Extended TrustChain implementation for Master thesis

This is the standalone implementation of the pairwise auditing mechanism for strategyproof information dissemination in distributed transaction based reputation systems.

Installation

pip install -r requirements.txt

Usage

The runner takes a configuration file as input an performs the experiment defined in that configuration. The configuration files contain information such as the number of nodes to spawn, the length of the experiment and the data to output.

./run.py execute --config configs/test.json

Process

Pairwise auditing leads to validation and dissemination of data. It works in the following way when considering agents Alice (A) and Bob (B).

  • A requests data from B, therefore B would like to check the trustworthiness of A
  • B requests full chain of A, shares own chain
  • A shares chain, verifies B's chain and sends difference, creates block
  • B checks the chain and if complete calculates the difference in blocks, which he sends to A, signing block
  • B calculates trustworthiness of A