Design and implementation of block cipher using neural network

Abstract EN

The last decade witnessed a great evolution on the fields of computer science, artificial intelligence, communication and data transmission.

This evolution draw the attention of specialists to design a modern cryptosystem for data encryption based on neural networks methodologies that are very hard, if not impossible, to be broken.

This paper employs cryptography scheme utilizes the neural networks in block cipher algorithms.

Taking into consideration different plaintexts block size.

The block frequency test has been show the successful of ciphertexts with difficulty to break the cipher process.

The application of the neural network in cryptography provides fast and high security system in block ciphering in comparison with traditional methods.

The C++ language is used for designing the programs of neural network to encryption and decryption algorithms.