Author : Dhia Kadhm Al-Shakarchy, Noor
journal of kerbala university,
Volume 8, Issue 2, Pages 13-21
Data Encryption Standard (DES) algorithm considers one of complicated algorithms that have users confidence for a long time. This algorithm enjoy with wide spread in business progress, banking, and governmental.
When DES algorithm entered running space, and many experiments and researches in cryptanalysis continuous to break this algorithm. In this research we using artificial neural networks to attack this algorithm by designing artificial neural network system simulating the main design of DES algorithm. The main idea of proposed system depends on represented the plaintext / ciphertext process; so that(Expansion permutation, and S-boxes substitution, and P-boxes permutation) process represented in proposed artificial neural network model; because the key is unknown and the purpose of proposed system obtained it. The proposed system used in two ways as cryptanalysis by provide a ciphertext as input to the network and the output obtained from the network system is the plaintext. The second used to the network is as cryptography system that’s by inputs plaintext as input to the network system and the output obtained from the network is ciphertext.