Digital watermarking methods applied to non-additive channels

In the design of a data hiding method, one of the basic requirement that has to be accounted for is robustness. Thus, a great effort has been spent by the data hiding community to develop algorithms to achieve robustness against the modifications that watermarked contents can undergo during their life cycle. In this thesis we have studied side informed data hiding methods in case of more complex non-additive channels, such as the power-law channel and on the linear time invariant filtering channel, that model most of the processings commonly applied to media contents.