New Quantum Cryptography Speed Record

Roland Piquepaille writes "Physicists from the National Institute of Standards and Technology (NIST) have established a world's speed record for 'unbreakable' encryption with their cryptographic system based on the transmission of single photons. With this kind of method, messages cannot be intercepted without detection, meaning transmission is always safe. The NIST 'quantum key distribution' (QKD) system was used between two buildings located 730 meters apart for transmitting a stream of photons at a rate of 1 million bits per second. While it might not look very fast, its 100 times faster than with previous quantum distribution systems. This overview contains more details and references about information theory."

Comments

Always?

[mrgrey]

meaning transmission is always safe

Always is a powerful word. Nothing is totally secure.

Re:Always?

[Anonymous Coward]

Indeed, but if it were possible to eavesdrop without detection, implications for physics would be just as great as for cryptology.

Ya cannae change the laws of physics
- Scotty, Chief Engineer

Re:Always?

[Anonymous Coward]

Blah, blah, blah. Haven't we gotten tired of these trolls? In the context of the transmission itself, it is, actually, totally secure. It's obvious to anyone without an icepick in their frontal lobe that there are other potential weaknesses. However, in this important respect, QC is provably secure in a way that classical crypto cannot be.

Feel free to look into the past 2-3 weeks of /. for a more eloquent response (and reresponse and rereresponse and...).

Re:Always?

[Anonymous Coward]

The use of 'always' in this context is similar to "An apple always falls downwards when you let it go."

Re:Always?

[MS_is_the_best]

QC relies on the ability to emit photons, and to known probability distribution of those photon emissions. The problem is, there is no hardware out there than can emit one and only one photon 100% of the time. I wouldn't be suprised if it turns out to be totally impossible to build hardware that does. (Like building hardware to perfectly measure a particle's position and speed is impossible.)

This is total nonsense. Are you a cryptographer afraid to loose your job, with no physical background? Then please read the article before you respond.

I agree that the text and title posted to Slashdot is kind of misleading. All this QC does is making a channel on which eavesdropping impossible, without detection. Point. And it is.

This has actually nothing to do with crypto (you can breathe again, your salary is safe), it can be used as a nice method for key exchange in a crypto -solution. The solution in total can be hacked (do something nasty on the sending or receiving end, but the transmission cannot be listened to undetected.

Re:Always?

[BalloonMan]

"With this kind of method, messages cannot be intercepted without detection, meaning transmission is always safe."

How about if I said, "With this kind of armored vehicle, passengers cannot be intercepted without detection, meaning transport is always safe." Now, the fallacy should be a bit easier to spot.

The passengers are not really safe at all, in fact they might be D.O.A., or maybe they just got interrogated along the way, or perhaps they were replaced by pod replicants. Whatever, if they were carrying sensitive information, you cannot "save" them or the information they were carrying, i.e. you cannot prevent a compromise of the data, just by proving that the vehicle was intercepted. At best you can say, "better call off the invasion, they're probably onto us now."

Re:Always?

[Anonymous Coward]

Well, yes, you can raise that philosophical objection to the concept of "always", and I would actually agree with you.

A better statement would be: "undetected eavesdropping on this quantum channel would require finding a physical situation that does not conform to the Schroedinger Equation". Instead of saying "X cannot be broken", one can say "X is at least as strong as Y", where Y is well-known to be very strong.

Man in the Middle?

[Allen Zadr]

While Quantum physics certainly allows for scientific detection of observation (which would help you detect if someone is merely viewing your stream)

However, with all technology, this could be a common pocket-sized device some-day. So, would this not also fall under the problem of Man-in-the middle attacks? Read the quantum stream (eliminating the existance of said stream), and recreate the stream to the other point. This would create a delay, but without other forms of detection, it would not necessarily be as safe as wires... (as wires, at least, can be physically secuired. Hard to secure open air).

Re:Man in the Middle?

[Cyclopedian]

I think your premise fails because you are using an established methods that worked for certain electrical and computer principles. Quantum Cryptography (QC) is something entirely different than what's been done in the past. Current methods cannot merely just be used on QC just because it worked in the past for other levels of physics.

-Cyc

Re:Man in the Middle?

[Kainaw]

So, would this not also fall under the problem of Man-in-the middle attacks?

The way to avoid the man-in-the-middle has to do with the filters for the photons. It is confusing in the code, but easier to understand from a completely fabricated example.

First, you need to understand that photons are becoming 1 and 0 based on spin. That spin is aligned so that 1 is 90 degrees off of 0. The filters have to be aligned as well (sure makes portable devices hard, but I'm sure we'll figure that out later). Assume we cycle through 8 filters. The first four look like + so that vertical is a 1 and horizontal is a 0. The next four look like x so that diagonal one way is a 1 and the other is a 0. If you shoot a photon aligned to + through a x filter, it will become either a 1 or 0, but not necessarily the correct value.

What does that mean? It means that you and I can decide to use the following filter sequence: x++xx+x++. Now, a man in the middle must use the same sequence or he will scramble the message. If he scrambles the message, he cannot retransmit it. Also, he cannot decode it because he doesn't know which bits are correct and which ones are incorrect.

Now, what if the man in the middle knows your filter sequence? Now you hit the key-sharing problem that cryptology has had since the start. There's no point in assuming that's a new problem.

Wouldn't this make DOS easier though?

[foidulus]

This is the thing I don't understand about quantum cryptography(maybe someone can explain it to me). If someone were to try to listen in, would you still be able to read the information being sent? If not, wouldn't this make DOS attacks relatively easy? The information isn't any good if you cannot transport it.

What about keyloggers and stuff?

[joda]

Even thought that in theory, the encrypted messages (or whatever is sent) can't be read, you still have the problems before and after encryption.
Especially these days with worms and trojans affecting even the most _secure_ environments (*bad memories about some american nuclear power plant*). You can expect someone somewhere to get some spyware or keylogging-thingie onto a sender or reviever's system. (or sometimes even enough with just getting it onto the network on each end in question.)
I recall visiting a webshop somewhere who sold a small (read less than half an inch) plug, which you put in between the keyboard and the comp, which could log several megs of typed in text. Later it's just to harvest ...

Maybe I'm just paranoid, but if you can't trust your coworkers 130% in these cases, you're still toast unless you put the machine (and yourself) in a vault and throw away the key. /joda

the weakest link in the chain

[WormholeFiend]

is human.

while it's true that cryptography like this improves security, those encrypted messages are still transmitted between people, and people are not corruption-proof.

Re:Implications for the Government?

[m.koch]

*(yeah, yeah, your favourite open source encryption is unbreakable, I know, but come on, the government isn't going to enter any 'break this encryption' contests to show what a kewl ha>or it is and thereby advertise the fact that communications using said encryption are not actually secure, is it?)

Pardon? The known encryption algorithms are insecure because the government doesn't say it can't break them? Reminds me of a little story where a man claps his hands to get rid of elephants in his house. The proof that it works? There are no elephants in his house.

Also it seems strange to imply that Schneier et al are just a bunch of idiots.