CNN-based Chaotic System Design And It’s Application Research

With the in-depth research of chaotic phenomena and bifurcation theory, researching on cellularneural networks bifurcation theory and application has increasingly become a hotpot. Based on thestructure of CNN, the deeply study had been carried on separately in theory and industry applicationdomain in this paper, the main contents are as follows:1） Based on the cell structure of the neural network, two kinds of chaotic model-hysteresischaotic system and MLC chaotic system are brought up. Due to the difficulty to directly writing thehysteresis function, this paper carries out a way of building model through SIMLINK modules inMATLAB. This article does not use the way of realzing the differential equation in traditionalliterature, integral part of, but directly calls state module in SIMLINK and achieving the constructionof the entire system via the module parameter settings.2） Based on the simple chaotic systems in (1), building improved chaotic or hyperchaoticsystem.We do the eigenvalue analysis (Lyapunov exponent, power spectra, Poincare Figure) to allchaotic systems, including chaotic attractors and new chaotic system3） Study the engineering application on the image encryption filed for chaotic system. Asoftware platform has been implemented with the chaotic system designed in this paper,,which can beapplied in image encryption/decryption field. It was verified that the encryption/decryption result isperfect, which can lead to the practice use of this encryption system.