Chaum-Pedersen ZKP

Introduction

Chaum-Pedersen zero knowledge proof implementation in Python. Quick implementation of the protocol is available chaum_pedersen.py. It serves only as demonstrational purposes for the background calculations.

In /app there is an implementation of client-server communication via gRPC.

Run locally

Prepare environment

scripts/update_proto.sh
scripts/venv_setup.sh
source venv/bin/activate
docker-compose up -d redis # or your local Redis instance

python3 app/victor/victor.py

Use client

Register

Client is accepting a number (secret), registers it with the server and returns the random prime number generated. This prime number should be stored, so it can be reused later on for logging in.

python3 app/peggy/peggy.py register -s ${SECRET}

Example request:

python3 app/peggy/peggy.py register -s 123

Example response:

157360668734434990285839916876951699033602410454308426433468085480284003817301482464165455037654048596347789589086255823400592679731982756027484164480954597575310097972589187852240388920290281011504939640293166111231509991974534860763163500734488993174560764639540033076095728568655453279638565710498651806239

Login

Client is accepting a number (secret), and a prime number. It checks with the server if it can login.

python3 app/peggy/peggy.py login -s ${SECRET} -p ${PRIME}

Example request:

python3 app/peggy/peggy.py login -s 123 -p 157360668734434990285839916876951699033602410454308426433468085480284003817301482464165455037654048596347789589086255823400592679731982756027484164480954597575310097972589187852240388920290281011504939640293166111231509991974534860763163500734488993174560764639540033076095728568655453279638565710498651806239

Example response:

True

Run in Docker

Prepare environment

scripts/update_proto.sh
docker-compose up

Use client

Register

Client is accepting a number (secret), registers it with the server and returns the random prime number generated. This prime number should be stored, so it can be reused later on for logging in.

docker exec peggy python3 app/peggy/peggy.py register -s ${SECRET}

Example request:

docker exec peggy python3 app/peggy/peggy.py register -s 123

Example response:

157360668734434990285839916876951699033602410454308426433468085480284003817301482464165455037654048596347789589086255823400592679731982756027484164480954597575310097972589187852240388920290281011504939640293166111231509991974534860763163500734488993174560764639540033076095728568655453279638565710498651806239

Login

Client is accepting a number (secret), and a prime number. It checks with the server if it can login.

docker exec peggy python3 app/peggy/peggy.py login -s ${SECRET} -p ${PRIME}

Example request:

docker exec peggy python3 app/peggy/peggy.py login -s 123 -p 157360668734434990285839916876951699033602410454308426433468085480284003817301482464165455037654048596347789589086255823400592679731982756027484164480954597575310097972589187852240388920290281011504939640293166111231509991974534860763163500734488993174560764639540033076095728568655453279638565710498651806239

Example response:

True

Run unit tests

./scripts/unit_test.sh

Further work and improvements

• Prime number can be saved in a database on client's side. This way client won't be required to provide the prime number;
• Bad weather unit tests;
• Integration, e2e and coverage tests;
• Improved env variables handling;
• Healthcheck endpoints;
• Randomly choose generators g and h. Right now I have them as environment variables with default values 4 and 9. However, it would be better to select random generators on start. More literature on this:
  • Cryptography: An Introduction (3rd Edition) by Nigel Smart
  • Notes on ZKP by Jorge L. Villar
  • Lecture on Proofs by Susan Hohenberger.