SBGen: A high-performance library for rapid generation of cryptographic S-boxes

Abstract

In the realm of cryptographic research, the generation of S-boxes with high nonlinearity and optimal cryptographic properties remains a critical challenge. This paper presents a novel approach to S-box generation, leveraging the strengths of heuristic optimization methods. Through a meticulous integration of Simulated Annealing (SA) and Hill Climbing (HC) algorithms with sophisticated cost functions, we introduce an innovative software tool that significantly advances the efficiency of generating highly nonlinear S-boxes. Our methodology is distinguished by its ability to consistently produce S-boxes that meet stringent security criteria, with a remarkable 100 % success rate and minimized computational overhead. A comparative analysis reveals that our approach outperforms existing methods in terms of the probability of generating target S-boxes and the average number of iterations required. The software's practical implications extend beyond theoretical advancements, offering a valuable resource for cryptographic system designers in their quest to fortify cipher systems against linear and differential cryptanalysis. By setting new benchmarks for nonlinearity and search efficiency, our work paves the way for future research in cryptographic S-box generation, highlighting the potential of combining heuristic techniques with domain-specific cost functions to achieve superior security outcomes.