Abstract. We demonstrate the existence of an efficient block cipher with the property that whenever it is composed with any non-perfect cipher, the resulting product is strictly more secure, against an ideal adversary, than the original cipher. We call this property universal security amplification, and note that it holds trivially for a one-time pad (a stream cipher). However, as far as we are aware, this is the first efficient block cipher with this property. Several practical implications of this result are considered.
Abstract. Absolute lower limits to the cost of cryptanalytic attacks are quantified, via a theory of guesswork. Conditional guesswork naturally expresses limits to known and chosen plaintext attacks. New inequalities are derived between various forms of guesswork and variation distance. The machinery thus offers a new technique for establishing the security of a cipher: When the work-factor of the optimal known or chosen plaintext attack against a cipher is bounded below by a prohibitively large number, then no practical attack against the cipher can succeed.As an example, we apply the technique to iterated cryptosystems, as the Markov property which results from an independent subkey assumption makes them particularly amenable to analysis.
We formally study iterated block ciphers that alternate between two sequences of independent and identically distributed (i.i.d.) rounds. It is demonstrated that, in some cases the effect of alternating increases security, while in other cases the effect may strictly decrease security relative to the corresponding product of one of its component sequences. As this would appear to contradict conventional wisdom based on the ideal cipher approximation, we introduce new machinery for provable security comparisons. The comparisons made here simultaneously establish a coherent ordering of security metrics ranging from key-recovery cost to computational indistinguishability.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.