Five-dimensional memristive Hopfield neural network dynamics analysis and its application in secure communication

Purpose This study aims to improve the complexity of chaotic systems and the security accuracy of information encrypted transmission. Applying five-dimensional memristive Hopfield neural network (5D-HNN) to secure communication will greatly improve the confidentiality of signal transmission and grea...

Full description

Saved in:
Bibliographic Details
Published in:Circuit world 2024-01, Vol.50 (1), p.67-81
Main Authors: Yin, Xinxing, Chen, Juan, Yu, Wenxin, Huang, Yuan, Wei, Wenxiang, Xiang, Xinjie, Yan, Hao
Format: Article
Language:English
Subjects:
Behavior
Chaos theory
Communication
Complexity
Control methods
Convergence
Coupling
Design
Dynamic characteristics
Feedback control
Liapunov exponents
Masking
Neural networks
Neurons
Signal transmission
Stability analysis
Synchronism
Time synchronization
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Purpose This study aims to improve the complexity of chaotic systems and the security accuracy of information encrypted transmission. Applying five-dimensional memristive Hopfield neural network (5D-HNN) to secure communication will greatly improve the confidentiality of signal transmission and greatly enhance the anticracking ability of the system. Design/methodology/approach Chaos masking: Chaos masking is the process of superimposing a message signal directly into a chaotic signal and masking the signal using the randomness of the chaotic output. Synchronous coupling: The coupled synchronization method first replicates the drive system to get the response system, and then adds the appropriate coupling term between the drive The synchronization error and the coupling term of the system will eventually converge to zero with time. The synchronization error and coupling term of the system will eventually converge to zero over time. Findings A 5D memristive neural network is obtained based on the original four-dimensional memristive neural network through the feedback control method. The system has five equations and contains infinite balance points. Compared with other systems, the 5D-HNN has rich dynamic behaviors, and the most unique feature is that it has multistable characteristics. First, its dissipation property, equilibrium point stability, bifurcation graph and Lyapunov exponent spectrum are analyzed to verify its chaotic state, and the system characteristics are more complex. Different dynamic characteristics can be obtained by adjusting the parameter k. Originality/value A new 5D memristive HNN is proposed and used in the secure communication
ISSN:0305-6120
1758-602X
0305-6120
DOI:10.1108/CW-05-2022-0135