Time - Frequency Analysis Techniques Applied to Automotive Noise and Vibration Signals

Automotive stationary noise and vibration signals are normally analyzed using Fourier methods. However, many noise and vibration signals are non-stationary (transient or time-varying). In those situations, the time characteristics of the signals can be lost using standard Fourier methods. Lately, time-frequency (TF) analysis methods have become more popular and are applied in many different areas of NVH (Noise, Vibration, and Harshness) in order to preserve the time-frequency information. The objective of this paper is to present some of the different time-frequency analysis tools, such as the Short Time Fourier transform (spectrogram), the Gabor Transform, the Wavelet transforms (scalograms), and the Wigner-Ville Distribution. Examples of application of these techniques to automotive non-stationary noise and vibration signals are presented.