Signal estimation in chaos using nonlinear prediction

The problem of parameter estimation in chaotic noise is considered in this paper. Based on the inherently deterministic nature of a chaotic signal — the short term predictability, a novel estimation approach called minimum nonlinear prediction error (M) technique is proposed. The parameters of a signal can be accurately estimated by minimizing the nonlinear prediction error of the output of an inverse filter of the received signal. Monte Carlo simulations are carried out to demonstrate the efficiency of the MNPE approach. It is shown that not only could the chaotic approach provide an accurate estimation, but it is more effective than the conventional statistic approach in the sense that the chaotic estimation approach has a smaller mean squares error (MSE).