Block Diagrams and Transfer

Functions
Just as with CT systems, DT systems are
conveniently described by block diagrams and
transfer functions can be determined from them. For
example, from this DT system block diagram the
z Transform Signal and System difference equation can be determined.
Analysis
Chapter 12

1
y[ n ] = 2 x[n ] ! x[n ! 1] ! y[n ! 1]
2

7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 2

Block Diagrams and Transfer Block Diagram Reduction

Functions All the techniques for block diagram reduction
From a z-domain block diagram the transfer function can introduced with the Laplace transform apply exactly to z
be determined. transform block diagrams.

1
Y( z) = 2 X( z) ! z !1 X( z ) ! z !1 Y( z)
2
Y( z) 2 ! z !1 2z ! 1
H( z ) = = =
X( z) 1+ 1 z !1 z + 1
2 2
7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 3 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 4

System Stability System Interconnections

Cascade

A DT system is stable if its impulse response

is absolutely summable. That requirement
translates into the z-domain requirement that
all the poles of the transfer function must lie Parallel
in the open interior of the unit circle.

7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 5 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 6

1
 System Interconnections Responses to Standard Signals
N( z)
If the system transfer function is H( z ) = the z
D( z)
z N( z)
transform of the unit-sequence response is Y( z) =
z ! 1 D( z)
which can be written in partial-fraction form as
N ( z) z
Y( z) = z 1 + H(1)
D( z ) z !1
Y( z) H1( z) H (z )
H( z ) = = = 1 N (z )
X( z) 1+ H1( z) H2 ( z) 1 + T( z) If the system is stable the transient term, z 1 , dies out
D( z)
z
and the steady-state response is H(1) .
T( z) = H1( z) H 2 ( z) z !1

7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 7 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 8

Responses to Standard Signals Responses to Standard Signals

Unit Sequence Response
Kz
Let the system transfer function be H( z ) = One-Pole System
z! p

z Kz Kz " z pz %
Then Y( z) = = $ ! '
z ! 1 z ! p 1 ! p # z ! 1 z ! p&

and y[ n ] =
K
1! p
(1 ! pn+1 ) u[n]

Let the constant, K be 1 - p. Then y[ n ] = (1 ! p n+1 ) u[ n ]

7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 9 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 10

Responses to Standard Signals Responses to Standard Signals

N( z)
Unit Sequence Response If the system transfer function is H( z ) = the z transform
D( z)
Two-Pole System of the response to a suddenly-applied sinusoid is
N( z) z[ z ! cos( "0 )]
Y( z) =
D( z ) z2 ! 2z cos( "0 ) + 1
The system response can be written as
" N ( z )%
y[ n ] = Z !1 $ z 1 ' + H( p1 ) cos((0 n + ) H( p1 )) u[ n ]
# D( z ) &
and, if the system is stable, the steady-state response is
H( p1 ) cos(!0 n + " H( p1 )) u[n ]
a DT sinusoid with, generally, different magnitude and phase.

7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 11 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 12

2
 Pole-Zero Diagrams and Pole-Zero Diagrams and
Frequency Response Frequency Response
For a stable system, the response to a suddenly-applied
sinusoid approaches the response to a true sinusoid (applied Let the transfer function of a DT system be
z z
for all time). H( z ) = =
z 5 ( z ! p1 )( z ! p2 )
z ! +
2
2 16
1+ j2 1! j2
p1 = p2 =
4 4

e j!
H( e j! ) =
e " p1 e j! " p2
j!

