CONSTRUCTION OF BIJECTIVE S BOX USING GENETIC ALGORITHM

Abstract

The cryptographic strength of the S-Boxes' resistance to existing cryptanalytic attacks deter- mines the block cipher's security. S-Boxes' nonlinearity is used to determine how resistant they are to linear approximation and di erential assaults. S-Boxes with a high nonlinearity are regarded to be cryptographically secure. Finding cryptographically strong S-Boxes with consid- erable nonlinearity gets computationally di cult as the size of S-Boxes grows. The GENETIC ALGORITHM is suggested here to recognise a function by associating a nite stream of input with an equal length stream output.The overview and idea of this thesis is to exploit the concept of Genetic Procedure for our and try to device an average soln which is better than worst one in terms of non linearity and the solution space is large than the best ones.For the demonstration purpose ,8X8 s-box is being created by incorporating the concept of non linearity as selection parameter and Genetic algorithm to increase the potential soln space for selection of bijective vectorial function and the nature /Behavior of solns revolves around the parents.