The pyfp library

Author

Hans Georg Schaathun

Title

Attacks on Kuribayashi's Fingerprinting Scheme

Journal

IEEE Transaction on Information Forensics and Security April 2014

Simulator source code

The pyfp library was developed to obtain experimental results for the paper. Download source code as tar ball.

Version 0.2 of the pyfp library has been augmented for the sake of another paper, currently under review.

The download file contains the following elements:

1. The pyfp library
2. Necessary modules from the pysteg library
3. The fptest.py script to run simulations.
4. Two batch files listing all the commands with parameters used to generate data for the paper.
   1. prod-orig/batch without interference removal
   2. prod-ir/batch with interference removal
5. The Bourne shell script batch.sh which can be used to submit all the jobs needed to a PBS or Torque queue. It must be amended to fit local directory structure.

The program runs a Monte Carlo simulator, so the output will be pseudo-random. Thus, obviously, the output will not exactly match the results in the paper.

The pyfp library provides functions to generate, embed, and attack digital fingerprinting systems. It has been created for research on the topic and is still underdocumented. Please ask for advice if you need it.