Linear Cryptanalysis of Reduced-Round Speck with a Heuristic Approach: Automatic Search for Linear Trails

摘要

Previous research on linear cryptanalysis with Speck has proved that good linear trails and a meaningful distinguisher for variants of Speck can be found. In this paper we use two different linear approximations of modular addition to search for even better linear trails. Also, we have added a heuristic to search for large bias approximations for the state conversion approach. We will explain how the automatic search works and discuss its performance. Finally we illustrate that linear approximations with large bias exist in variants of Speck.