Quantum Computers

[ecoli]

Can anyone explain to me how quantum computers work?

[Verusamore]

I think how it works is obvious to anyone who knows the definition , as to me

I have no idea of quantum or how it works but I think it has something to do with 'entanglement' ,' chaos' .

 

I want one , though I don't think it would be retailed for desktop computing until networking becomes invulnerable since it would be way to much power for one to have .

There is a concept being discussed in security which has also something to do with 'quantum encryption' & 'entanglement' .

[Jordan14]

Yep it's all about entanglement, the mysterious link between the electron and the positron which decay from the pion.

The elctron and the positron however far away from each other form a bond and what ever the electron doesthe positron does the exact opposite, this makes communication in a computer a damn lot faster, I think i'n right.

Theres a debate going on at the moment related to these links, go and google pseudo-telepathy.

[ecoli]

so information could be computer countless times faster on chips that are countless times faster. Sounds important in developing nanobots

[Jordan14]

Here's a good link:

 

http://www.newscientist.com/hottopics/quantum/quantum.jsp?id=ns99993114

[5614]

that is a good link, i can give more info if people need it, but i wont write more unless you ask!

 

can i just remind you guys that there still must be a 'standard' median to transport the (to make it simple lets say:) "the entanglement data" at the moment a laser is used, i think (personal opinion) that one day radio waves could be used and therefore satelites etc. but the fact is that quantum computers are still limited to the speed of light (aka 'c').

[1veedo]

As far as encryption goes, the uncertainty principle works on your side.

 

The key to the encryption required to crack data is always either standard, in which case you use google to hack it, or it is included someware in the file, in which case you apply software to search for it. So, as it is, anything and everything no matter how ‘secure’ it is can be cracked.

 

In a quantum encrypted message any attempt to read the data would only give part the data and some more data that is corrupt. Plus, the receiver of the data would know right away what 'pure' info was obtained.

[ecoli]

A quantum computer could crack an encripted system in mere seconds, while today's best computers would take eons. It would be able to go through all the prime/ semi-prime numbers in seconds.

[5614]

how long would it take a quantum computer to break an encryption code set by another quantum computer?

[Jordan14]

I presume it would take the same time a normal computer would to break an encryption make by another normal computer. Because it's not the complexity of the coding it's the difference between the stadard of the two computers, so comparing to standard computers is the same as comparing to quantum computers. I think.

[1veedo]

I dont think so...?

 

I'm pretty sure that it's impossible to receive data if you're not the intended recipicant.

 

At the end you recieve the data, but along the way, unless you can trick into receiving the data yourself, any attempts to gain it would result in 'partial' forms of information.

 

Quanumt enrcyption is not like a regular message with half the amount of information required.

[timo]

how long would it take a quantum computer to break an encryption code set by another quantum computer?

 

Current encryption methods work because it´s very hard to find the prime factors of the product P of two large primes. Don´t ask me how that actually works - I´m just repeating what I´ve heard everywhere and I don´t know the details. The length of this prime, however, is small compared with the length of the message. If you are able to break down the publicly known P into it´s primes it seems like you can easily decode the message. Normal computers are quite slow doing this.

On Quantum Computers, however, algorithms have been proposed that would run much more efficiently and would do the job in a reasonable time.

 

If you for example chose a key with the same length as the message, a simple XOR-operation will destroy all information stored in the message except it´s size. This can be considered as completely unbreakable (test: Try to find the message of 1MB of random numbers). The problem here is that you have to transport your key to the reviever as well as the actual message.

 

That´s where Quantum-Cryptography comes into play: Because it is possible to determine if someone intercepted your message sent by QM-ways you can simply send the key to the reciever. If someone intercepted the key you will notice that and simply send another one untill it hasn´t been intercepted. Now if you successfully sent your key without anyone intercepting it you can simply decode your message with is, send it per E-Mail and let the other one decode it with the key only the two of you have.

Above method of QCryptography was referred to as the BB 84 Protocol by the lecturer of my Quantum Information lecture if that helps anyone.

 

To sum iit up: It is impossible to break a message that was encrypted by above means of Quantum-Cryptography, regardless of what kind of computer you use.

[5614]

To sum iit up: It is impossible to break a message that was encrypted by above means of Quantum-Cryptography, regardless of what kind of computer you use.

 

whilst your explanation good and your details good too and though im in no position to prove it wrong...

 

all i will say is that:

 

"no matter how good a code you make, it can always be cracked.... its just a matter of time"

 

that is a famous saying said by some random dude who was probably very famous, i just called him famous because i (at this moment) dont have a clue who said it!!! all the same, it is quite a famous saying and there are several similar ones like it, but you get what i mean.

 

so im not saying you're wrong because i dont understand quantum cryptology well enough, but unless there is good (easy for me to understand!) proof, i dont think i will differ from the above saying - in other words i disagree with you (the quoted part) but i cant prove you wrong!

