1
$\begingroup$

I need a construction for the following: Given a group $\mathbb{G}$ of order $p$, enable a party to commit to a vector $(x_1,\ldots,x_N)\in\mathbb{G}^n$ in a way that, in a later phase, the party can open $\sum_{i=1}^N c_i\cdot x_i$ for some public coefficients $c_1,\ldots,c_N\in\mathbb{Z}_p$, while proving succinctly that this opening is done correctly.

KZG commitments can be seen as an example of this for $\mathbb{G} = \mathbb{Z}_p$ (and for 'structured coefficients' $\gamma^0,\ldots,\gamma^{N-1}$). However, for my use-case I need this for more general groups: concretely for elliptic curve groups and class groups.

The notion of Functional Commitments seems to be what I need, but I find this paper insufficient as it seems it only deals with the case $\mathbb{G} = \mathbb{Z}_p$.

Any pointers would be very welcome. Thanks!

$\endgroup$
2
  • $\begingroup$ Your question only mention succinct openings; do you also need the commitments themselves to be succinct? $\endgroup$ Commented Jul 3 at 21:12
  • $\begingroup$ Thanks @GeoffroyCouteau. Yep, also need the commitments to be succinct (perhaps with some non-succinct setup of course) $\endgroup$
    – Daniel
    Commented Jul 5 at 13:32

1 Answer 1

0
$\begingroup$

You can try looking at more general functional commitments, which work for functions $f: \mathcal{X} \to \mathcal{Y}$ for any finite sets $\mathcal{X}, \mathcal{Y}$. See these paper by de Castro and Peikert. This one is lattice based and has transparent setup!

I imagine for class groups, a functional commitment for vectors with entries in $\mathbb Z_p$ should suffice, because the ring of integers of any number field is a free module over $\mathbb Z$ (i.e any ring of integer element can be written as a $\mathbb Z$ linear combination of some basis elements). That is, you can uniquely represent any ring of integer element as a vector of integers with respect to the basis. You can take class group element representatives to be some ring of integer elements (i.e vectors with integer entries).

$\endgroup$
1
  • $\begingroup$ Thanks for the reference. Yeah I imagined if worst comes to worst, we could make do with a generic commit-and-proof-style approach. Generic functional commitments would fit the role, and this reference is indeed good in this front. I was just hoping to get something that actually exploits the group properties. $\endgroup$
    – Daniel
    Commented Jul 5 at 13:34

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.