Design and circuit implementation of a novel 5D memristive CNN hyperchaotic system

Chunbo Xiu,Jingyao Fang,Yuxia Liu
DOI: https://doi.org/10.1016/j.chaos.2022.112040
2022-05-01
Chaos, Solitons and Fractals
Abstract:In order to enrich the dynamic characteristics of cellular neural network (CNN) and reveal the influence of memristor nonlinearity on the dynamic behavior of the network, a novel five-dimension memristive cellular neural network hyperchaotic system is designed by replacing the linear resistors in the output module of cellular neural network with two flux-controlled memristors, and the hardware circuit design of the system is completed. Based on Lyapunov exponent spectrum and attractor phase trajectories, the effects of system parameters and initial value on the dynamic characteristics of the memristive cellular neural network model are studied, and the generation conditions of different chaotic attractors and coexisting chaotic attractors are explained. Kolmogorov entropy is used to measure the chaotic degree of the system when the parameters are in different intervals. In this way, the criteria of parameter selection for system application are given. In particular, the effect of Gaussian white noise on the dynamic behavior of memristive CNN chaotic system is studied. The hardware circuit design and characteristic analysis of memristive cellular neural network are completed by the circuit simulation software, and the physical realizability of chaotic characteristics of the memristive cellular neural network model is verified. Furthermore, a secure communication application example based on the hyperchaotic system is given.
What problem does this paper attempt to address?