vacp2p / research

Thinking in code
MIT License
62 stars 4 forks source link

Waku-Rln-Relay Extension (Journal submission) #100

Open staheri14 opened 2 years ago

staheri14 commented 2 years ago

Below is a list of action items, extracted from the reviews received from DINPS 22, to be taken for the extended version of the waku-rln-relay article for journal submission. These items have not been addressed in the initial submission due to the page limit and the fact that the project was still in progress and not deployed on a large scale.

  1. Maintaining Merkle tree in a storage efficient manner suited for resource-limited devices
  2. Benchmarking and performance evaluation, especially the proof verification time.
staheri14 commented 2 years ago

cc: @oskarth We already have a rough solution for the first item, I suggest prioritizing it for this/next next quarter and publishing it as a short paper.

oskarth commented 2 years ago

Makes sense to me, not obvious to me how much is covered by prior art here, but it is something we should address regardless.

staheri14 commented 2 years ago

Another extension idea: How to break user anonymity by making proofs against an old version of the Merkle tree. Basically, what can go wrong if the index of one's identity public key gets (or the range of the index) exposed to/inferred by the network.