Accurate estimates of the data complexity and success probability for various cryptanalyses (Q2430686)

This paper deals with statistical attacks against ciphers when the attacker has a number of binary samples of plaintext/ciphertext pairs and attempts to obtain some information on the key. It contains: {\parindent=6mm\begin{itemize}\item[{\(\bullet\)}] a general framework to estimate the number of samples that are required to perform a statistical cryptanalysis; \item[{\(\bullet\)}] an algorithm that computes the number of samples which are required for achieving some given error probabilities; \item[{\(\bullet\)}] an explicit formula that gives a good estimate of the number of required samples; \item[{\(\bullet\)}] a formula for the success probability which generalizes the result obtained by \textit{A. A. Selçuk} [J. Cryptology 21, No. 1, 131--147 (2008; Zbl 1147.68510)]. \end{itemize}} The considerations are based on the binomial distribution and asymptotic expansions of the beta distribution.