Robust Audio Watermarking Scheme Based on Short Time Fourier Transformation and Singular Value Decomposition

Digital watermarking has been widely used for copyright protection of digital contents. This paper proposes a new watermarking scheme based on short time Fourier transformation and singular value decomposition. In our proposed watermarking scheme, spectrum of the original audio signal is computed first by short-time Fourier transformation (STFT). Prominent spectral peaks are detected from the frequency spectrum of each frame using a peak detection algorithm. Singular value decomposition (SVD) is applied to the selected prominent peaks of each frame represented in matrix form. Watermarks are then embedded into the highest singular value of each matrix of the selected prominent peaks of each frame. Watermarks are extracted by performing the inverse operation of watermark embedding process. Simulation results indicate that the proposed watermarking scheme is highly robust against various kinds of attacks such as noise addition, cropping, re-sampling, re-quantization, MP3 compression, and achieves similarity values ranging from 28 to 32. In addition, our proposed scheme achieves signal-to-noise ratio (SNR) values ranging from 20 to 31 dB.