A DWT-Based Audio Watermarking Algorithm Using Wavelet Component Proportional Coefficients for Balanced Robustness and Inaudibility

Abstract

Research on audio watermarking has concentrated predominantly on robustness, yet most existing algorithms fail to sustain high robustness under conditions of complete inaudibility, and reconciling this trade-off between imperceptibility and attack resilience remains a central challenge in the field. To address this problem, this paper proposes a digital watermarking algorithm based on wavelet component proportional coefficients, designed explicitly to balance robustness and inaudibility. This paper proposes a digital watermarking algorithm based on wavelet component proportional coefficients. First, the algorithm computes the wavelet coefficients by Discrete Wavelet Transformation (DWT), then gets the wavelet coefficients proportion of the two sub-segments of every segments in each frame, then produces the wavelet coefficients of all sub-segments by two modular addition; finally, the watermark is embedded by inverse DWT based on the wavelet coefficients scaled by the coefficient proportions before and after the process. The simulation results show that the proposed algorithm can provides the good inaudibility and robustness to conventional signal processing, such as MP3 compression, low-pass filtering, resampling, etc.