Image encryption based on feistel network for the new multiscroll chaotic hyperjerk system

This study introduces a novel multi-scroll chaotic hyperjerk system characterized by multistability and enhanced dynamics through initial condition-based offset boosting control. The proposed system generates multiple coexisting attractors, enabling flexible and controllable chaotic behavior suitable for secure applications. Analytical and numerical analyses confirm the chaotic and multistable properties of the system. To demonstrate practical applicability, an image encryption scheme based on the Feistel network and the proposed chaotic system is developed. The encryption algorithm incorporates permutation and diffusion mechanisms to enhance confusion and diffusion properties. Experimental results demonstrate that the proposed scheme achieves strong randomness, effective pixel decorrelation, and high resistance against statistical and differential attacks while maintaining computational efficiency. These findings confirm the suitability of the proposed system for secure image encryption and chaos-based communication applications.