0
$\begingroup$

I'm new to cryptography, sorry if this question is dumb.

I've just read the paper of Dwork and Naor Pricing via Processing or Combatting Junk Mail and understand the example of squaring on $\mathbb{Z}/p\mathbb{Z}$. Now I'm wondering how to apply the proof of work for a different kind of data, for example, I want to check that

if someone has converted successfully a wav audio file to mp3

Is there any work (papers, books) on this kind of problem?

Thanks for any hint.

$\endgroup$

1 Answer 1

1
$\begingroup$

If you want to determine whether a file has undergone format conversion, I'd venture that no algebraic approach (such as squaring over integers modulo $p$) is likely to be directly useful.

Assume the standard POW settings, so my understanding is that someone who has converted file.wav to file.mp3 is to use proof of work to demonstrate this conversion to another party.

Also, under the standard setting, one cannot just display the two files which a second party can check are two encodings of the same clip. The mechanism is used to stop spammers, so last thing we want is for them to submit the two files for someone (the spam target) to check.

There are at least two problems with this idea:

  1. POW can demonstrate the quantity of work done but not the aim of the work done.
  2. The actual algorithm for the conversion, regardless of what kind of encoding is used, cannot be usefully converted to an algebraic function with the properties required for POW, such as a Fiat-Shamir type of function, e.g., squaring modulo an integer.
$\endgroup$
1
  • $\begingroup$ Thank you. I initially think of some features (e.g. Fourier transformation result of some random parts) of the audio file which can be "easily" checked to know whether the result is genuine. $\endgroup$ Commented Nov 8, 2023 at 16:02

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.