Proposing two quantum audio watermarking schemes with improved robustness

Abstract- As an important security technology, recently quantum watermarking attracted wide research attention. Up to now, several methods have been proposed for quantum image watermarking, while there are a few achievements in the domain of quantum audio watermarking. This study presents two quantum audio watermarking schemes, with the aim of improving robustness. The first scheme embeds a watermark qubit into odd number of qubits of host audio, and employs majority-voting policy to extract the correct watermark qubit. In the second proposed scheme, k audio samples are grouped as a frame, which holds one qubit of watermark. In order to embed a qubit into a frame, parameter r, which is sum of frame amplitudes, is calculated in module 2k. The amplitudes are then adjusted to set parameter r in the median of two ranges, [0, 2k-1-1], and [2k-1, 2k-1] to represent embedding ├ |0⟩ or ├ |1⟩. The parameter r is recalculated in extracting phase, and based on belonging to a range, the extracted qubit is determined. For every procedure of the proposed schemes, the quantum circuit and complexity analysis are presented. The circuit complexity of both proposed schemes is linear. Experimental results show that the proposed schemes offer promising trade-offs in terms of robustness and transparency.