Design And Development of Deep Learning Model of Chaotic Signal Processing for Efficient Data Encryption

This project explores the chaotic generation of signals and the encryption and decryption of these signals using a deep learning model. The signal generation involves the use of two map generators: the Chebyshev map generator and the Logistic map generator. The proposed deep learning model can be used to develop new and improved chaotic signal modulation and demodulation schemes for secure and robust data encryption in a variety of applications, such as communication, financial transactions, and military communications. The synchronization of chaotic sequences is necessary for coherent Chaos Shift Keying (CSK) schemes, yet it is still nearly impossible in disruptive environments. In order to overcome this limitation, an LSTM based receiver is utilized for online modulated signal classification after being trained offline via chaotic maps across a limited number of channel realizations.