Results on the spatio-temporal dynamics of a first order cell CNN

Cellular Neural Networks (CNN's) type analog parallel architectures have been and are studied both for their potential in high-speed signal processing applications and for their interesting spatio-temporal dynamics. Basically, such architectures consist of arrays of identical linear or nonlinear cells identically coupled by means of cloning templates. It is known that such architectures can exhibit spatio-temporal filtering and pattern formation. The dynamics of the class of CNN's discussed in this paper has been so far studied mainly for symmetric templates. Here we study the CNN behavior for non-symmetric templates of first and second order with transistor level simulations. The roles of the cell structure and the connection template are discussed and models for the spatial modes dynamics are presented.