Scholarlink

<p>In this paper, a novel memristor-driven dual-neuron chaotic system based on periodic perturbation is proposed to address the security challenges in multi-image transmission for the Internet of Things. The proposed chaotic system integrates memristive coupling and periodic perturbation into a dual-neuron model. It generates deterministic pseudo-random sequences with high initial value sensitivity and excellent statistical properties, making it suitable for image encryption. Its nonlinear dynamic behaviors are analyzed via phase trajectories, Lyapunov exponent spectra, attractor basins, bifurcation diagrams, and spectral entropy, confirming the effective chaotic intervals appropriate for generating pseudo-random sequences. Based on these sequences, the proposed encryption framework integrates self-describing three-dimensional fusion, as well as 3D transform-domain operations including three-dimensional integer wavelet transform and dynamic triaxial Josephson permutation, which further enhance the diffusion capability in spatial orientations and improve resistance against statistical and structural attacks, while incorporating inter-channel permutation and dynamic DNA codon S-box diffusion. The scheme supports batch encryption of color images of different sizes, and the embedded redundant metadata eliminates the need for external decryption parameters. Simulation results show that the ciphertext information entropy is close to the ideal value, the adjacent pixel correlation is near zero, the NPCR and UACI values approach the theoretical expectations, and the scheme exhibits favorable robustness under noise and occlusion attacks. These results demonstrate that the proposed scheme provides an effective and secure approach for color multi-image encryption.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A novel dual-neuron memristor-driven system and its application in dynamic DNA codon S-boxes multi-image encryption

  • 期刊论文

  • 发表日期：

    2026年5月28日

查看全文
  • Li Xiong,
  • Huiyang Bian,
  • Silong Zhou,
  • Xinlei An,
  • Qigui Yang

摘要

In this paper, a novel memristor-driven dual-neuron chaotic system based on periodic perturbation is proposed to address the security challenges in multi-image transmission for the Internet of Things. The proposed chaotic system integrates memristive coupling and periodic perturbation into a dual-neuron model. It generates deterministic pseudo-random sequences with high initial value sensitivity and excellent statistical properties, making it suitable for image encryption. Its nonlinear dynamic behaviors are analyzed via phase trajectories, Lyapunov exponent spectra, attractor basins, bifurcation diagrams, and spectral entropy, confirming the effective chaotic intervals appropriate for generating pseudo-random sequences. Based on these sequences, the proposed encryption framework integrates self-describing three-dimensional fusion, as well as 3D transform-domain operations including three-dimensional integer wavelet transform and dynamic triaxial Josephson permutation, which further enhance the diffusion capability in spatial orientations and improve resistance against statistical and structural attacks, while incorporating inter-channel permutation and dynamic DNA codon S-box diffusion. The scheme supports batch encryption of color images of different sizes, and the embedded redundant metadata eliminates the need for external decryption parameters. Simulation results show that the ciphertext information entropy is close to the ideal value, the adjacent pixel correlation is near zero, the NPCR and UACI values approach the theoretical expectations, and the scheme exhibits favorable robustness under noise and occlusion attacks. These results demonstrate that the proposed scheme provides an effective and secure approach for color multi-image encryption.