| As a fast-growing emerging subject in areas such as mathematics and digital signal processing, wavelet analysis and its application provided a new perspective and a whole new way of thinking and analysis methods for these areas. It offered a gradual approximation subspace for function space theory of mathematics, these subspace have different degree of approximation, and gradually approaching to the whole space, so we can approximate the original function with different accuracy and resolution through a series of functions. It also provided a local analysis method in time and frequency domain for signal processing, and when the time-frequency window moved in the entire frequency plane, the bandwidth would change adaptively with frequency changes, therefore, the wavelet is known as "mathematical microscope". After the explore and study from many disciplines and fields, the important mathematical form of wavelet has basically established, so it’s more solid of the theoretical basis, and become the intersection of many areas of research. These results make the wavelet theory to enrich and improve continuously, therefore it has been widely used in many professional disciplines and fields.To filter out noise from the original signal is an important part of signal processing, although some de-noising results can be received by the methods now being widely used as smooth analysis and Fourier analysis. However, in some aspects like sensitivity, accuracy and speed, etc. are still unable to get satisfactory results and there are many shortcomings in these methods. Because the wavelet analysis can be achieved both in the time domain and frequency domain signal analysis, and could effectively break down a variety of time-varying signal to accomplish the separation of signal and noise efficiently and finally obtain satisfactory de-noising results. In this paper, the application of wavelet analysis in the vehicle gear data processing had been studied and discussed intensively.Vehicle gear data processing played an important role during the experiment of engine system, according to the current engine speed and vehicle speed the current gear of transmission could be determined. But the signal acquisition process will inevitably be doped with noise, in this paper, wavelet de-noising approach for signal de-noising in order to get more accurate gear data. And MATLAB was used in writing programs to achieve a combination algorithm of the Haar wavelet, Symlets wavelet and the threshold de-noising algorithm and get good results.Based on the problems in signal processing of vehicle gear, wavelet theory and the theory of de-noising based on wavelet analysis had been studied in this paper. A variety of traditional wavelet de-noising methods commonly used such as wavelet threshold de-noising method, wavelet de-noising based on stationary and de-noising based on wavelet packet were analyzed in detail and carried out in MATLAB environment for comparison test. And on the basis of the theory, several de-noising method such as the Haar wavelet, Symlets wavelet and wavelet threshold applicated in signal de-noising in the vehicle gear, and experimental of vehicle gear signal de-noising based on MATLAB simulation platform has been achieved. Experimental and simulation results showed that de-noising based on wavelet analysis can retain the original characteristics of the signal, while play inhibitory effect to noise. And the comparison tests showed that in the same experimental and parameters conditions the de-noising method through a combination of the above types of wavelet could get better de-noising results compared to the traditional wavelet-based threshold method. |
PDF Full Text Request