The Feistel construction is a fundamental technique for building pseudorandom permutations and block ciphers. This paper shows that a simple adaptation of the construction is resistant, even to algorithm substitution attacks—that is, adversarial subversion—of the component round functions. Specifically, we establish that a Feistel-based construction with more than \(337n/\log (1/\epsilon )\) rounds can transform a subverted random function—which disagrees with the original one at a small fraction (denoted by \(\epsilon \) ) of inputs—into an object that is crooked-indifferentiable from a random permutation (or ideal cipher), even if the adversary is aware of all the randomness used in the transformation. Here, n denotes the length of both the input and output of the round functions that underlie the Feistel cipher. We also provide a lower bound showing that the construction cannot use fewer than \(2n/\log (1/\epsilon )\) rounds to achieve crooked-indifferentiable security.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Crooked Indifferentiability of the Feistel Construction

  • Alexander Russell,
  • Qiang Tang,
  • Jiadong Zhu

摘要

The Feistel construction is a fundamental technique for building pseudorandom permutations and block ciphers. This paper shows that a simple adaptation of the construction is resistant, even to algorithm substitution attacks—that is, adversarial subversion—of the component round functions. Specifically, we establish that a Feistel-based construction with more than \(337n/\log (1/\epsilon )\) rounds can transform a subverted random function—which disagrees with the original one at a small fraction (denoted by \(\epsilon \) ) of inputs—into an object that is crooked-indifferentiable from a random permutation (or ideal cipher), even if the adversary is aware of all the randomness used in the transformation. Here, n denotes the length of both the input and output of the round functions that underlie the Feistel cipher. We also provide a lower bound showing that the construction cannot use fewer than \(2n/\log (1/\epsilon )\) rounds to achieve crooked-indifferentiable security.