Analysis of Cost function using Genetic algorithm to construct balanced Boolean function
Pratap Kumar Behera, Sugata Gangopadhyay · 2018
The security of symmetric cryptosystem depends upon the cryptographic properties of Boolean function, e.g. balancedness, high nonlinearity, and low autocorrelation used as primitives in their designs. The problem of finding such Boolean functions satisfying multiple cryptographic properties is computationally hard, since the search space consisting of all n variable Boolean functions are 22n. The most common methods used for constructing Boolean functions are the random generation, algebraic construction and evolutionary techniques. In this paper, we use the Genetic algorithm to construct balanced Boolean function with Clark's cost function with high nonlinearity and low autocorrelation. Our main focus is to analyze the Clark's cost function with different values of tuning parameter using Genetic algorithm and compares the results obtained with the nonlinearity as a cost function.