I understand that the main reason why the RCTTypeFull scheme is not applied to multi-input transactions is that the true signing keys in the MLSAG algorithm are on the same row, which impacts traceability (one input identified, all identified).

Has anybody invested time in trying to reformulate the MLSAG algorithm to circumvent this limitation? At first sight it shouldn't be terribly difficult ...

I know that StringCT is coming, but in the meantime the generalization would cut space needs by half ....

  • Can you please provide more clarity on your "first sight" impression? Dec 20, 2017 at 16:54
  • Isn't it just a matter of re-shuffling indexes? It would probably be more tricky to put on paper than to actually implement it in the code, I think
    – Kurt
    Dec 21, 2017 at 9:01
  • If you shuffle indexes, your commitment sum will be incorrect.
    – Luigi
    Dec 21, 2017 at 17:26


Your Answer

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

Browse other questions tagged or ask your own question.