Wavelet-Based Hybrid Audio Watermarking Using Statistical Mean Manipulation and Spread Spectrum

In this paper, we propose a Discrete Wavelet Transform (DWT) - based hybrid audio watermarking using Statistical Mean Manipulation (SMM) and Spread Spectrum (SS) technique. The host audio is decomposed by DWT to produce the signal in low frequency subband and high frequency subband, where SMM embeds the watermark into low frequency in the first subband, and SS embeds the watermark into high frequency in the selected subband. The embedding process using the SMM technique is the insertion process by modifying the average of the audio signal in one frame according to the watermark, thus it modifies the audio in the low-frequency subband. The SS technique modulates the watermark before it is embedded into the host audio in the higher selected frequency subband. This combination technique produces the robust watermarking method to the signal processing attack, such as Low Pas Filter (LPF), resampling and audio compression while maintaining high watermarked audio quality and watermark payload.