Slashdot Mirror
Single-Photon LED: Key To Uncrackable Encryption?
nut writes: "The BBC are carrying this story of new type of LED so precise that it can emit just one photon of light each time it is switched on. It has been developed by scientists from Toshiba Research Limited and the University of Cambridge. It is described in the journal Science, although I can find no mention of it on their website. One of the applications of this is supposedly uncrackable encryption, due to the law of indeterminacy. This application is described fully in 'The Code Book', by Simon Singh, although the method was only theoretical at the time the book was first published."
The article is unfortunately a little light on details. The application of these devices seems to be for sharing key material for an OTP. Seems that it could be considerably more practical than the quantum entanglement of particles methods previously discussed.
The line can't be tapped, because if you intercept the photons, you can't re-create the signal. Read an article on Quatum Cryptography.
-Mark
I've been following this technology with great interest. There seems to be a fundamental problem: it is point to point. Its applications will be fairly limited.
It seems to me, at least in terms of networks, that this would really be used to secure lines between networks, clusters, or individual computers. But on today's public Internet, this isn't really an issue. Of course, I would rather use this technology than to not have lines protected with quantum indeterminism.
Most security people are more concerned about platform security than link security. If this technology can be used to reinforce something used for platform security, then boo yeah! Otherwise, this is cool, but I'm not going to get a heart condition over it.
The only platform benefit I see is reducing the need to perform expensive computations to encrypt and decrypt data. Let the link take care of that and thus increase performance. Of course, how many nodes on the Internet only want to talk to their nearest neighbor? And how many routers and such are between them and their nearest neighbor? It might not even be possible to secure the link between a node and its nearest neighbor in most cases.
I doubt this technology will impact current Internet infrastructure all that much. We'll see.
Here's the Science Magazine Abstract
----Abstract-----
Electrically Driven Single Photon Source
Zhiliang Yuan 1, Beata E. Kardynal 1, R. Mark Stevenson 1, Andrew J. Shields 1,Charlene J. Lobo 2, Ken Cooper 2, Neil S. Beattie 3, David A. Ritchie 2, Michael Pepper 3
1 Toshiba Research Europe Limited, Cambridge Research Laboratory, 260 Cambridge Science Park, Milton Road, Cambridge, CB4 0WE, UK.
2 Cavendish Laboratory, University of Cambridge, Madingley Road, Cambridge, CB3 0HE, UK.
3 Toshiba Research Europe Limited, Cambridge Research Laboratory, 260 Cambridge Science Park, Milton Road, Cambridge, CB4 0WE, UK; Cavendish Laboratory, University of Cambridge, Madingley Road, Cambridge, CB3 0HE, UK.
Electroluminescence from a single quantum dot within the intrinsic region of a p-i-n junction is demonstrated to act as an electrically driven single photon source. At low injection currents the dot electroluminescence spectrum reveals a single sharp line due to exciton recombination, while another line due to the biexciton emerges at higher current. The second order correlation function of the diode displays anti-bunching under a DC drive current. Single photon emission is stimulated using sub-nanosecond voltage pulses. These results suggest that semiconductor technology can be used to mass-produce a single photon source for applications in quantum information technology.
-----End Abstract-----
If anyone has access to Science Online (http://www.sciencemag.org) you can download the PDF reprint at this link: here.
"One touch of Darwin makes the whole world kin." George Bernard Shaw
The application refers to its use in quantum cryptography. It doesn't render the encryption process uncrackable, but makes it able to detect that someone is eavesdropping and/or has broken the encryption. With current methods, you can't tell if someone has broken your key and read your message. Using quantum cryptography, you can tell when someone has read your message.
(It all goes along the lines of you can't observe something without changing it. If someone along the way intercepts the message and observes it, they will change the message and you can detect THAT on the other end.)
Where the wind blows, the tumbleweed goes.
1. The line can still be hacked, because it is possible to put a TEE into the circuit, just as long as STDOUT looks like STDIN.
Actually, if the predictions of quantum mechanics are correct, this is not possible.
The way that this works is not intuitive at all, so don't worry if you don't understand it. Einstein, Podalsky, and Rosen published a famous paper showing that quantum mechanics necessarily leads to these kind of effects.
Their goal was actually to show that quantum mechanics was unacceptable as a physical theory because they did not believe nature could possibly behave this way. But as far as we can tell, nature really does work in these mysterious ways.
More accurately, Quantum encryption IS OTP. The quantum part comes in when you generated the pad.
The one-time pad (Vernam cipher), however, is uncrackable. It has been used very heavily since it was first introduced (1917) and, beyond being arguably the simplest automated cipher ever devised, is still being proven to be completely 100% uncrackable. Unfortunately, since the key lengths are at least as long as the message, and the keys can only be used once, exchanging keys can be a bit burdensome. Quantum cryptography is basically concerned with ways of exchanging pads securely. If our current understanding of the Heisenburg principle is correct, then current quantum cryptography (in combination with OTP's) is 100% uncrackable.
The failures of previous ciphers, especially public-key ones, is due to underestimating the difficulty (or "intractability") of certain computational tasks, but no one would have ever claimed that they were COMPLETELY secure, just secure ENOUGH. The Vernam cipher does not rely on computation (beyond addition mod 2), and is completely uncrackable.
Actually, under the right circumstances the human eye can detect a single photon. However, due to the preprocessing done by the brain this signal doesn't actually reach any conscious part of your brain (for lack of better terms). But you don't need that many photon's, 10 or 20 should be perfectly detectable under the right circumstances.
Actually that is incorrect.
You'll have to look for a description of it, but it is in fact in impossible to eavesdrop and then resend the information. There is a very good description in "The Code Book" by Simon Singh. I'm not sure where else you would look.
"The Code Book", at least the british version, does describe that this unbreakable quantum encryption actually had several sucessful attempts befor this special LED appeared. I believe it was sucessfully done though the air at up to one mile. I would quote but since I'm moving the book is packed up. If you don't own the book, go buy it. It's a very good read.
This application is described fully in 'The Code Book', by Simon Singh, although the method was only theoretical at the time the book was first published."
Uhm... I believe this is wrong. The book was issued in 1999, and it contains this sentence in chapter 8:
Moreover, one paragraph further we see:
One of us is wrong -- either I'm reading this from an edited version of "the Code Book", although nowhere does it say "second edition", or the original poster needs to re-check his facts.
If you open yourself to the foo, You and foo become one.
 
