In this paper, we use a tool that computes exact values for expectations and standard deviations of random variables involved in generic attacks on various Feistel-type schemes in order to get a better study of these attacks. This leads to the improvement of previous attacks complexities: either we need less messages than expected or we can attack more rounds. These improvements are given for different sizes of the inputs. We also show that for rectangle attacks, there are more differential paths than presented in previous attacks and this strengthens the attacks.
CITATION STYLE
Volte, E., Nachef, V., & Marrière, N. (2017). Improvements of attacks on various feistel schemes. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 10311 LNCS, pp. 321–344). Springer Verlag. https://doi.org/10.1007/978-3-319-61273-7_16
Mendeley helps you to discover research relevant for your work.