Adaptive model-free synchronization of different fractional-order neural networks with an application in cryptography

In this paper, an adaptive model-free control method is designed to synchronize a class of fractional-order neural networks which has a vast application in engineering and industry. The theoretical and analytical concepts of the method are based on the fractional-order version of the Lyapunov stability theorem and using adaptive control theory. Moreover, it is worth to mention that, because of using of boundedness property in states of chaotic systems, there is no trace of nonlinear/linear dynamic terms of the system in the control approach. Also, for the application point of view, a new crypto-system algorithm is proposed based on the designed adaptive model-free method for encryption/decryption of unmanned aerial vehicle color images. Plus, numerical simulations are created to emphasize the usability of the method and algorithm. Finally, this point should be emphasized that security analysis including key space analysis, key sensitivity analysis, histogram analysis, information entropy analysis and correlation analysis of the crypto-system are provided to confirm the results of the crypto-system.