Title :
An effective evolutionary strategy for bijective S-boxes
Author :
Chen, Hua ; Feng, Deng-Guo
Author_Institution :
State Key Lab of Inf. Security, Inst. of Software Chinese Acad. of Sci., Beijing, China
Abstract :
Being as unique nonlinear components of block ciphers, S-boxes control the security of the cryptographic algorithms. The design of S-boxes with genetic algorithms is a recent research focus. For the popular bijective S-boxes, an effective evolutionary strategy is given in this paper, including fitness function, breeding strategy and hill climbing algorithm. Under this strategy, an effective genetic algorithm for 8 × 8 bijective S-boxes is provided and a large number of S-boxes with high nonlinearity and low difference uniformity can be obtained.
Keywords :
cryptography; genetic algorithms; bijective S-boxes; block ciphers; breeding strategy; cryptographic algorithms; evolutionary strategy; fitness function; genetic algorithms; hill climbing algorithm; nonlinear components; security control; Algorithm design and analysis; Boolean functions; Chaos; Cryptography; Galois fields; Genetic algorithms; Information security; Linearity; Testing;
Conference_Titel :
Evolutionary Computation, 2004. CEC2004. Congress on
Print_ISBN :
0-7803-8515-2
DOI :
10.1109/CEC.2004.1331158