MKB4062 – PENGOLAHAN SINYAL DIGITAL

INFINITE IMPULSE RESPON

FILTER DIGITAL
 INFINITE IMPULSE RESPONSE ;
FILTER FORMAT

An IIR filter is described using the difference equation

gives the IIR filter transfer function as

2

where bi and ai are the (M+1) numerator and N denominator coefficients,

respectively. Y(z) and X(z) are the z-transform functions of the filter input x(n)
and filter output y(n).
 EXAMPLE

Given the IIR filter

determine the transfer function, filter length, nonzero coefficients, and impulse response.
Solution:
Applying the z-transform and solving for a ratio of the z-transform output over input, we
have

We also identify the nonzero numerator coefficients and denominator coefficient as

b0 = 0,2 ; b1 = 0,4 ; a1 = 0,5
 EXAMPLE (NEXT…)

To determine the impulse response, we rewrite the transfer function as

Using the inverse z-transform and shift theorem, we obtain the impulse response
as

The obtained impulse response has an infinite number of terms, where the first
several terms are calculated as
h(0) = 0,2 ; h(1) = 0,7 ; h(2) = 0,25
 DESIGN OF IIR FILTER

At this point, we can make following remarks:

1. The IIR filter output y(n) depends not only on the current input x(n) and past
inputs x(n-1), ., but also on the past output(s) y(n-1), ., (recursive terms). Its transfer
function is a ratio of the numerator polynomial over the denominator polynomial,
and its impulse response has an infinite number of terms.
2. Since the transfer function has the denominator polynomial, the pole(s) of a
designed IIR filter must be inside the unit circle on the z-plane to ensure its stability.
3. Compared with the finite impulse response (FIR) filter, the IIR filter offers a much
smaller filter size. Hence, the filter operation requires a fewer number of
computations, but the linear phase is not easily obtained. The IIR filter is preferred
when a small filter size is called for but the application does not require a linear
phase.
 DESIGN OF IIR FILTER

• The digital filter design problem is to determine the coefficients bi and ai so that
H(z) satisfies the given specifications.
• The purpose of designing a digital IIR filter is to determine the transfer function
H(z) which approximates the prototype analog filter.
• Two methods can be used to map the analog filter to the equivalent digital filter:
the impulse-invariant method and the bilinear transform.
• The impulse-invariant method preserves the impulse response of the original
analog filter by sampling its impulse response, but has inherent aliasing problems.
• The bilinear transform can preserve the magnitude response characteristics of the
analog filters, and thus is better for designing frequency-selective IIR filters.
 BILINEAR TRANSFORMATION DESIGN
METHOD

The IIR filter design method with this

bilinear transformation makes a digital
filter by transforming the H(s) equation
from an analog filter, so the process of
designing an IIR filter with a bilinear
transformation method starts from the
process of designing an analog filter. This
analog filter is then transformed into a
digital filter by the bilinear transformation
method, as shown in Figure 1.
 BILINEAR TRANSFORMATION DESIGN
METHOD

• Analog filters have a frequency

response of up to f = ꝏ, while
digital filters are only up to fs/2.
In other words, the point f = ꝏ
on the analog filter is shifted to
the point fs/2. This process will
result in a shift in the cut-off
point (fc). The fc frequency point
on the digital filter will be shifted
slightly to the left, as shown in
figure 2.
 BILINEAR TRANSFORMATION DESIGN
PROCEDURE

Now we can summarize the BLT design procedure.

• Given the digital filter frequency specifications, prewarp the digital frequency
specifications to the analog frequency specifications.
For the lowpass filter and highpass filter:

For the bandpass filter and bandstop filter :

4
and
 BILINEAR TRANSFORMATION DESIGN
PROCEDURE

• Perform the prototype transformation using the lowpass prototype HP(s)

From lowpass to lowpass :
5
From lowpass to highpass :

6
From lowpass to bandpass :

7
From lowpass to bandstop:

8
 BILINEAR TRANSFORMATION DESIGN
PROCEDURE

• Substitute the BLT to obtain the digital filter

9
 EXAMPLE

The normalized lowpass filter with a cutoff frequency of 1 rad/sec is given as

Use the given HP(s) and the BLT to design a corresponding digital IIR lowpass
filter with a cutoff frequency of 15 Hz and a sampling rate of 90 Hz.
Solution:
First, we obtain the digital frequency as
 EXAMPLE (NEXT…)

We then follow the design procedure:

• Calculate the pre warped analog frequency as
 EXAMPLE (NEXT…)

• Then perform the prototype transformation (lowpass to low pass) as follows:

This yields an analog filter:

• Apply the BLT, which yields

THANKS