TOPICS  IN DIGITAL SIGNAL PROCESSING 
C.  S.  BURRUS  and  T.  W.  PARKS:  DFTIFFT  AND  CONVOLUTION 
ALGORITHMS 
Rice  University 
T. W.  PARKS and  C. S.  BUR RUS:  DIGITAL FILTER  DESIGN 
Cornell  University and  Rice  University 
J.  TREICHLER,  R.  JOHNSON,  JR.  and  M .   LARIMORE:  THEORY AND 
DESIGN  OF ADAPTIVE FILTERS 
Cornell  University and Applied Signal Technology,  Inc. 
W.  KOHN:  DIGITAL  CONTROL (in preparation) 
Rice University 
Digital Filter Design 
T. W. Parks 
School of  Electrical Engineering 
Cornell University 
Ithaca, New York  14853 
C. S. Burrus 
Department of  Electrical and Computer Engineering 
Rice University 
Houston, Texas 77251 
A WILEY-INTERSCIENCE PUBLICATION 
JOHN WlLEY & SONS,  Inc. 
New York  Chichester  Brisbane  Toronto  Singapore 
Copyright 0 1987 by  Texas Instruments Incorporated 
Published  by  John  Wiley & Sons, Inc. 
All  rights reserved. Published  simultaneously in Canada. 
Reproduction  or translation of  any part of  this work 
.  beyond that permitted  by  Section 107 or 108 of  the 
1976 United  States Copyright Act without  the permission 
of  the copyright  owner is unlawful. Requests for 
permission or further information should be  addressed  to 
the Permissions  Department, John Wiley & Sons, Inc. 
Library  of  Congress Cataloging-in-Publication  Data: 
Parks, T. W. 
Digital filter design. 
"A  Wiley-Interscience publication." 
Includes bibliographies and index. 
1.  Electric filters, Digital-Design  and 
construction.  L  Burrus, C. S.  11.  Title. 
TK7872.F5P37 
ISBN  0-471 -82896-3 
1987 
621.3815'324 
Printed in the United  States of  America 
86-32500 
To our parents 
William and Mildred Parks 
and 
Aleta  Huffman 
Preface 
This digital  filter design  book  is  addressed  to  the  mathematician, scientist, or 
engineer who  has  an  understanding  of  continuous-time  signals  and  who  has 
been introduced to discrete-time signal analysis. 
The  main  topic of  this  book  is  the  frequency-domain  analysis  and  design 
of  linear,  constant  coefficient,  digital  filters.  The  book  is  divided  into  two 
major parts: finite-duration impulse-response (FIR) filters and infinite-duration 
impulse-response (IIR) filters. Each part  consists of  a complete, self-contained 
treatment  of  the  corresponding  filter  type.  All  aspects  of  each  filter  type  are 
discussed. Each  part  begins  with  a discussion  of  filter  properties,  which  leads 
into material  on  design  of  the  filter  to meet  frequency-domain  specifications. 
This aspect of  filter design is called the approximation problem  and makes up a 
major  portion  of  the book. Each of  the two parts concludes with a chapter on 
realization  pro- 
implementation  of  the  filter  with  fixed-point  arithmetic-the 
blem. The chapters on implementation both include a detailed design example 
that  presents  a step-by-step design and  implementation of  a typical filter. The 
design examples begin  with  the  frequency-domain  specifications for  the  filter 
and conclude with a listing of the assembly language program for implementing 
the filter on a signal-processing chip (the TMS32010 from Texas Instruments). 
The  book  begins  with  an  introductory  chapter  that  reviews  the  concepts  of 
frequency-domain  analysis  of  discrete-time  systems  and  states  the  major 
problems in digital filter design. The final chapter summarizes the main  results 
in the book  with a discussion of  the unique characteristics  of  the FIR and IIR 
filter  types.  An  appendix  with  listings of  ten  FORTRAN  programs  for  filter 
design is included. 
This book  may  be  used  in  several ways. For some applications one might 
viii 
Preface 
turn to the appendix, run the appropriate design program to get the coefficients 
of  a  filter  that  meets  given  frequency-domain specifications, then  turn  to the 
listing in  the  design example, insert  the  coefficients in  the listing, and  run  the 
program on a TMS32010. If all goes smoothly in this process, one may not need 
to read and completely understand the theory in the book. However, if, as often 
happens, the problem one is faced with is not exactly covered by the programs in 
the appendix, then with  some reading of  the theory, one can probably  modify 
the  appropriate  design  program  or  write  a  special  program  to  obtain  the 
appropriate  filter coefficients.  Even  if  the  coefficients can  be  obtained  from  a 
program in the appendix, the implementation in the design example may not be 
exactly what  one  wants.  For  example,  the  filter  may  take  too  much  time  to 
execute or may require too much memory or may have undesirable quantization 
effects.  Again,  some  time  spent  in  reading  the  theory  in  the  chapters  on 
implementation  should  allow  the  reader  to  develop an  appropriate  implem- 
entation of  the desired digital filter. 
This book would not have been written without the support and encourage- 
ment of Texas Instruments, Inc. We would especially like to thank Mike Hames, 
who  has  always been  ready  with  a  smile  and  a  helping hand  when  all  of  us 
realized just  how much work is involved in writing a book. Maridene Lemmon 
has  continued  to  patiently  correct  and  improve  our  writing  styles  and  has 
carefully read through countless revisions of the manuscript. The engineers at TI 
have read early versions of  the text and helped correct our errors. 
We would  like to thank  Professor H. W. Schiissler who helped  us  begin to 
understand  the issues in digital filter implementation when he  was on leave at 
Rice University. Some of  our examples are taken from his notes. We would like 
also to thank  Cole Erskine for working out the  two detailed design examples 
and for providing the necessary TMS32010 code. Jim Kaiser and Dick Roberts 
provided us  with  very  thorough  reviews of  the  manuscript  and  made  several 
good suggestions, which we have incorporated  in the text. 
We  appreciate  the  long hours  of  reading put  in  by  our  graduate  students 
Doug Jones and Henrik  Sorenson, who have made many good suggestions for 
improving the book. Thanks also is given to the students at Rice University in 
our digital signal-processing courses who have helped us develop this book over 
the years. 
Contents 
Part One 
INTRODUCTION 
1.  Introduction to Digital Filters 
1.1.  Properties of  Discrete-Time  Systems 
1.2.  Linear, Stationary, Discrete-Time  Systems 
1.3.  Frequency  Response and Transfer Functions 
1.4.  Digital Filter  Design 
1.4.1  The Approximation  Problem 
1.4.2  The Realization  Problem 
1.5.  Properties of  FIR and  IIR  Filters 
References 
Part Two 
FINITE IMPULSE-RESPONSE (FIR) FILTERS 
2.  Properties of Finite Impulse-Response Filters 
2.1.  Frequency-Domain  Description  of  FIR Filters 
2.2.  Linear-Phase  FIR Filters 
2.2.1.  Four Types of  Linear-Phase FIR Filters 
2.2.2.  Calculation  of  FIR Filter  Frequency  Response 
2.2.3.  Zero Locations for  Linear-Phase FIR Filters 
References