7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 13 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 14

Pole-Zero Diagrams and The Jury Stability Test

Frequency Response Let a transfer function be in the form, H( z ) =
N( z)
D( z)
where D( z) = a D z D + aD!1 z D!1 +! + a1z + a0

Form the “Jury” array

1 a0 a1 a2 ! a D!2 aD!1 aD
2 aD aD!1 aD!2 ! a2 a1 a0
3 b0 b1 b2 ! bD!2 bD!1
4 bD!1 bD!2 bD!3 ! b1 b0
5 c0 c1 c2 ! cD!2
6 cD!2 cD!3 cD!4 ! c0
" " " " #
2D ! 3 s0 s1 s2

7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 15 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 16

The Jury Stability Test Root Locus

The third row is computed from the first two by Root locus methods for DT systems are like root
a0 aD a0 aD!1 a0 aD!2 a0 a1
locus methods for CT systems except that the
b0 = , b1 = , b2 = , ! , bD!1 = interpretation of the result is different.
aD a0 aD a1 aD a2 aD a D!1

The fourth row is the same set as the third row except in CT systems: If the root locus crosses into the
reverse order. Then the c’s are computed from the b’s in
the same way the b’s are computed from the a’s. This
right half-plane the system goes unstable at that
continues until only three entries appear. Then the system gain.
is stable if
D(1) > 0 (!1)D D( !1) > 0 DT systems: If the root locus goes outside the
aD > a0 , b0 > bD!1 , c0 > cD!2 , !, s0 > s2 unit circle the system goes unstable at that gain.

7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 17 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 18

3
 Simulating DT Systems with DT Simulating DT Systems with DT
Systems Systems
The ideal simulation of a CT system by a DT system would have One approach to simulation is to make the impulse response of
the DT system’s excitation and response be samples from the CT the DT system be a sampled version of the impulse response of
system’s excitation and response. But that design goal is never the CT system.
achieved exactly in real systems at finite sampling rates.
h[ n ] = h( nTs )

With this choice, the response of the DT system to a DT unit

impulse consists of samples of the response of the CT system to a
CT unit impulse. This technique is called impulse-invariant
design.

7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 19 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 20

Simulating DT Systems with DT Simulating DT Systems with DT

Systems Systems
When h[ n ] = h( nTs ) the impulse response of the DT system is a In impulse-invariant design, even though the impulse response is a
sampled version of the impulse response of the CT system but the sampled version of the CT system’s impulse response that does not
unit DT impulse is not a sampled version of the unit CT impulse. mean that the response to samples from any arbitrary excitation
will be a sampled version of the CT system’s response to that
A CT impulse cannot be sampled. First, as a practical matter excitation.
the probably of taking a sample at exactly the time of occurrence
of the impulse is zero. Second, even if the impulse were sampled All design methods for simulating CT systems with DT systems
at its time of occurrence what would the sample value be? The are approximations and whether or not the approximation is a good
functional value of the impulse is not defined at its time of one depends on the design goals.
occurrence because the impulse is not an ordinary function.

7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 21 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 22

Sampled-Data Systems Sampled-Data Systems

An ADC simply samples a signal and produces numbers. A
Real simulation of CT systems by DT systems usually common way of modeling the action of a DAC is to imagine
sample the DT impulses in the DT signal which drive the DAC are
the excitation with an ADC, process the samples and then instead CT impulses of the same strength and that the DAC
produce a CT signal with a DAC. has the impulse response of a zero-order hold.

7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 23 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 24

4
 Sampled-Data Systems Sampled-Data Systems
Consider the response of the CT system not to the actual signal,
The desired equivalence between a CT and a DT system is x(t), but rather to an impulse-sampled version of it,
#
illustrated below. x! (t ) = $ x(nT )! (t " nT ) = x(t ) % f
n="#
s s s comb( fst )
The response is
$ $
y( t ) = h( t ) ! x " (t ) = h(t ) ! % x[m ]" (t # mT ) = % x[m]h(t # mT ) s s
m =#$ m=#$
where x[ n ] = x( nTs ) and the response at the nth multiple of Ts
"
is
y( nTs ) = # x[ m] h(( n ! m)Ts )
m=!"
The design goal is to make y d (t ) look as much like y c ( t ) as The response of a DT system with h[ n ] = h( nTs ) to the excitation,
possible by choosing h[n] appropriately. x[ n ] = x( nTs ) is
#
y[ n ] = x[n ] ! h[n ] = $ x[m ]h[n " m ]
m ="#

7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 25 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 26

Sampled-Data Systems Sampled-Data Systems

Modify the CT system to reflect the last analysis.
The two responses are equivalent in the sense that the values
at corresponding DT and CT times are the same.

