On the Robustness of the Delay-Based Fingerprint Embedding Scheme

TL;DR

This paper evaluates the robustness of delay-based fingerprint embedding against desynchronization attacks and proposes methods to enhance synchronization, including adding a synchronization fingerprint and using relative delay, supported by analysis and experiments.

Contribution

It introduces two novel techniques to improve robustness of delay-based fingerprinting against desynchronization attacks, which were not previously addressed.

Findings

Adding synchronization fingerprint improves robustness.

Using relative delay enhances synchronization accuracy.

Proposed methods significantly increase attack resistance.

Abstract

The delay-based fingerprint embedding was recently proposed to support more users in secure media distribution scenario. In this embedding scheme, some users are assigned the same fingerprint code with only different embedding delay. The algorithm's robustness against collusion attacks is investigated. However, its robustness against common desynchronization attacks, e.g., cropping and time shifting, is not considered. In this paper, desynchronization attacks are used to break the delay-based fingerprint embedding algorithm. To improve the robustness, two means are proposed to keep the embedded fingerprint codes synchronized, i.e., adding a synchronization fingerprint and adopting the relative delay to detect users. Analyses and experiments are given to show the improvements.