Utilization of Artificial Neural Networks for Fast Encryption and Decryption of Messages
Ayman Haggag · International Journal of Advanced Trends in Computer Science and Engineering · 2020
This paper aims to provide a fast method to decrypt short encrypted messages that consist of 10 letters by using the technique of trained Feed-Forward Artificial Neural Network (FF-ANN).The use of artificial neural network is one of the methods of artificial intelligence, a method that mainly depends on increasing the training of the network on many and many examples.These examples must take the same context of the main purpose of this research so that the results of the network are very close to the desired and desired realism.We give the feature of our trained FF-ANN and show how we train our network using 50 encrypted messages using our proposed method to achieve the best results.Our collection of test patterns of cryptographic messages took a shorter decryption time compared to the solution using mathematical methods.We can apply this method presented in this research in the large messages that the use of mathematics for encryption and decryption is very difficult and consumes a very long time, but the use of the neural network will be easy as a method and the expectation for time consumption to be little, especially with good network training which may be of interest in next researches.