Feature Extraction of Acoustic Signals Based on Complex Morlet Wavelet

摘要: Abstract This article studies feature extraction of acoustic signals based on complex Morlet wavelet. Since, parameter optimization is the important and difficult point wavelet application. In this article, a new 2-parameter process algorithm proposed, i.e., ascertaining parameter, getting optimized value other then repeating until system performance index satisfied. Based optimal scalogram obtained from applying transform to typical emission (AE) signal, region segmented location method designed judge exactly number AE signal frequencies whose exact values are calculated. By way, error induced by misjudgment misreading can be avoided effectively. Finally, compared with results traditional feature. Simulation show that such methods improve precision have good engineering value.