Switched State Controlled-CNN: An Alternative Approach in Generating Complex Systems with Multivariable Nonlinearities Using CNN

Abstract

It has been already shown in the literature that extended State Controlled-Cellular Neural Network (SC-CNN) could be used to approximate the behavior of complex dynamics with multivariable nonlinearities by modifying the nonlinearity. In this study, a different approach to imitating the multivariable nonlinearities in complex dynamical systems by switching the states of SC-CNN is given. For this purpose, first of all, Rossler type system dynamics are generated by switching the states of Cellular Neural Networks (CNNs). In the second study, Sprott-94-f type system is presented. All systems are numerically and experimentally investigated in order to verify the convenience of the standpoint.