Some background: Physics student, in my 4th year of a 5 year masters degree, with a strong focus on quantum physics: last friday, Fr'ex, I handed in my report on a computer simulation I wrote for Quantum Ghost Imaging. I'm writing this 'cause the descrription of Quantum Encryption in the books bugs the hell out of me. I'll stop after this, honest. But the real tech is cool as hell, and not all that complicated.
First thing you need to know: In any given axis (X, Y, or Z) somthing can be spin 'up' or spin 'down'. This is a quantum property. Always either up or down. You find out which when you measure.
What makes this interseting is that measuring the spin in /one /axis - say the X axis - can be thought of as randomizing the spins in the other axis (In a lies to children sense) So that.. Let's suppose you measure 'up' for X. Then Measure X again: Still up. You then measure Y: It's got a 50/50 chance of being up or down. Let's say we get down. You then measure X again. It has a 50/50 chance of being up or down /even though the only thing you've done is check to see what the value is in a different direction/. This is the important bit.
Now, we can measure spin. I'm not gonna get into how.
Now. I generate a stream of particles - photons, say. And I randomly switch between, oh, X and Z, measuring their spin in that axis, before sending them to you.
You randomly read their spin in X and Z as well.
Once we've sent, say, a million photons, we openly compare our lists of X and Z, Discarding all the ones where we didn't measure in the same axis: so XX is kept, ZZ is kept, XZ and ZX are discarded.
Still with me?
Finaqlly, we take a random sample from our results of 'up' and 'down' (or '0' and '1') and compare them to see if we measured the same thing. if we did? Then we know that /no-one/ intercepted the message, and can use the data sa a one-time pad: That is, random data I add onto a message to encrypt it. Used once, then discarded.
Why can I do this?
Because anyone who tried to intercept the signal would /also/ need t orandomly guess X and Z, and then send on the data - but the value would only be the same if /all three/ people measured the same axis: If even one of them measured differently from the others, then we're back to a 50/50 variance. Thus changing the data we compared.
So you know if anyone is reading the encryption pattern you're making other than the person you intend.
And.. well. As of 10 years ago, I recall a paper by BT wherein they had done this through air - a beam of light - over a distance of several kilometres. So.
none of this nonsense about qbits being needed, please!
/sciencelesson
/rant
Welcome! These forums will be deactivated by the end of this year. The conversation continues in a new morph over on Discord! Please join us there for a more active conversation and the occasional opportunity to ask developers questions directly! Go to the PS+ Discord Server.
Quantum Encryption
Mon, 2012-03-26 21:26
#1
Molikai
Quantum Encryption
Tue, 2012-03-27 04:08
#2
Arenamontanus
Re: Quantum Encryption
In my case it is people ignoring the causal problems of FTL in the setting. Maybe we should form a self-help group.
I guess the problem in EP descriptions is that there are three kinds of qubits around:
* The quantum entanglement between photons or other particles that are used in quantum cryptography. (OK with known physics)
* The quantum entangled pairs of particles used in the QE comm FTL devices (likely breaking physical law as we know it)
* The qubits inside quantum computers. (Not yet fully implemented, but likely allowed by physics)
These are very different things, although related by being "quantum". The first two are entangled particles, the third involves quantum superposition (a bit being in a mixture of on and off states) too.
—
Fri, 2012-03-30 01:55
#3
cenrae
Re: Quantum Encryption
And just how large are qubit reservoirs? Text says relatively large and expensive to transport.
Fri, 2012-03-30 04:11
#4
Arenamontanus
Re: Quantum Encryption
The devices in the gear section are apparently man-portable. I tend to think the small reservoir is about a kilogram and the big perhaps 10 kg.
Like antimatter, it is not the qubits that are heavy but the containment keeping them isolated. Or it might be that the reservoirs have a simply enormous number of qubit pairs, but the probability of them surviving the processing needed to try to transmit something means they have to be used up at a great rate. Billions of qubits lost for every classical bit transmitted via QE.
