Introduction

A time history of a noise or vibration signal is just a direct recording of an acoustic pressure fluctuation, a displacement, a velocity, or an acceleration waveform with time – it allows a view of the signal in the time domain. A basic noise or vibration meter would thus provide a single root-mean-square level of the time history measured over a wide frequency band which is defined by the limits of the meter itself. These single root-mean-square levels of the noise or vibration signals generally represent the cumulative total of many single frequency waves since the time histories can be synthesised by adding single frequency (sine) waves together using Fourier analysis procedures. Quite often, it is desirable for the measurement signal to be converted from the time to the frequency domain, so that the various frequency components can be identified, and this involves frequency or spectral analysis. It is therefore important for engineers to have a basic understanding of spectral analysis techniques. The appropriate measurement instrumentation for monitoring noise and vibration signals were discussed in section 4.3 in chapter 4. The subsequent analysis of the output signals, in both the time and frequency domains, forms the basis of this chapter.

Just as any noise or vibration signal that exists in the real world can be generated by adding up sine waves, the converse is also true in that the real world signal can be broken up into sine waves such as to describe its frequency content.