International Conference on Software Technology and Engineering (ICSTE 2012)
13 Theoretical Analysis on Robustness of Spread Spectrum Audio Watermarking
Download citation file:
- Ris (Zotero)
- Reference Manager
Robustness is one of the most important performances of audio watermarking algorithm. To analyze the robustness performance of direct sequence spread spectrum (DSSS) based audio watermarking algorithm, the bit error rate (BER) formulas of detected watermarking under original audio and additive white Gaussian noise (AWGN) interferences are deduced. According to the statistical model of DSSS-based audio watermarking system, which is established by adopting the statistical theory of signal detection, the robustness is in direct proportion to the length of spreading sequence and embedding power of watermark, and in inverse proportion to the variances of the transform coefficients of original audio and channel noise. What’s more, the communications model of DSSS-based audio watermarking system is designed according to the spread spectrum communications theory, and the BER formula indicates that it’s decided by the watermark power and processing gain. These expressions are helpful for evaluation of robustness performance and selection of optimum parameters.