 

in other words thats my personal view, feel free to prove me wrong, but thats what i think!

[timo]

May strongly depend on if that guy called a file of random numbers a code I think (that´s all you have left when you apply a random key of the same length to a message). Have to leave now, but I´ll try to go into more detail later.

[5614]

well when a computer is making random... it isnt REALLY random as it is based on an algorithm. so sure, it may be linked to the clock and be immensly complicated, but the fact that it is based on an equation means it can be cracked.

 

but i'll wait for "more detail later" until i say more.

 

i'll just repeat though, that you know more than this than i do, so all im doing is saying "i think... - feel free to argue, its just my view with no proof!"

[ecoli]

I'm sure people used to think that using large prime numbers was unbreakable code. Who knows what future technologies have in store.

[timo]

well when a computer is making random... it isnt REALLY random as it is based on an algorithm.

 

Oh, if you agree that it´ll work if the numbers are really random, then I can save my time making thoughts about how to explain it (I had to look up a few things so I´d prefer not doing so). Indeed, computers only create pseudo-randoms but above encryption method simply assumed you take real randoms. They don´t have to be made with a computer.

 

Of course I could think of few (=no) methods for creating large ammounts of (pseudo-) randoms other than using a computer so the problem would indeed lie in the random generator (apart from other problems like that I could simply steal the information from any of the computers).

As for real randoms: This somehow rings a bell. I think it´s one of the advantages of Quantum Computers that they are able to produce real randoms. Assume you prepare a state [math] \frac{1}{\sqrt{2}} \left( |0> + |1> \right) [/math] with |0> and |1> being perpendicular states (polarization of a photon for example). If you now measure on |0> it´ll give you a count 50% of the time without any correlations between the different measurements.

Well, that´s all idealized. The real problems lie in the actual building of Quantum Computers. The theoretical background and some algorithms are allready there.

[timo]

I'm sure people used to think that using large prime numbers was unbreakable code. Who knows what future technologies have in store.

 

It´s another level of unbreakable, at least. I can write a program that factorizes large numbers within 5-20 minutes if the numbers fit into an integer supplied by that language. Methods of factorizing large numbers can be rather trivial. The method only works because my program would probably run a few years before giving a result.

 

The other one is unbreakable even in theory. Unbreakable, however, -by how I use the term- does not mean that there´s absolutely no chance of getting the information. If, for example, you intercept the message "sd4a" which was sent by above means, just guessig the original message was your favorite four letter word has at least a finite chance of being correct. But that´s not breaking the encryption and your chances drastically decrease with the length of the message.

[5614]

Oh, if you agree that it´ll work if the numbers are really random

 

well thinking about it now, you couldnt use a set method to crack a code with random numbers in it, however, you could always use trial and error, it'd take a very long time, but assuming it is a code e.g. 1 = Z or whatever, just a lot harder! it is still crackable.... just takes a while and the computer would need to recognise when it had cracked the encryption.

[timo]

>> and the computer would need to recognise when it had cracked the encryption.

 

That´s the major point. If I send 100 Bytes that you don´t have any additional infos on, every of the 8^100 possibilities could be the message. You could as well just intercept the length of the transmission and make guesses what it could be or not intercept anything and simply guess.

[5614]

the fact that it would take a VERY long time does not stop the fact that it'd be theoretically possible to do so.

 

as some would say: "no code is uncrackable given an amount of time"

[timo]

We seem to have different views on crackable. If I take the set of all possible messages (it´s even a countable one) and say "your message is in there, I cracked your code" then I have cracked each code even before the actual message is written.

 

EDIT:

To clearify on my example with the 100 Bytes. You do know the message is definitely one of those of the set of all messages containing 100 Bytes (size of this set is 8^100). However, you have absolutely no way to know which one it is because they all do have the same probability (1/(8^100)) to be the right one.

[5614]

well yeah!... what im saying is that you give a computer a code then it'd be able to crack it just by going though each individual option... it is possible.

 

a quantum computer would do that even quicker - (it was always do-able, just quicker now)

 

now the only hard bit is for the computer to recognise when the code has been cracked... i dont know how it would be done - well, if it were text in a certain langauge (or coding ie. programming coding ie. java or something) basically anything which exists (ie. language (spoken or programming)) then it'd be easy.

 

but if you took an (almost) random selection of letters and then encoded that it'd be harder to know when you have cracked the code - as the end result would still be (near random) letters aka nothing useful or nothing recogniseable, so you just wouldnt know.

[Molotov]

This might be a little OT but does quantum entanglement allow for faster than light speed data transmission?

 

Like you could send a probe to Saturn with a "quantum transmitter" on it that was entangled with an identical unit on earth. Would this allow real time transmission to the probe?

[5614]

no as there needs to be a classical mode of communication to (keep it simple) transport the "entanglement data".

 

at the moment lasers are used... however in theory, anything could be used.