Score:1

Non-uniform plaintext distribution -- AES attack

jo flag

Are their any attacks on AES (any mode) in which before the attack is mounted, the adversary is aware that the input distribution is non-uniform? Most of the work assume that the plaintext is uniformly distributed. This is with reference to the following paper- https://eprint.iacr.org/2019/491.pdf

miraunpajaro avatar
in flag
Usually there is no assumption that the plaintext is distributed in any way. Definitely not in the standard definitions.
nivedita avatar
jo flag
Many authors assume that the plaintext is uniformly distributed. Please find the link below: https://eprint.iacr.org/2019/491.pdf
kr flag
@nivedita: You have linked a *single* article. Why are you saying "*many* authors assume..."?
miraunpajaro avatar
in flag
@nivedita I'm not familiar with side channel analysis. However, in cryptography, AES is supposed to be secure against chosen message attacks. This implies that the plaintext is chosen by an adversary. And an adversary may choose any plaintext it wishes, and may sample from any distribution it likes. Also in the real world, plaintexts are definitely not uniform. (E.g, if its english the letter t is more common that z). So plaintext being uniform is not usually a reasonable assumption.
I sit in a Tesla and translated this thread with Ai:

mangohost

Post an answer

Most people don’t grasp that asking a lot of questions unlocks learning and improves interpersonal bonding. In Alison’s studies, for example, though people could accurately recall how many questions had been asked in their conversations, they didn’t intuit the link between questions and liking. Across four studies, in which participants were engaged in conversations themselves or read transcripts of others’ conversations, people tended not to realize that question asking would influence—or had influenced—the level of amity between the conversationalists.