Extracting periodic driving signal from chaotic noise *

Abstract

Abstract After periodic signals pass through some nonlinear systems, they are usually transformed into noise-like and wide-band chaotic signals. The discrete spectrums of the original periodic signals are often covered by the chaotic spectrums. Recovering the periodic driving signals from the chaotic signals is important not only in theory but also in practical applications. Based on the modeling theory of nonlinear dynamic system, a “polynomial-simple harmonic drive” non-autonomous equation (P-S equation) to approximate the original system is proposed and the approximation error between P-S equation and the original system is obtained. By changing the drive frequency, we obtain the curve of the approximation error vs. drive frequency. Basedon the relation between this curve and the spectrums of the original periodic signals, the spectrum of the original driving signal is extracted and the original signal is recovered.