Wavelet Transform Applied on Vehicle Air Flow Meter (AFM) Signal De-Noising

Abstract

In engine compartment, it is difficult to eliminate the effect of interfering signals on vehicle AFM signal by the traditional processing method. According to the characteristic of AFM signal, this paper decomposed the noise signal for 4 layers by selecting Daubechies (db5) wavelet in Matlab based on wavelet theory. After decomposing, and then employed the soft-threshold de-noising method to eliminate noise. At last this paper used the inverse wavelet transform to reconstruct signals for signal de-noising and rebuilding. The test results show that this method can eliminate the noises of AFM signal effectively.