In this paper, we propose an aperiodic stochastic resonance (ASR) signal processor for communication systems based on a bi-stable mechanism. This processor can detect base-band binary pulse amplitude modulation (PAM) signals. In this processor related parameters can be adjusted. The adjustment mechanism is explained from the perspective of the conventional noise-induced nonlinear signal processing. To demonstrate this processors usability, based on it we implemented a digital image watermarking algorithm in the discrete cosine transform (DCT) domain. In this algorithm, the watermark and the DCT alternating current (ac) coefficients of the image are viewed as the input signal and the channel noise, respectively. This phenomenon that the detection bit error ratio (BER) of the system suffering from certain attacks is lower than that of the system not suffering from any attack is systematically analyzed. © 2008 Springer Berlin Heidelberg.
|Title of host publication||Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)|
|Publication status||Published - 2008|
- Digital watermarking
- Signal processing
- Stochastic resonance