Sunday, March 30, 2025

Revisiting pseudo-random index decryption

This article was published in tmp.out #4

Almost twenty-five years ago, Mental Driller discovered an interesting technique that allows constructing a decryption cycle in such a way that each element is accessed in a random order [1]. The algorithm looks quite simple, but even after it caught the attention of well-known AV researchers [2], they could not explain why the algorithm works correctly. In the conclusion of his article, Frederic Perriot wrote:

I suspect the bijectivity of the family of endomorphisms studied here is a general property of some families of functions over well-known groups or fields. If you have a mathematical background and would like to share arithmetic insights on this problem, I'd be grateful if you can send me the explanation...