## Abstract

Much of the discussion in Chapters 1 to 4 has centred on what can be classified as deterministic signals (both continuous and discrete). Such signals can be described as ‘weighted sums of complex exponentials’ and are thus highly predictable in the following sense: given the Fourier transform of a signal we can work out exactly what the value of that signal would be at any time *t*. In practical applications other signals are encountered which are not amenable to such a description and are not exactly predictable. Such signals are often termed as ‘noise’. However, using concepts from probability theory, the Fourier/Laplace descriptions can be modified to accommodate these ‘noise’ signals.

## Preview

Unable to display preview. Download preview PDF.

## Copyright information

© Bernard Mulgrew, Peter M. Grant and John S. Thompson 1999