Wavelet Analysis Of The Inclination Sensor Signal Processing

After more than 10 years of development, wavelet theory has made many very important achievements, analysis methods becoming more and more plentiful ,application range being increasingly broad. The success or failure of Wavelet analysis depends on particular practical issues. How to select the optimal wavelet type and approach, according to the characteristics of signal and processing need, has been one of the important issues in the field of wavelet theory and application research.First of all, this paper studies the latest theory of wavelet, after systematic and deep discussion of the wavelet signal processing methods, it provides a new denoising methods and gives experimental test results. On the other hand, though very high accuracy can be done now, but angle sensor products are still very vulnerable to outside interference. Therefore, this article focuses on the applications of wavelet analysis in angle sensor signal processing, including the angle sensor signal denoising and signal identification when the sensor is inspired by different signals. During denoising processing research which depends on the one-dimensional wavelet transform, this paper gives a new function for signals whose noise variance is already known. As for signals with unknown variance, a GCV rule methods combining genetic algorithm is proposed. As signal of tilt sensor would be mixed with acceleration influence during dynamic measurement, possibility of distinguish acceleration influence with wavelet is discussed.