Towards joint decoding of binary Tardos fingerprinting codes

TL;DR

This paper introduces a practical iterative joint decoder for binary Tardos fingerprinting codes, leveraging side-information and universal linear decoding, demonstrating strong performance in large user scenarios.

Contribution

It presents the first practical implementation of a joint decoder for probabilistic fingerprinting codes that does not rely on specific collusion assumptions.

Findings

Demonstrates improved decoding performance over previous methods

Supports large user bases with robust accuracy

Easily incorporates soft watermarking outputs

Abstract

The class of joint decoder of probabilistic fingerprinting codes is of utmost importance in theoretical papers to establish the concept of fingerprint capacity. However, no implementation supporting a large user base is known to date. This article presents an iterative decoder which is, as far as we are aware of, the first practical attempt towards joint decoding. The discriminative feature of the scores benefits on one hand from the side-information of previously accused users, and on the other hand, from recently introduced universal linear decoders for compound channels. Neither the code construction nor the decoder make precise assumptions about the collusion (size or strategy). The extension to incorporate soft outputs from the watermarking layer is straightforward. An extensive experimental work benchmarks the very good performance and offers a clear comparison with previous state-of-the-art decoders.