Está en la página 1de 17

DIGITAL FILTER DESIGN

1

2 .Spectral Transformations of IIR Digital Filters • Objective . bandpass or bandstop filter. • z-1 has been used to denote the unit delay in the prototype lowpass filter G L(z) and ẑ-1 to denote the unit delay in the transformed filter GD(ẑ) to avoid confusion. highpass.Transform a given lowpass digital transfer function G L(z) to another digital transfer function G D(ẑ) that could be a lowpass.

zˆ = e • Transformation from z-domain to ẑˆ z  F ( z ) domain given by • Then GD  zˆ   GL  F  zˆ   3 .Spectral Transformations of IIR Digital Filters • Unit circles in z.and ẑ-planes defined by j jˆ z  e .

if z  1    1. if z  1 • Therefore 1/F(ẑ) must be a stable allpass function  1   l* zˆ    . ˆ  l  F  zˆ  l 1  z 1 L l  1 4 .Spectral Transformations of IIR Digital Filters • Fromz  F ( zˆ) z  F  zˆ  . if z 1  F  zˆ    1. hence  1. thus .

Lowpass-to-Lowpass Spectral Transformation • To transform a lowpass filter GL(z) with a cutoff frequency ωc to another lowpass ˆ c a cutoff frequency filter GD(ẑ) with . the transformation is 1 1   zˆ z 1  F  zˆ   zˆ   • On the unit circle we  jˆ have e   j e   jˆ 1e  1   Which yieldstan   2     tan  ˆ 2   1   5 .

2593z   1  0.5 dB ripple.Lowpass-to-Lowpass Spectral Transformation • Solving we get  sin   c  ˆ c  2  sin   c  ˆ c  2  • Example .35π. 6 .0662  1  z 1   1  0. • Redesign the above filter to move the passband edge to 0.3917 z 2  which has a passband from dc to 0.25π with a 0.6763z 1 3 1  0.Consider the lowpass digital filter GL  z   0.

the desired lowpass transfer function G is  zˆ   G  z  ˆ D L z 1  0.05    0.1934 z  1 0.1934 sin  0.3  • Hence.1934 zˆ 1 1 7 .Lowpass-to-Lowpass Spectral Transformation • Here sin  0.

bandpass-to-bandpass and bandstop-to-bandstop transformations.Lowpass-to-Lowpass Spectral Transformation • The transformation lowpass-to-lowpass 1   zˆ z   F  zˆ  zˆ   1 1 can also be used as highpass-tohighpass. 8 .

9 .Lowpass-to-Highpass Spectral Transformation • Desired transformation zˆ 1   z  1   zˆ 1 1 • The transformation parameter α is given by cos   c  ˆ c  2    cos   c  ˆ c  2  where ωc is the cutoff frequency of the ˆ c lowpass filter and is the cutoff frequency of the desired highpass filter.

6763z  0.3917 z  1 3 L 1 1 2 with a passband edge at 0.Transform the lowpass filter 0.15a   passband 0.3468 1  • The desiredz transformation is 1  0.3468 zˆ 1 10 .4  with at 0.0662  1  z  G  z   1  0. • Here.25π to a highpass filter edge    cos cos  0. 1 ˆ z  0.Lowpass-to-Highpass Spectral Transformation • Example .2593z   1  0.55π.3468  0.

3468 zˆ 1 1 11 .Lowpass-to-Highpass Spectral Transformation • The desired highpass filter is GD  zˆ   GL  z  zˆ 1  0.3468 z  1 0.

0 12 .Lowpass-to-Highpass Spectral Transformation • The lowpass-to-highpass transformation can also be used to transform a highpass filter with a cutoff at ωc to a lowpassˆfilter with a cutoff at . c • And transform a bandpass filter with a center frequency at ω0 to a bandstop filter with a centerˆ frequency at .

Lowpass-to-Bandpass Spectral Transformation • Desired transformation 2 1   1 zˆ  zˆ   1  1 1 z    1 2 2 1 zˆ  zˆ  1  1  1 2 13 .

are the desired upper and lower cutoff frequencies of the bandpass filter. 14 .Lowpass-to-Bandpass Spectral Transformation • The parameters α and β are given by cos   ˆ  ˆ  2   cos   ˆ  ˆ  2  c2 c1 c2 c1   cot   ˆ c 2  ˆ c1  2  tan  c 2  where ωc is the cutoff frequency of ˆ c1 and ˆ c 2 and the lowpass filter.

Lowpass-to-Bandpass Spectral Transformation • Special Case .The transformation ˆ c  ˆ c if can be simplified 2  c1 • Then the transformation reduces to 1 ˆ z  1 1 ˆ z  z 1   zˆ 1 ˆ 0 where  cos  ˆ 0  with denoting the desired center frequency of the bandpass filter. 15 .

Lowpass-to-Bandstop Spectral Transformation • Desired transformation 2 1 1   zˆ  zˆ  1  1  1 z  1   2 2 1 zˆ  zˆ  1 1  1  2 16 .

Lowpass-to-Bandstop Spectral Transformation • The parameters α and β are given by cos   ˆ  ˆ  2   cos   ˆ  ˆ  2  c2 c1 c2 c1   tan   ˆ c 2  ˆ c1  2  tan  c 2  where ωc is the cutoff frequency of ˆ c1 and ˆ c 2 and the lowpass filter. 17 . are the desired upper and lower cutoff frequencies of the bandstop filter.