—
Fri, 2012-03-30 11:03
#5
Molikai
Re: Quantum Encryption
Ah.. My point is, REAL LIFE, You do not need some 'Qubit reservoir'. You need a Laser and a Type 2 crystal cabable of doing Spontaneous Parametric Down-conversion - It's about a centimetre long, probably.
There. You can create entangled photons. Nto that you /need/ to, since QU does not NEED entangled photons.
it just takes regular photons and a device capable of measuring their spin. (Stern Gerlach device)
And is then used as outlined above. To create what in cryptography are called 'one time pads'.
Sun, 2012-04-01 10:13
#6
Decivre
Re: Quantum Encryption
No, the reservoirs are used for FTL communications. Some as-of-yet-unrevealed process breaks the entanglement and transmits data in the process. This requires reservoirs, as you need at least as many entangled pairs as you wish qubits of data to transmit. And you cannot simply produce them on the spot as you need each reservoir of qubits to be in both the transmission point and reception point... one qubit is used up on each end.
For quantum encryption, only a single entangled pair is required. This is why quantum encryption devices can be very small... there is even a radio in the core book that can be as small as a necklace and serve to quantum encrypt your transmissions. You still need each piece of the pair on both ends of the communication for it to work, and you can't entangle particles over long distances... so far as I know.
—
Transhumans will one day be the Luddites of the posthuman age.
[url=http://bit.ly/2p3wk7c]Help me get my gaming fix, if you want.[/url]
Sun, 2012-04-01 14:21
#7
Foucault
Re: Quantum Encryption
Well, you'd also have problems of with the photon one time pad creator because of the diffraction of your laser. I think this is mentioned somewhere in the corebook that you need huge bloody phased array lasers and detectors do to this sort of thing. But those things at least cannonically exist in EP.
And I'm also totally with you there on the science in the game needing a support group. I'm a 3rd year PhD student in physics, working on high intensity laser interaction with matter. It's going to take some real effort to any sort of laser weapon to work in air, and don't get my started on plasma confinement. But the real problem is that I've run Eclispe Phase for two different groups, both of where were over 75% physics PhD students.
Now sometimes they don't realize the physics implications of what they do, like when one group decided to kill all of the Lost terrorist/child-kidnappers on a Venutian Aerostat by venting the atmosphere. They belatedly rememberd that this would sent the aerostat plummeting to the ground. So instead of sitting in the maintenence airlock until all the Lost are dead they have to attempt to get to the egocasters before they suffocate. And they don't have enough vacsuits for everyone. Yea, all those egos bit it.
One of those players also was working with the Super-Kamiokande neutrino data at the time. We had a long discussion about how we thought neutrino comms were BS, to the point where I replaced all of them with laser comms in my game. But then here comes that paper out of Fermilab where they do it!
But on the other hand the other group realized how sparse the Jovian Trojans really are and how bright even a ship not burning engines has to be. Then they called in some favors to get some high-resolution telescopic imagry of the Song-Cai-Flower. As Arenamontanus' earlier white-paper showed, you just can't hide in space. So we've basically decided that every faction knows where every other ship is in the Solar System, at all times. As well as it's approximate size and engine class. There are just Corps and Anarchist Collectives that make this their buisness to know, and sell or share it.
I certainly wouldn't give it up though. Running EP with such a science-litterate crowd is certainly a challenge. However, they actually understand a lot of the obsecure material in the setting pretty well and can do some damn clever things.[/Thread-jacking]
—
R-Rep: 2
Mon, 2012-04-02 01:54
#8
Arenamontanus
Re: Quantum Encryption
Maybe we should set up a new thread to discuss the fun of big lasers. I enjoyed thinking about beams for my space battle thing, but there are so many other issues to think about - what you can do with phased arrays, cooling, nonlinear optics, ...
I also have a neutrino physicist in my Oxford group (T2K experiment, also at Super-Kamiokande) and he was equally surprised. Score one for Eclipse Phase!
Sometimes it is simply delightful. Like yesterday, when a player perked up at my description of a big superconductor power conduit in the enemy base on an icy moonlet and just said "superconductor quenching" with an evil smile...
—