4 ms·
There is one public key system that is pretty much that, it's called Merkle Puzzles. It's kind of non-practical though. No one has found a system that is pra
by YZF 13y ago
There is one public key system that is pretty much that, it's called Merkle Puzzles. It's kind of non-practical though. No one has found a system that is practical and provably secure. You need the right kind of problem. Pretty much all the problems we have today that can be used for public key cryptography are thought to be hard but can't be proved to be hard. Under the assumption they are hard the system is proved to be secure. Symmetric cryptography (e.g. AES) is in a slightly different boat but again security proofs are based on assumptions about the function used and I don't think anyone has found a function that is provably secure (but there are various constructions that are used that have some provable characteristics). Also under some base security assumption there are further constructions of ciphers that can be proved secure.
- mjn 13y agoTo expand on why it's non-practical: the gap between the "easy" and "hard" directions is only quadratic, whereas key-exchange protocols based on integer factorization and other such unproven candidates for one-way functions are conjectured to have a super-polynomial gap. With such a small gap it's much harder to produce a setup that is on the one hand actually usable by the parties, but on the other hand intractable for the attacker. You might wonder if something closer to Merkle's approach could be improved to produce a larger gap, but the answer turns out to be no: http://www.boazbarak.org/Papers/merkle.pdf http://www.boazbarak.org/Papers/merkle.pdf