Abstract
Introducing the memcapacitor into the Cellular Neural Network (CNN), the Memcapacitor-Cellular Neural Network (MC-CNN) model with infinitely many equilibrium points is constructed. A series of dynamical behaviors of the MC-CNN are investigated by various nonlinear system analysis means. It is shown that the system has a large maximum Lyapunov exponent in a specific parameter range. And with the variation of parameters, the system is able to produce many different phase trajectories of the attractor. Multistability is also found in the system. The pseudo-randomness of the MC-CNN is calculated by Spectral Entropy (SE) complexity algorithm. The final hardware results proves the physical realizability of the system. The MC-CNN model is intended to provide guidance for neural networks and cryptographic strategies based on the memcapacitor.
