Introduction to Spectral Analysis of Non-Stationary Random Signals - Spectral Analysis: Parametric and Non-Parametric Digital Methods - page 231

Chapter 9
Introduction to Spectral Analysis of
Non-Stationary Random Signals
The spectral analysis of non-stationary random signals can be made, as in the
stationary case , by using two large categories of methods:
- non-parametric methods;
- parametric methods.
In each of these large categories, we also find two types of methods specific to
the non-stationary case. Actually, in non-stationary context, the spectral analysis can
be made:
- either by using the stationary tools and by adapting them to the non-stationary
nature of signals;
- or by introducing new techniques.
Thus, the reader will find three parts in this chapter: a first part which makes it
possible to set out the problem of defining the “evolutive spectrum”, a second part
which tackles the non-parametric techniques and the third part which creates a
panorama of the parametric methods used in a non-stationary context.
A more complete panorama of the tools for spectral analysis of non-stationary
random signals can be found in [HLA 05].