8Digital Ladder Filters

The filter structures presented in the previous chapters are deduced directly from their Z-transfer functions, with the coefficients applied to the multiplying circuits being those of the powers of Z−1. More elaborate structures can be developed.

In analog filtering, structures exist which allow filters with very low ripple and excellent selectivity to be constructed using passive components of limited precision. In digital filtering, these properties can be translated into a reduction in the round-off noise and in the number of bits representing the coefficients.

Analog filter networks are based on cascading two port circuits whose properties will be considered first [1].

8.1 Properties of Two-Port Circuits

The general two-port circuit terminated by the resistors R1 and R2 is shown in Figure 8.1, together with the currents I and voltages V at ports 1 and 2. This circuit, assumed to be linear, is defined by its impedance matrix z, which establishes the relations between the variables, generally written in reduced form as:

upper R equals upper R 1 equals upper R 2 semicolon upsilon equals StartFraction upper V Over StartRoot upper R EndRoot EndFraction semicolon i equals upper I StartRoot upper R EndRoot

Then,

(8.1)upsilon equals italic z i

with:

z equals Start 2 By 2 Matrix 1st Row 1st Column z 11 2nd Column z 12 2nd Row 1st Column z 21 2nd Column z 22 EndMatrix

The values z12 and z21 are the transfer impedances of the two-port circuit. It is reciprocal if z12 = z21.

If ...

Get Digital Signal Processing, 10th Edition now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.