In introductory physics, this is where they tell you that light is a particle and a wave, then about Schrodingers Cat, and about Heisenberg uncertainty (the more exactly you know the position of a particle, the less exactly you know its momentum, and analogous relationships with wavelength, etc).
 
Wow!, say all the young physics students. The world is inherently unknowable! Take /that/, determinists!
 
Sadly, the young physics students do not understand. The paradoxes "explained" by the above arise from the fact that a photon is /not/ a particle. It is also /not/ a wave. It's something else. But it's really useful to describe as a particle - sometimes. Other times, it's useful to describe it as a wave. We have reams and reams of equations and theorems to deal with particles and waves, so when we can model a photon as one of them, life is easy. However, since both the wave model and the particle model are inherently wrong, if you set up an experiment properly, you can produce what seems to be a paradox. Heisenberg uncertainty merely describes the breakdown of the two models mathematically. Schrodinger's Cat is an /analogy/ only - it describes a phenomenon that only applies to things like photons and electrons.
 
Interestingly, once you measure a particle/wave, you change it - since it is impossible to measure something without interacting with it. The first explanation most people hear is that when you measure a photon as a particle, there's something about a waveform collapsing, and it "becomes" a particle. This is easy to understand, but is, unfortunately, pure rubbish. If you measure it as a particle, you will get some results that are consistent with it being a particle, and you will change something about it. That's all.
 
So to get to the encryption (although I'm sure this is already (-oo, offtopic)) here's how it works: find a particle that will change in some way measurable if snooped on. Have the sender and receiver each come up with a random sequence (polarizations). Using your photons, find the common choices in the random number streams. Now - if the photon is snooped on, (measured too early) you can tell. Even if you don't notice the snooping, unless the snooper picked the same sequence of common choices, (s)he's left with nothing. And that's the end of my post.