sus0pid / bowtie-artifacts

BowTie Artifacts

This repo shows the artifacts of the paper Stopping Silent Sneaks: Defending against Malicious Mixes with Topological Engineering. We developed two tools:

• routesim to evaluate the topologies on their expected security metrics (i.e., expected time to first compromise/number of messages to first compromise for the clients) of typical dynamic email-like usage;
• Mixnets topology generator (MTG) to produce the mixnet topologies and calculate the security/performance metrics (e.g., end-to-end compromised path, guessing entropy, and queuing delay) based on the generated topologies.

Tools

1. routesim

To enable our evaluation of time to first compromise metric based on realistic Mixnets usage, we implement routesim in rust to support the dynamic, multi-message user scenario, aiming at estimating user's resilience against client enumeration. It takes the output generated by MTG as the input data, and evaluates the security level of clients over time. Please click the link above to read the detailed instructions of routesim.

2. MTG

We implemented a Mixnet topology generator incorporating the four mixnet construction algorithms in Python. It generates the mixnet topological configuration and the amount of fully compromised traffic that the adversary can see. Please click the link above to see detailed information about MTG.