Based On Wavelet Transform And Chaos Theory Of Weak Signal Detection Method

Detecting weak signals in noise background research has become a cutting-edge disciplinesmeasurement field, but the field of engineering applications research problems. The weak signaldetection characteristics: First, the weak signal detection in low signal-to-noise ratio; second, theweak signal detection required to detect fast and real-time. The existing traditional detectionmethods are mostly the linear detection methods, but are limited by the input signal-to-noise ratioof the weak signal can not be accurately detected, there is an urgent need to find a new method fordetecting.1992, since the concept was first proposed by Birx this theory has aroused the attentionof researchers lines, and its sensitivity to weak periodic signal and noise immunity characteristicsmake it become a weak signal detection a unique and effective way, with a very large potentialapplication value.In this paper, on the basis of the relevant literature, comparative analysis of three methods ofwavelet de-noising, including: Forced de-noising, default threshold de-noising processing andgiven soft or hard threshold de-noising processing, Finally comprehensive advantages anddisadvantages to choose the default threshold value de-noising treatment method for noise underthe background of weak signal de-noising Selected db4wavelet for six decomposition, M filewritten in the MATLAB simulation software to get the wavelet de-noising before and after thewavelet de-noising signal simulation graph and the figure of low frequency coefficient and highfrequency coefficient, can be directly observed details of wavelet de-noising effect.This article discusses the chaos definitions, characteristics and chaotic systems discriminatingmethod, the discrimination including: phase plane method, timing diagrams, the Lyapunov index,Poincare law and fractal dimension calculation. Depending on the simulation need to select thecombination of phase plane method and timing diagram discriminant method to accurate determinethe state of the system. This paper describes the chaos theory in several typical chaotic oscillatormodel, including the Duffing oscillator model, the Vander-Duffing oscillator model, the Lorenzoscillator model, the Chen oscillator model, Rossler oscillator model and Birkhoff-shaw oscillatormodel, and gives the corresponding simulation graphics through the MATLAB software simulationgraphics.This article details the introduction of based on Duffing oscillator weak signal detection,including the Duffing oscillator to detect weak signal of amplitude, the Duffing oscillators todetect the weak signal frequency and the weak signal phase, in the analysis of the principle ofrespectively, through the build model in MATLAB Simulink, simulation graphics are given respectively. Application Melnikov function of Duffing oscillators to detect the weak signalfrequency, with the oscillator detection frequency comparison, graphics display results more clear,more precise, Finally adopts the wavelet de-noising and the combination of Duffing oscillatorsystem for weak signal detection, given the principle of the chaotic signal wavelet de-noising, andwere compared with use of Duffing oscillator system to detect weak signal amplitude andfrequency, the detection threshold and signal-to-noise ratio compared to prove combine thewavelet transform and chaos theory system on the weak signal detection more efficient, moreaccurate, particularly evident in the relatively large noise power of the weak signal detectiontheory is applied to engineering practice laid a certain foundation.