Hybrid encryption based on a generative adversarial network

In today's world, encryption is crucial for protecting sensitive data. Neural networks can provide security against adversarial attacks, but meticulous training and vulnerability analysis are required to ensure their effectiveness. Hence, this research explores hybrid encryption based on a gene...

Full description

Bibliographic Details
Published in:Indonesian Journal of Electrical Engineering and Computer Science
Main Author: Amir I.; Suhaimi H.; Mohamad R.; Abdullah E.; Pu C.-H.
Format: Article
Language:English
Published: Institute of Advanced Engineering and Science 2024
Online Access:https://www.scopus.com/inward/record.uri?eid=2-s2.0-85195200213&doi=10.11591%2fijeecs.v35.i2.pp971-978&partnerID=40&md5=f466be2f9c4274c2b15f40b2641ffb96
id 2-s2.0-85195200213
spelling 2-s2.0-85195200213
Amir I.; Suhaimi H.; Mohamad R.; Abdullah E.; Pu C.-H.
Hybrid encryption based on a generative adversarial network
2024
Indonesian Journal of Electrical Engineering and Computer Science
35
2
10.11591/ijeecs.v35.i2.pp971-978
https://www.scopus.com/inward/record.uri?eid=2-s2.0-85195200213&doi=10.11591%2fijeecs.v35.i2.pp971-978&partnerID=40&md5=f466be2f9c4274c2b15f40b2641ffb96
In today's world, encryption is crucial for protecting sensitive data. Neural networks can provide security against adversarial attacks, but meticulous training and vulnerability analysis are required to ensure their effectiveness. Hence, this research explores hybrid encryption based on a generative adversarial network (GAN) for improved message encryption. A neural network was trained using the GAN method to defend against adversarial attacks. Various GAN training parameters were tested to identify the best model system, and various models were evaluated concerning their accuracy against different configurations. Neural network models were developed for Alice, Bob, and Eve using random datasets and encryption. The models were trained adversarially using the GAN to find optimal parameters, and their performance was analyzed by studying Bob's and Eve's accuracy and bits error. The parameters of 8,000 epochs, a batch size of 4,096, and a learning rate of 0.0008 resulted in 100% accuracy for Bob and 52.14% accuracy for Eve. This implies that Alice and Bob's neural network effectively secured the messages from Eve's neural network. The findings highlight the advantages of employing neural network-based encryption methods, providing valuable insights for advancing the field of secure communication and data protection. © 2024 Institute of Advanced Engineering and Science. All rights reserved.
Institute of Advanced Engineering and Science
25024752
English
Article
All Open Access; Gold Open Access
author Amir I.; Suhaimi H.; Mohamad R.; Abdullah E.; Pu C.-H.
spellingShingle Amir I.; Suhaimi H.; Mohamad R.; Abdullah E.; Pu C.-H.
Hybrid encryption based on a generative adversarial network
author_facet Amir I.; Suhaimi H.; Mohamad R.; Abdullah E.; Pu C.-H.
author_sort Amir I.; Suhaimi H.; Mohamad R.; Abdullah E.; Pu C.-H.
title Hybrid encryption based on a generative adversarial network
title_short Hybrid encryption based on a generative adversarial network
title_full Hybrid encryption based on a generative adversarial network
title_fullStr Hybrid encryption based on a generative adversarial network
title_full_unstemmed Hybrid encryption based on a generative adversarial network
title_sort Hybrid encryption based on a generative adversarial network
publishDate 2024
container_title Indonesian Journal of Electrical Engineering and Computer Science
container_volume 35
container_issue 2
doi_str_mv 10.11591/ijeecs.v35.i2.pp971-978
url https://www.scopus.com/inward/record.uri?eid=2-s2.0-85195200213&doi=10.11591%2fijeecs.v35.i2.pp971-978&partnerID=40&md5=f466be2f9c4274c2b15f40b2641ffb96
description In today's world, encryption is crucial for protecting sensitive data. Neural networks can provide security against adversarial attacks, but meticulous training and vulnerability analysis are required to ensure their effectiveness. Hence, this research explores hybrid encryption based on a generative adversarial network (GAN) for improved message encryption. A neural network was trained using the GAN method to defend against adversarial attacks. Various GAN training parameters were tested to identify the best model system, and various models were evaluated concerning their accuracy against different configurations. Neural network models were developed for Alice, Bob, and Eve using random datasets and encryption. The models were trained adversarially using the GAN to find optimal parameters, and their performance was analyzed by studying Bob's and Eve's accuracy and bits error. The parameters of 8,000 epochs, a batch size of 4,096, and a learning rate of 0.0008 resulted in 100% accuracy for Bob and 52.14% accuracy for Eve. This implies that Alice and Bob's neural network effectively secured the messages from Eve's neural network. The findings highlight the advantages of employing neural network-based encryption methods, providing valuable insights for advancing the field of secure communication and data protection. © 2024 Institute of Advanced Engineering and Science. All rights reserved.
publisher Institute of Advanced Engineering and Science
issn 25024752
language English
format Article
accesstype All Open Access; Gold Open Access
record_format scopus
collection Scopus
_version_ 1812871794656280576