作者: Christian Rechberger , Christophe De Cannière
DOI:
关键词: Computer science 、 Theoretical computer science 、 Function (mathematics) 、 Inversion (discrete mathematics) 、 Reversing 、 Random graph 、 Margin (machine learning) 、 Hash function 、 Property (programming) 、 SHA-1
摘要: In this paper, we examine the resistance of popular hash function SHA-1 and its predecessor SHA-0 against dedicated preimage attacks. order to assess security margin these functions attacks, two new cryptanalytic techniques are developed: Reversing inversion problem: idea is start with an impossible expanded message that would lead required digest, then correct until it becomes valid without destroying property. P 3 graphs : algorithm based on theory random allows conversion attacks compression less effort than traditional meet-in-the-middle approaches. Combining techniques, obtain preimage-style shortcuts for up 45 steps SHA-1, 50 (out 80).