<p>This paper proposes a novel hybrid image encryption scheme that combines a memristor-based hyperchaotic system, the Trivium stream cipher, DNA-inspired operations, and a lightweight 3DES layer into a tightly integrated multi-layer architecture. The memristor-based chaotic system ensures a large key space and complex dynamics, while the Trivium cipher enhances randomness, and the DNA operations introduce strong confusion and diffusion properties. The addition of 3DES further reinforces resistance against brute-force and structural attacks. Extensive experiments demonstrate the effectiveness of the proposed algorithm: the information entropy of encrypted images approaches the ideal value (7.999), the correlation coefficients between adjacent pixels are reduced to nearly zero (≈&#xa0;0.001), and differential attack metrics achieve NPCR values above 99.58% and UACI values above 33%. Furthermore, the scheme exhibits robustness against data loss and noise (up to 20% salt-and-pepper noise) with minimal degradation in decrypted images. Hardware implementation on the NVIDIA Jetson TX2 confirms the practicality of the method for embedded and edge-computing environments. These results indicate that the proposed scheme provides a secure, efficient, and hardware-friendly solution for modern image encryption applications.</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 memristor-based hyperchaotic hybrid encryption system with DNA for image encryption on the Jetson TX2

  • Hasan Ulutas

摘要

This paper proposes a novel hybrid image encryption scheme that combines a memristor-based hyperchaotic system, the Trivium stream cipher, DNA-inspired operations, and a lightweight 3DES layer into a tightly integrated multi-layer architecture. The memristor-based chaotic system ensures a large key space and complex dynamics, while the Trivium cipher enhances randomness, and the DNA operations introduce strong confusion and diffusion properties. The addition of 3DES further reinforces resistance against brute-force and structural attacks. Extensive experiments demonstrate the effectiveness of the proposed algorithm: the information entropy of encrypted images approaches the ideal value (7.999), the correlation coefficients between adjacent pixels are reduced to nearly zero (≈ 0.001), and differential attack metrics achieve NPCR values above 99.58% and UACI values above 33%. Furthermore, the scheme exhibits robustness against data loss and noise (up to 20% salt-and-pepper noise) with minimal degradation in decrypted images. Hardware implementation on the NVIDIA Jetson TX2 confirms the practicality of the method for embedded and edge-computing environments. These results indicate that the proposed scheme provides a secure, efficient, and hardware-friendly solution for modern image encryption applications.