A novel 8D hyperchaotic framework for robust and key-dependent substitution box design
摘要
In response to the escalating frequency of sophisticated hacking and phishing attacks, robust data encryption and information security have become paramount. Modern adversaries increasingly target vulnerabilities within cryptographic systems to compromise sensitive data. As a critical component of symmetric-key block ciphers, the S-Box is essential for introducing confusion and obfuscating the relationship between plaintext and ciphertext. This study proposes a novel method for generating key-dependent, dynamic S-Boxes leveraging a high-dimensional 8D hyperchaotic map. The cryptographic integrity of the resulting S-Box is rigorously validated through extensive statistical analyses, including nonlinearity, strict avalanche criterion (SAC), bit independence criterion (BIC), and differential/linear probability tests. Comparative evaluations reveal that the proposed S-Box achieves an average nonlinearity of 111.75, an ideal SAC value of 0.5, BIC–NL (103.4), BIC–SAC (0.5040), LP (0.125), and DP (0.0391). Absence of any fixed point and opposite fixed point is another trait of the proposed S-Box. Similarly, the standard side-channel resistance measures with transparency order value of 7.846, signal-to-noise-ratio of 9.178, and a confusion coefficient variance of 0.127 suggest that the resistance to side-channel leakage of power is very effective. These results demonstrate that the proposed design is at par with the current state-of-the-art S-Boxes, highlighting its significant potential for practical implementation in high-security environments. To evaluate its practical applicability, the proposed S-box is incorporated into a new color image encryption scheme. Standard evaluation metrics, including entropy, correlation coefficient, PSNR, NPCR, and UACI have values 7.999, 0.0008, 8.9436, 99.63, and 33.42 respectively. These results demonstrate that the proposed image encryption scheme achieves strong security characteristics and effective resistance against statistical and differential attacks.