Dynamical Behavior of Three-Order Cellular Neural Network with Application in Image Secure Communication

The paper analyzes dynamic behaviors of three- order Cellular Neural Network (3-order CNN). It shows that 3-order CNN is symmetric with regard to the origin; the system is dissipative and asymptotic motion settles onto an attractor. The system has diverse chaotic attractors with different parameters. So 3-order CNN can be applied in secure communications owing to its chaotic behaviors. The paper combines 3-order CNN with DES to propose a scheme of image secure communication. The results of security analysis indicate that this scheme not only has a large key space but also has a very sensitivity with respect to the initial conditions of 3-order CNN and the key of DES.