Re:NSA seeks to build quantum computer that could crack most...
by sillysaurus2 on Thursday January 02, 2014 @05:56PM
(#7002978)
What would you use as an RSA replacement? Beyond that, is there a technique that's still relatively simple like RSA (humor me) but impervious to publicly-known quantum attack vectors like Shor's algo?
[ Parent
| on HN ]
Re:NSA seeks to build quantum computer that could crack most...
by tptacek on Thursday January 02, 2014 @06:07PM
(#7003042)
Here's a good place to start: http://pqcrypto.org/
(This is Daniel J. Bernstein).
Re:NSA seeks to build quantum computer that could crack most...
by rdtsc on Friday January 03, 2014 @12:30AM
(#7004314)
I was going to jump in and say ECC
https://en.wikipedia.org/wiki/Elliptic_curve_cryptography
but, searching I found that it is apparently vulnerable to quantum computing attacks (wikipedia points me to: Nielsen, Michael A.; Chuang, Isaac L. Quantum Computation and Quantum Information. p. 202) and also found this:
http://www.mathcs.richmond.edu/~jad/summerwork/ellipticcurve...
Re:NSA seeks to build quantum computer that could crack most...
by scythe on Friday January 03, 2014 @02:56AM
(#7004583)
Elliptic curve cryptography is logically equivalent to prime-factorization cryptography in a sense: both are special cases of the hidden subgroup problem. DLP is also a version of the hidden subgroup problem. Any variant of HSP can be broken by a quantum computer using a variant of Shor's algorithm.
Re:NSA seeks to build quantum computer that could crack most...
by venomsnake on Friday January 03, 2014 @04:45AM
(#7004756)
How broken is the question - NSA cracks 2048 keys we just double the key size and are safe for 3-4 more years, or making it so trivial it doesn't matter how big the key size is.
Re:NSA seeks to build quantum computer that could crack most...
by yk on Friday January 03, 2014 @01:45AM
(#7004470)
In one of the crypto talks at the 30C3 the speaker mentioned that quantum computers are actually a greater danger to ECC, because of the smaller key sizes. ( I think this sounds plausible, but I am not an expert.)
Re:NSA seeks to build quantum computer that could crack most...
by betterunix on Thursday January 02, 2014 @09:47PM
(#7003868)
The Learning With Errors problem is promising:
https://en.wikipedia.org/wiki/Learning_with_errors
When last I checked there was still a lot of debate about choosing secure parameters for LWE-based cryptosystems, so it would be pretty risky to start using that right now.