The Research on Identification for Electromagnetic Interference in Automobile Based on WPD and MLPNN

The technique for recognizing and identifying disturbed signals based on wavelet packet decomposition (WPD) and multilayer perceptron neural network (MLPNN) was proposed. It was greatly reduced the volume of computation after Parseval´s theorem energy rule and feature extraction of the disturbed signals emerged by the equipment called EM-Test which could bring confirmed automotive interferential signals on automobile. A neural network was also developed for fast interferences identification.