Signal Processing in C

CHAPTER 14

Filtering and Z-Transforms

14.1 INTRODUCTION

This chapter introduces the concepts of poles and zeros as the components of linear filters. Every linear, time-invariant differential equation may be characterized in terms of the poles and zeros of its frequency response function. The same notion carries over into discrete-time systems. Discrete-time versions of the continuous-time poles and zeros are used to build arbitrary discrete-time filters (assuming linearity and time invariance).

This chapter begins by analyzing the continuous-time case, starting from the general differential equation and continues by developing the discrete-time solutions, computing the gains of these discrete-time systems, and comparing them to the continuous-time systems.

14.2 POLES AND ZEROS FROM DIFFERENTIAL EQUATIONS

14.2.1 The All-Pole Equations

Every linear differential equation can be translated into the language of linear filters and vice versa ; they are equi valent concepts. Usually, physical systems are initially modeled with linear differential equations , from which the filter can be calculated. In this text we are concerned only with time -invariant systems. The general form of an nth-order. time-independent, linear differential equation is

In terms of the differential operator D, the derivative with respect to time, this can be written as

where q is the polynomial

The method of ...