Abstract | ||
---|---|---|
Neural network cryptography is an interesting area of research in the field of computer science. This paper proposes a new model to encrypt/decrypt a secret code using Neural Networks unlike previous private key cryptography model that are based on theoretic number functions. In the first part of the paper, we propose our model and analyze the privacy and security of the model thereby explaining why an attacker with a similar neural network model is unlikely to pose a threat to the system. This proves that the model is pretty secure. In the second part of the paper, we experiment with the neural network model using two different ciphertexts of different length. Parameters of the network that are tested are different learning rates, optimizers and step values. The experimental results show how to enhance the accuracy of our model even further. Furthermore, our proposed model is more efficient and accurate compared to other models for encryption and decryption. |
Year | DOI | Venue |
---|---|---|
2017 | 10.1007/978-3-319-76351-4_33 | HYBRID INTELLIGENT SYSTEMS, HIS 2017 |
Keywords | Field | DocType |
Encryption,Decryption,Neural networks | Cryptography,Computer science,Neural cryptography,Encryption,Theoretical computer science,Artificial neural network,Public-key cryptography | Conference |
Volume | ISSN | Citations |
734 | 2194-5357 | 0 |
PageRank | References | Authors |
0.34 | 6 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Sayantica Pattanayak | 1 | 13 | 0.90 |
Simone A Ludwig | 2 | 1309 | 179.41 |