Iterative learning identification for chaos communication

An approach for secure communication using chaotic systems is proposed, where the information signal to be transmitted is over a finite time interval. A parameter of the chaotic system is modulated by the information signal, and the mechanism for generating the transmitting signal is a pre-specified nonlinear function. Iterative learning identification is applied to recovery the masked signal, which features complete recovery along the entire interval. In this paper, fully-saturated learning law is proposed and a sufficient condition for the convergence and robustness is derived, by which the learning gain can be chosen. Moreover, two techniques, the transmitting-signal-involved parameter modulation and the composite nonlinear masking, are presented for security enhancement.