New bounds and constructions for authentication/secrecy codes with splitting (Q1176594)

The article deals with the problem of communicating certain information to a receiver in such a way as to minimize an opponent's success to mislead the receiver by sending a fraudulent message to the receiver hoping to have it accepted as authentic. Using the mathematical authentication model introduced e.g. in \textit{G. J. Simmons} [Lect. Notes in Comp. Sci. 196, 411-432 (1985; Zbl 0575.94011 )] authentication codes with splitting are investigated. First, bounds for the probability that the opponent can deceive the receiver are given for the variety of cases where the opponent is assumed to observe a certain number of messages before sending his fraudulent message. Then the construction of authentication codes which meet one or more of these bounds is presented. In all cases considered the codes are unconditionally secure, i.e. it is assumed that any opponent has unlimited computational resources.