Cryptography System Implementation Using Neurocomputational Model and Deep Learning
Valencia Ramos, Rafael Alejandro · 2020
The management and protection of information were always topics of interest for society and of great importance in the current digital age. The vast majority of people around the world have handled important and sensitive information through various platforms ranging from bank accounts to social media. All platforms have ensured that the information flowing through them is kept safe from malicious attacks. This has given rise to a vast and important field in computing called cryptography. There are several cryptographic algorithms that allow information to remain secure, and these have been divided into two groups, symmetric key and asymmetric key algorithms. The operation of asymmetric key algorithms is based on the manipulation of very large prime numbers, which provides a high level of security, but also implies a high computational time. This work proposes a cryptographic system based on artificial neural networks, implemented through the use of deep learning techniques. The method used the synaptic weights of an autoencoder neural network as encryption and decryption keys, avoiding the use of large prime numbers. The proposed solution allowed the initial and final synaptic weights to have a high level of randomness, without affecting the overall performance of the network. The theoretical security analysis indicated that the proposed methodology was robust and difficult to break. The experimental results confirmed that the proposed system performs the encryption and decryption of data in a low computational time, with respect to traditional algorithms such as RSA, ElGamal, ECC and Paillier.