A large class of multimedia and biomedical signals can be modeled as Autotegessive （AR） random processes. Pefformance of watermarking embedding algorithms utilizing this host model is still left unexplored. The auth...A large class of multimedia and biomedical signals can be modeled as Autotegessive （AR） random processes. Pefformance of watermarking embedding algorithms utilizing this host model is still left unexplored. The authors investigate the decoding perform-nance of Spread Spectrum （SS） embedding algorithm in the standard Additive White Gaussian Noise （AWGN） channel with the host signal being modeled as AR process. The SS embedding algorithm also use linear interference cancelation in the subspace spanned by watermark pattern. They study the influence of design parameters on the decoding performance. The analytic result is verified by Monte Carlo simulation on synthesized AR process. The result may be helpful to design watermarking system for speech, biomedical and image signals.展开更多
基金supported by research project of“SUSTSpring Bud”:the research on decoder under desynchronization attack for data hiding systems
文摘A large class of multimedia and biomedical signals can be modeled as Autotegessive （AR） random processes. Pefformance of watermarking embedding algorithms utilizing this host model is still left unexplored. The authors investigate the decoding perform-nance of Spread Spectrum （SS） embedding algorithm in the standard Additive White Gaussian Noise （AWGN） channel with the host signal being modeled as AR process. The SS embedding algorithm also use linear interference cancelation in the subspace spanned by watermark pattern. They study the influence of design parameters on the decoding performance. The analytic result is verified by Monte Carlo simulation on synthesized AR process. The result may be helpful to design watermarking system for speech, biomedical and image signals.
