S-box construction algorithm based on neural networks and genetic algorithms

To address the challenge of high computational complexity in determining the cryptographic characteristics of S-boxes created through genetic algorithms, a novel method for S-box design is introduced that incorporates neural network algorithms. Initially, S-boxes exhibiting lower differential uniformity and greater nonlinearity are identified using neural network algorithms. Following this, a highly efficient algorithm for creating S-boxes with superior cryptographic properties is developed by merging it with genetic algorithms. Experimental results indicate that the integration of neural networks greatly enhances the efficiency of S-box construction utilizing genetic algorithms.