Please use this identifier to cite or link to this item: http://hdl.handle.net/10609/81610
Title: Redes neuronales aplicadas al criptoanálisis del Advanced Encryption Standard
Author: Novas Otero, Pedro
Tutor: Isern, David  
Others: Universitat Oberta de Catalunya
Ventura, Carles  
Abstract: This study will explore the application of neural networks to the cryptanalysis of the Advanced Encryption Standard algorithm. The proposed cryptanalysis consists of training a neural network with examples comprised of pairs of plaintexts and their respective cyphertexts, so that it can subsequently obtain the correct cyphertext having access only to the plaintext. The study shows learning capacity, albeit reduced, when the neural networks are tested against a reduced-round version of the algorithm. However, when the full-round version of the algorithm is used, the tests show randomness. Therefore, according to the results obtained, the proposed cryptanalysis cannot be considered an effective attack against the algorithm studied.
Keywords: convolutional neural network
criptanalysis
AES
Document type: info:eu-repo/semantics/bachelorThesis
Issue Date: 5-Jun-2018
Publication license: http://creativecommons.org/licenses/by-nc-nd/3.0/es/  
Appears in Collections:Bachelor thesis, research projects, etc.

Files in This Item:
File Description SizeFormat 
pnovasTFG0618memoria.pdfMemoria del TFG3,27 MBAdobe PDFThumbnail
View/Open
pnovasTFG0618presentación.pdfPresentación del TFG1,08 MBAdobe PDFThumbnail
View/Open