are. Quantum Entangling photons is kind of like splitting the coin in half.

Re:A little unclear on entanglement (1)

Kartu (1490911) | about 2 years ago | (#41356775)

Nope: http://en.wikipedia.org/wiki/No-communication_theorem [wikipedia.org]

From this it is argued that, statistically, Bob cannot tell the difference between what Alice did and a random measurement (or whether she did anything at all).

I can't state I fully understand it though.

Re:A little unclear on entanglement (2)

ikaruga (2725453) | about 2 years ago | (#41357879)

I'm more curious to know how entanglement is even relevant to this study. I'm just an engineer, not a physicist, but from my understand of the article, they basically used an almost standard laser communication. However, in this case with two slightly different beams(different polarization) for transferring the encryption key. The reason they called this "quantum" was simply because they were reading the key signals by directly measuring the photon state due to polarization. That is important because, if someone tried to intercept the signal they'd necessarily use some sort of lens or mirror that would change the signal polarization. This change is measurable and if detected, the keys must be changed. AFAIK, no entanglement phenomena is used in this study. The "New scientist" article mentions entanglement when mentioning that 143km "teleportation" study (sorry, but that hardly looked like any teleportation at all, more like quantum state transfer using laser beams/optical communications). And by mentioning a complete independent Nature study, they can get more clicks. Sorry if there are any mistakes in my post.

Re:A little unclear on entanglement (1)

Shavano (2541114) | about 2 years ago | (#41357887)

The speed of light applies because the information travels with the entangled particles.

Re:A little unclear on entanglement (0)

Anonymous Coward | about 2 years ago | (#41357891)

Doesn't entanglement also imply faster-than-light communications between the two quantum nodes? Or does the speed of light still apply to the entangled systems?

Think of this quantum bullshit as a transactional memory system. Within the domain of your transaction you can fiddle with all kinds of crap the outside world can't see. It is only when the transaction is committed anyone else can understand the outcome. The semantics are a little different but what can one expect.. all analogies suck.

The same tired old concept is used by god to maintain coherency in the verse keeping any further "neo" outbreaks contained...no more FTL, boxes with more room inside than they consume outside or sheets of dark matter cookies baked by Larry Ellison himself.

I love those who blabber about perfect security while overlooking the fact you still need "classical" keying material to establish trust relationship between parties. Steal any of the "classic" keys and quantum MITM becomes just as trivial to pull off as the non-quantum version.

Re:A little unclear on entanglement (1)

azav (469988) | about 2 years ago | (#41362427)

Yes, if you consider distance to be a factor. No, if you consider this two entangled systems to really be collapsed into one system but in >1 location.

I wonder if we will be able to have more than a pair of items entangled at one time?

Assuming we are only using binary systems, envision this:
Entangle several items, say 7, and add one more and then attach these to a reader the tabulates the results from the first 7 when the result of the fifth goes from 0 to 1. After tabulation, this can be passed on to another item for display or processing. The fifth bit is then set back to 0, which a transmitting system reads as a license to send 7 more bits of information.

Assuming we are only using 7 bits. This allows the transmission of low ASCII characters (vals 32 - 126) encoded in binary.

Assuming the two ends of the entangled devices are stable, fly one to Mars and send it an ASCII character.

Viola. You've got faster than light communication. This would enable faster than light communication over distances that we can transfer each "end" of these devices to.

Now, if the connection stays entangled after reading the value, that's the catch.

You're all welcome to be part of my patent. Well, since it's disclosed, I guess it can't be patented now.

Of course, please feel free to tell me just how much crack I'm smoking.

That BS again (-1, Redundant)

gweihir (88907) | about 2 years ago | (#41356123)

No, it does not work, this is merely a stunt.
No, it is not needed at all.
No, it does not improve security. Because it requires perfect functioning of machinery, it decreases security.

No, I do not want any more stories about this nonsense.

I will believe it... (3, Funny)

kiriath (2670145) | about 2 years ago | (#41356169)

When Sheldon is excited about it.

Re:I will believe it... (1)

Anonymous Coward | about 2 years ago | (#41356983)

This is a joke, right? Nobody here actually watches this show, right? If i'm wrong, man I need to stop reading all your guy's comments. Wow.

Re:I will believe it... (0)

Anonymous Coward | about 2 years ago | (#41361291)

This is a joke, right? Nobody here actually watches this show, right? If i'm wrong, man I need to stop reading all your guy's comments. Wow.

Um, the op never said -what- show he was talking about...

BB84 protocol does not use entanglement (4, Informative)

Anonymous Coward | about 2 years ago | (#41357057)

This experiment implemented the BB84 protocol with attenuated coherent laser pulses (with a view to use decoy-states to close the photon number splitting attack on attenuated coherent states). The ideal implementation would use single photons (which are highly non-classical states). The protocol does not utilize entanglement. The E91 protocol was the first QKD protocol to propose using entanglement for this purpose.

moD down (-1)

Anonymous Coward | about 2 years ago | (#41357787)

Give BSD cred1t

Solution without a problem (2)

Mathinker (909784) | about 2 years ago | (#41359123)

QKD is a solution to a non-problem. Even if the current public key algorithms would be broken (via a revolutionary advance in the the field of quantum computing, or in some other, yet unknown way), there exist backup algorithms for which there are no known quantum algorithms which break them.

Re:Solution without a problem (1)

fatphil (181876) | about 2 years ago | (#41359793)

It's also in some ways a non-solution to a non-problem, as it has the issue of being unable to distinguish evesdropping from noise. And as noise always happens, you can never be sure some of it isn't actually evesdropping.

Re:Solution without a problem (1)

Anonymous Coward | about 2 years ago | (#41363467)

The security assumption is that all noise is assumed to be due to eavesdropping. The noise level (e.g. quantum bit error rate) is used to determine the level of privacy amplification required to factor out any possible third party correlations with the final key. The tolerable error rate depends on the exact QKD protocol, e.g. for BB84 with perfect single photon sources it is at least 11% depending on the coding used in the privacy amplification step. Hence QKD in the presence of noise is feasible.

Check for New Comments
Slashdot Login

Need an Account?

Forgot your password?

Submission Text Formatting Tips

We support a small subset of HTML, namely these tags:

  • b
  • i
  • p
  • br
  • a
  • ol
  • ul
  • li
  • dl
  • dt
  • dd
  • em
  • strong
  • tt
  • blockquote
  • div
  • quote
  • ecode

"ecode" can be used for code snippets, for example:

<ecode>    while(1) { do_something(); } </ecode>