Design and Evaluation of Two Symmetrical Evolutionist-Based Ciphering Algorithms

In this paper, we present two symmetric evolutionist-based ciphering algorithms (SEC and SEC-EX) where the second is an extension of the first. Our evolutionist approach is modeled as a combinatorial optimization problem. The paper addresses the coding of the chromosomes, the identification of a fitness function, and the choice of the adapted genetic operators. In SEC we treat directly the characters of plaintext. While in SEC-EX we develop a new technique, which bars the way in front of all the statistical analysis attacks. In fact, instead of treating characters we treat the binary blocks constituting the plaintext coded in binary. Both of the algorithms generate a resistant session encryption key. The performance of the two algorithms has been evaluated using several experiments. We have also compared our algorithms to the well-known ciphering algorithms such as DES, IDEA, and RSA. The experimental results show that our algorithms have the fastest deciphering time and an average ciphering.