Research On Weak Signal Extraction In A Strong Noise Environment Based On Wavelet Analysis

The reliability of experimental data plays a vital role on the model parameter identification, but there exists noises in all systems. How to improve the authenticity of the experimental data is particularly important. Often, the useful signal is relatively weak and submerged in strong noise, so more attention has been paid to the problem of how to measure the weak signal in strong noise. The Wavelet Transform has good characters of multi-resolution and time-frequency localization, it resolve a lot of problems which had not been done by Fourier Transform, and the Wavelet Transform methods provide powerful tools for signal processing and offers an effective way for measure the weak signal in strong noise.In this paper, the acceleration signal of the slider on the cantilever was extracted by using the wavelet transform method. Firstly, the theory of wavelet transform and weak signal extraction are introduced and the feasibility of the weak signal extraction from strong noise environment based on wavelet transform is verified by the MATLAB simulation experiments. Then, collected the image information of the slider by the high-speed photography experiment platform. Analysised the image information by Proanalyst and calculated the acceleration-time data of the slider which submerged in strong noise because of the experimental environment and the differential operator. Finally, by contracting the results of friction model parameter identification by using the wavelet transform method and Butterworth method, the superiority of the wavelet transform method in weak signal extraction from the strong noise environment is verified and some advanced results are also obtained.