What question did this study set out to answer?

This research aims to improve symmetric-key cryptography using deep learning techniques to address key management and security issues.

June 3, 2026Open Access

Symmetric-Key Cryptography Using Deep Learning Techniques

Puntos clave

This research aims to improve symmetric-key cryptography using deep learning techniques to address key management and security issues.
Developed an intelligent encryption system combining ANN, CNN, and RNN with AES encryption.
Utilized Python, TensorFlow, Keras, and PyCryptodome for implementation.
Evaluated system performance using CICIDS and NSL-KDD cybersecurity datasets.
The intelligent AES system achieved a validation accuracy of 96.4%.
Demonstrated reduced time consumption during encryption and decryption phases.
Enhanced attack detection capabilities compared to conventional AES systems.

Resumen

With the increasing growth of cyber threats and intelligent attack techniques, there is a perpetual demand for stronger and adaptive Systems. Although traditional symmetric-key cryptographic systems enable efficient encryption, they also suffer from key management, attack, time complexity, and adaptive security issues. The introduction of the securities markets of emerging economies as well Islamic capital markets is a logical outcome of globalization. A combination of the model takes place within the new system, such as the Artificial Neural Network (ANNs), Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), and the encryption of Advanced Encryption Standard (AES). The implemented system makes use of Python, TensorFlow, Keras and PyCryptodome library. Also, it is evaluated with the help of cybersecurity datasets such as CICIDS and NSL-KDD. According to the experimental results, the validation accuracy of the proposed intelligent AES system is up to 96.4%. It consumes less time during the encryption and decryption phases. Moreover, it also enhances computational efficiency. Furthermore, it can detect attacks better than the conventional AES systems. As a result, deep learning technology can help in the enhancement of adaptive security, intelligent key analysis, and encryption optimization

Me gusta

Guardar

Ver artículo completo