A feedback mode with MN cells in cellular neural networks (CNN) is equivalent to a network which possesses 2MN cells, a neighborhood of mirror-like strucutre, space-variant templates and without feedback links as well as without input templates. The stability of the feedback mode and transformations of CNN with the neighborhood of mirror-like structure are discussed.
Chua L O, Yang L. Cellular neural networks: Theory and applications. IEEE Trans. on CAS, 1988, CAS-35(10): 1257-1290.[2]Special Issue on Cellular Neural Networks and Applications. IEEE Trans. on CAS, 1993, CAS-40(3): 145-224.[3]Special Issue on Cellular Neural Networks and Applications. Int[J].J. Circuit Theory Appl.1992, 20(5):449-637[4]Chua L O, Roska T. Stability of a class of nonreciprocal cellular neural networks. IEEE Trans. on CAS, 1990, CAS-37(12): 1520-1527.[5]Chua L O, Wu C W. On the universe of stable cellular neural networks[J].Int. J. Circuit Thery Appl.1992, 20(5):497-517[6]Savaci F A, Vandewalle J. On the stability analysis of cellular neural networks.IEEE Trans. on CAS, 1993, CAS-40(3): 213-215.[7]Harrer H.Multiple layer discrete-time cellular neural networks using time-variant templates. IEEE Trans.on CAS, 1993, CAS-40(3): 191-199.