Control of chaotic neural networks based on contraction mappings

The control chaos method based on contraction mappings is applied to small discrete neural networks for stabilizing a desired periodic orbit embedded in the chaotic attractor by a external input. The suggested method utilizes only an approximate location of the desired periodic orbit, one of the merits of this method is that the linearization of the system near the stabilized orbit is not required. Typical examples of two- and three-dimensional neural networks are given, and the validity of the method is shown by the numerical simulation.