3.3 Authentication and Identification

3.3 Authentication and Identification

Authentication protocols are designed to provide two or more specific entities communicating over an open network to achieve some cryptographic goals such as confidentiality,data integrity, entity authentication, message authentication, not repudiation, and key authentiation.

An authenticaiton is one of the most inportant of all informaion security ob jectives. As discussed in [14] it must be easy for anyone to recognize the signature as authentic, but impossible for anyone other than the legitimate signer to produce it.In public key cryptography there is an especially easy way to identify oneself in such a way that no one could be simply pretending to be you. Let A (Alice) and B (Bob) be two users of the system. Let f_A be the enciphering transformation with which any user of the system sends a message to Alice, and let f_B be the same for Bob. For simplicity, we assume that the set M of all possible plaintext message units and the set C of all possible ciphertext message units are equal, and are the same for all users. Let M be Alice’s “signature” (perhaps including an identification number, a statement of the time the message was sent, ets.). It would not be enough for Alice to send Bob the encoded message f_B(M ), since everyone knows how to do that, so there would be no way of knowing that the signature was not forged. Rather, at the beginning (or end) of the message Alice transmits f_B f_A^( -1)(M ). Then, when Bob deciphers the whole message, including this part by applying f_B^-1, he finds that everything has become plaintext except for a small section of of jibberish, which is f_A^(-1)(M ). Since Bob knows that the message is claimed to be from Alice, he applies f_A (which he knows, since Alice’s encihpering key is public), and obtains M. Since no one other than Alice could have applied the function f_A^(-1) which is inverted by fA, he knows that the message was from Alice.