Notes on Digital Signal Processing: Practical Recipes for Design, Analysis and Implementation

Note 67. Parametric Modeling of Discrete-Time Signals

Many sampled real-world signals can be modeled as the output of a discrete-time filter that is driven by a simple input sequence. Many techniques for the analysis and characterization of signals are based upon such signal models. This note serves as an introduction to the basic model classes that are explored in subsequent notes.

Parametric modeling is the name given to the concept of modeling a discrete-time signal as the output of a discrete-time filter that is driven by a specified input sequence—usually either white noise or the unit-sample function. Many techniques used in the areas of statistical signal processing, spectral estimation, and adaptive filtering are built upon the theoretical ...