Colluder Detection for Minimum Collusion Attacks

We investigate the problem of colluder identification for digital fingerprinting systems under the minimum collusion attack. Formulating the colluder detection as a binary hypothesis testing problem, we derive the log-likelihood ratio test. Utilizing the approximate distribution of the extreme order statistics, we obtain a low-complexity detector with an intuitively appealing form. Simulations show that the proposed detectors achieve significant performance improvements over the existing correlation-based detectors