2 ms·
So, to summarise: * Choose an Elliptic Curve (EC) and a large prime; * The rational points on the EC modulo the prime give you a group; * Your private key is
by ColinWright 7y ago
So, to summarise:
* Choose an Elliptic Curve (EC) and a large prime;
* The rational points on the EC modulo the prime give you a group;
* Your private key is a large integer N;
* Perform Diffie-Hellman-Merkle-Williamson (DHMW) in the group;
* You now have a shared point in the EC;
* Use a symmetric cipher using the shared point as the key.
Performing DHMW goes as follows:
* A and B agree on a point P in the EC;
* A and B each choose a large number as their secret key;
* For simplicity, or confusion, we'll say that X's secret key is X;
* They compute AP and BP as their public keys and exchange them;
* When A receives BP they compute A(BP), B does likewise;
* Now they share a point: (AB)P.
If you have the language of groups, and understand DHMW in that context, ECDHMW is trivial, as is (this version of) EC PK-Cryptography.
TL;DR : ECC is simply Symmetric Cryptography using a key that's been agreed by using DHMW in an Elliptic Curve group.
- archgoon 7y agoFor those like me, who are confused by the acronym DHMW, it's another name for 'Diffie-Helman'. Diffie and Helman felt that the entire scheme was based on work that Ralph Merkle did (same guy as Merkle Trees), and so should be included in the scheme name, and I think that Williamson refers to Malcolm J. Williamson, a GHCQ employee, who had secretly developed the same system 7 years before Diffie and Helman. What I don't know is why Williamson and not Cocks and Ellis are part of the acronym.
- ColinWright 7y agoCocks and Ellis developed what we know as RSA, Malcolm Williamson developed what most people call Diffie-Hellman. I did give the full version before the acronym on the fifth line of my comment, so if people are confused it might be why I'm referring to DHMW and not just DH. Reading up about it all it's pretty clear that Merkle and Williamson should get more credit than they do.
- dwoozle 7y agoNobody should get credit for inventing something in secret. You make your decision — invent for corporations or the military industrial complex and get the big bucks or the jingoistic frisson, or for academia and get the accolades. No way a person should double dip.
- ZenPsycho 7y agoso you would revoke credit from Turing?
- cortesoft 7y agoDid anyone else discover what he did before his work became public?
- ZenPsycho 7y agoyep, eniac and univac were the the first publicly known computers. it became known later that turing’s work predated eniac.
- dTal 7y agoCredit for what? The theoretical work for which Turing is most known - the concept of Turing machines and undecidability - was published publicly and before the war. He did not contribute to the design of a physical computer until after the war. Colossus, the code-breaking "computer", was not of his design (and the man who did design it is not at all famous for it).
- ColinWright 7y agoIt's not a question of credit, it's a question of attribution and historical accuracy.