Self-synchronized audio watermark in DWT domain

Robust audio watermarks present some challenges. In this paper, we propose a DWT-based audio watermarking algorithm to resist shifting and cropping. Synchronization codes are embedded into audio with informative watermark, and thus the watermark generated has the ability of self-synchronization. To achieve good robustness performance, we embed the synchronization codes and the watermark into the low-frequency coefficients in DWT domain. By exploring the time-frequency localization capability of DWT to decrease the computation load for searching synchronization codes effectively, we resolve the conflict between robustness and computation complexity of synchronization code. In addition, we analyze the performance of the proposed algorithm in terms of data payload, error probability of synchronization codes, SNR and BER. The experimental results show the watermark is robust against common signal processing manipulations and attacks, such as Gaussian noise corruption, requantization, cropping and MP3 compression.