Continuous media fingerprinting against time-varying collusion attacks

We analyze time-varying collusion attacks over continuous media such as music and movies. Inspired by communication systems, we propose countermeasures for time-varying collusion attacks to improve the performance of the proposed Multi-Carrier Code Division Multiple Access (MC-CDMA)-based fingerprinting system. In this paper, we focus on audio signals as continuous media. We first characterize the audio signals using the just-noticeable-difference (JND) of the human perceptual system model, called the human auditory system (HSA), at the fingerprint embedding, and provide the overall system model of our fingerprinting system as a framework for the analysis of time-varying collusion attacks. Then, we analyze the characteristics of time-varying collusion attacks in terms of their interference to and dynamics on audio. We provide advanced detectors as a countermeasure of interference and interleaving/repetition coding as a countermeasure of variation to improve the performance of the MC-CDMA-based fingerprinting system against time-varying collusion attacks.