Diffie-Hellman is a cryptographic (see Cryptography) protocol for key exchange, allowing Alice and Bob to agree on and construct a secret key over an insecure communication channel. The protocol is based on the [Diffie-Hellman Problem]? related to discrete logarithms. It is considered to be secure if an appropriate mathematical group is used. |
Diffie-Hellman key exchange is a cryptographic protocol that allows two communicators (conventionally named Alice and Bob) to agree on a secret key over an insecure communication channel. The protocol is based on the [Diffie-Hellman problem]? related to discrete logarithms. |
However it is not secure if a special type of attack called the "Man in the middle" attack is possible. This attack assumes the attacker is able to modify messages between Alice and Bob as well as read them. |
It is considered to be secure if an appropriate mathematical group is used. However it is vulnerable to the [man in the middle attack]? in which the attacker is able to modify messages between Alice and Bob as well as read them. Diffie-Hellman key exchange was invented in 1975 or 1976 during a collaboration between [Whitfield Diffie]?, [Martin Hellman]? and Ralph Merkle and was the first public proposal for establishing a shared secret over an unprotected communications channel. It had been discovered by Malcolm Williamson of GCHQ in the UK some years previously, but GCHQ chose not make it public until 1997, by which time it had no influence on research. There are many others now proposed or in use, and some of them are apparently immune to "Man in the middle" attacks. The method was followed shortly afterwards by the invention of public key cryptography using asymmetric algorithms. |
Diffie-Hellman was proposed circa 1976 and was the first public proposal for a cryptographic method that did not rely on Alice and Bob already having a shared secret before they start. It had been discovered by Malcolm Williamson of GCHQ in the UK some years previously, but GCHQ chose not make it public. Cryptographic methods of this sort are often called asymmetric algorithms. There are many others now proposed or in use, and some of them are apparently immune to "Man in the middle" attacks. |