The cool thing about quantum mechanics is that a man-in-the-middle attach won't work!
The reason for this is the Heisenberg Uncertainty Principle, which states that there are certain attributes of a particle where we can't know both of them perfectly. The classic example of this is position and energy.
The relevant attributes here, though, are angular momentum (spin) in the "left/right" and "up/down" directions. Think of a particle as two arrows, one pointed either left or right, and the other pointed either up or down. The Heisenberg Uncertainty Principle states that if I know with 100% accuracy that a particle is pointed left (or right), I can't know anything about if it's pointed up or down, and vice versa. In fact, if I measure the left/rightness of a particle, the up/downness gets set randomly to be either of them.
So quantum communication works more or less like this: We agree that left and up correspond to a 1, and right and down correspond to a zero. So if I want to send a 1, I randomly choose a either a left or an up, and send it to you. Then, on your end, you randomly choose an axis to measure on, and after I've measured, we tell each other (through a public channel) which axis we measured on. If we agree, we use the bit, otherwise, we throw it away.
Ok, so now, what if Eve comes in the middle and steals our electron, reads it, and sends it again. Well, she doesn't know what axis to measure on! So if I send it on the up/down axis, Eve measures it on the right/left axis, and you measure it on the up/down axis, you have a 50% chance of getting a wrong answer. So if we take a few bits at random and publish them, we can get a pretty good idea if someone's snooping.
Also, Eve can't copy our particle, because if she could, she could read one axis from one particle and the second from the other, which would violate the Uncertainty Principle (this result is known as the No-Cloning theorem).
That only works if the public channel is trustworthy. If Eve can man-in-the-middle it too, she can just conduct separate instances of the protocol with Alice and Bob, forwarding on the classical data.
I like to think that we'll uncover a new phenomena for communication, but be troubled the the incredible amount of noise in the medium, only to discover it's the channel for intergalactic WiFi.
Actually, a bunch of commercial quantum crypto systems were hacked a few years ago, in spite of being "theoretically unhackable". The details are at http://www.techrepublic.com/blog/security/quantum-hacking-cr..., but basically you can convince avalanche diodes that they're only receiving one photon by blinding them, and this lets you bypass the security inherent to the No-Cloning theorem and the Uncertainty Principle that quantum crypto relies on .
[+] [-] pranjalv123|14 years ago|reply
The reason for this is the Heisenberg Uncertainty Principle, which states that there are certain attributes of a particle where we can't know both of them perfectly. The classic example of this is position and energy.
The relevant attributes here, though, are angular momentum (spin) in the "left/right" and "up/down" directions. Think of a particle as two arrows, one pointed either left or right, and the other pointed either up or down. The Heisenberg Uncertainty Principle states that if I know with 100% accuracy that a particle is pointed left (or right), I can't know anything about if it's pointed up or down, and vice versa. In fact, if I measure the left/rightness of a particle, the up/downness gets set randomly to be either of them.
So quantum communication works more or less like this: We agree that left and up correspond to a 1, and right and down correspond to a zero. So if I want to send a 1, I randomly choose a either a left or an up, and send it to you. Then, on your end, you randomly choose an axis to measure on, and after I've measured, we tell each other (through a public channel) which axis we measured on. If we agree, we use the bit, otherwise, we throw it away.
Ok, so now, what if Eve comes in the middle and steals our electron, reads it, and sends it again. Well, she doesn't know what axis to measure on! So if I send it on the up/down axis, Eve measures it on the right/left axis, and you measure it on the up/down axis, you have a 50% chance of getting a wrong answer. So if we take a few bits at random and publish them, we can get a pretty good idea if someone's snooping.
Also, Eve can't copy our particle, because if she could, she could read one axis from one particle and the second from the other, which would violate the Uncertainty Principle (this result is known as the No-Cloning theorem).
[+] [-] comex|14 years ago|reply
[+] [-] stewie2|14 years ago|reply
If the two atoms are linked via entanglement, why are they also connected by optical fiber?
if aiming a laser at the first atom will change its state, then shouldn't the state of the second atom change too at the same time?
" That photon zooms along the optical fiber to other optical cavity containing the other atom."
photon zooms? that means on the other side of the optical fiber, the second atom absorbs more than just one photon?
[+] [-] jean-|14 years ago|reply
I believe that "zoom" means "to travel very quickly" in the sentence you quote.
[+] [-] powertower|14 years ago|reply
Wouldn’t this break the entanglement?
That photon is going to be absorbed/re-emitted countless amounts of times as it "travels" though the optical fiber.
Every time this happens, the wave function is collapsed.
[+] [-] mjjones0|14 years ago|reply
[+] [-] wicknicks|14 years ago|reply
[+] [-] pfraze|14 years ago|reply
Browsing the UWW will be crazy.
[+] [-] indiecore|14 years ago|reply
May I be the first to say, challenge accepted.
[+] [-] pranjalv123|14 years ago|reply
[+] [-] Monotoko|14 years ago|reply
[+] [-] ktizo|14 years ago|reply