Hidden coexisting hyperchaos of new memristive neuron model and its application in image encryption

The neuron models have been widely applied to neuromorphic computing systems and chaotic circuits. How-ever, discrete neuron models and their application in image encryption have not gotten a lot of attention yet. This paper first presents a novel neuron model with significant chaotic characteristics, by coupling a memristor into the proposed neuron, a memristive neuron model is further obtained. Relevant control parameter-relied dy-namical evolution is demonstrated using several numerical methods. The explorations manifest that memristor can boost chaos complexity of the discrete neuron, resulting in hyperchaos, infinite coexisting hidden attractors and attractor growing. Particularly, the NIST test verifies the generated hyperchaotic sequences exhibit high com-plexity, which makes them applicable to many applications based on chaos. Additionally, digital experiments based on developed hardware platform are designed to implement the memristive neuron model and get the hyperchaos. We also propose a new encryption scheme to apply the memristive neuron to the application of image encryption. The evaluation results show that the conceived algorithm appears excellent security charac-teristics and can effectively protect the information security of images. (c) 2021 Elsevier Ltd. All rights reserved.

Automated summary

This paper presents a novel neuron model by coupling a memristor to obtain a memristive neuron model. The study shows that memristor can enhance the chaos complexity of the discrete neuron, resulting in hyperchaos. Additionally, a new encryption scheme for image encryption is proposed, which exhibits excellent security characteristics.