摘要 |
PROBLEM TO BE SOLVED: To solve such problem of a conventional discrete-time type cellular neural network (DT-CNN) method that S/N is no good, and a stable conversion to an equilibrium point cannot be ensured because the gain adjustment of a sample point is performed after an unequally spaced sample sequence is replaced by an equally spaced sample sequence as it is. SOLUTION: In order to constitute the DT-CNN having a desired characteristic, DT-CNNs in which samples of an input sample sequence and an output sample sequence are equally spaced with the same interval, and templates are symmetric are constituted, and a DT-CNN in which the samples of the output sample sequence are unequally spaced is constituted on the basis of the above DT-CNNs. |