Then multiply the impulse responses of both systems by Ts

7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 27 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 28

Sampled-Data Systems Sampled-Data Systems

In the modified CT system,
& $ ) $
y( t ) = x! ( t ) "Ts h( t ) = ( % x( nTs )! ( t # nTs ) + " h(t )Ts = % x( nTs ) h( t # nTs )Ts
' n=#$ * n=#$ Summarizing, if the impulse response of the DT system is
chosen to be Ts h( nTs ) then, in the limit as the sampling rate
In the modified DT system,
" " approaches infinity, the response of the DT system is exactly
y[ n ] = # x[m]h[n ! m] = # x[m ]T h((n ! m)T ) s s
the same as the response of the CT system.
m=!" m =!"

where h[ n ] = Ts h( nTs ) and h(t) still represents the impulse

Of course the sampling rate can never be infinite in practice.
response of the original CT system. Now let Ts approach zero. Therefore this design is an approximation which gets better as
# #
the sampling rate is increased.
lim y( t ) = lim
Ts !0 Ts !0
$ x(nT ) h(t " nT )T
s s s = & x(% ) h(t " % )d%
n="# "#

This is the response, y c ( t ) , of the original CT system.

7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 29 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 30

5
 Digital Filters Digital Filters
• Practical CT filters have infinite-duration
• Digital filter design is simply DT system impulse responses, impulse responses which
design applied to filtering signals never actually go to zero and stay there
• A popular method of digital filter design is to • Some digital filter designs produce DT filters
simulate a proven CT filter design with infinite-duration impulse responses and
• There many design approaches each of which these are called IIR filters
yields a better approximation to the ideal as • Some digital filter designs produce DT filters
the sampling rate is increased with finite-duration impulse responses and
these are called FIR filters

7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 31 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 32

Digital Filters Digital Filters

Impulse and Step Invariant Design

• Some digital filter design methods use time-

domain approximation techniques
• Some digital filter design methods use
frequency-domain approximation techniques

7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 33 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 34

Digital Filters Digital Filters

Impulse and Step Invariant Design Impulse and Step Invariant Design
Impulse invariant: Impulse invariant approximation of the one-pole system,
L!1 Z 1
H s ( s) h( t )
Sample
h[ n ] H z ( z) H s ( s) =
s+a
yields
1
Step invariant: H z ( z) =
1 1 ! e! aTs z !1
!
s H s ( s) L!1 h !1 (t )
Sample
h !1 [n ]
H s ( s)
s

z "1
!
Z z z
H (z ) H z ( z)
z !1 z
7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 35 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 36

6
 Digital Filters Digital Filters
Impulse and Step Invariant Design Impulse and Step Invariant Design
1 1
Let a be one and let Ts = 0.1 in H z ( z) = Step response of H z ( z) =
1 ! e! aTs z !1 1 ! e! aTs z !1 Notice scale difference

Digital Filter Digital Filter

Impulse Response Step Response

CT Filter CT Filter
Impulse Response Step Response

7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 37 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 38

Digital Filters Digital Filters

A CT step excitation is not an impulse. So what should the
Why is the impulse response exactly correspondence between the CT and DT excitations be now? If the
right while the step response is wrong? step excitation is sampled at the same rate as the impulse response
was sampled, the resulting DT signal is the excitation of the DT
This design method forces an equality system and the response of the DT system is the sum of the
between the impulse strength of a CT responses to all those DT impulses.
excitation, a unit CT impulse at zero,
and the impulse strength of the
corresponding DT signal, a unit DT
impulse at zero. It also makes the
impulse response of the DT system,
h[n], be samples from the impulse
response of the CT system, h(t).

7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 39 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 40

Digital Filters Digital Filters

If the excitation of the CT system were a sequence of CT unit Impulse and Step Invariant Design
impulses, occurring at the same sampling rate used to form h[n],
Impulse invariant approximation of
then the response of the DT system would be samples of the
response of the
CT system.
s
H s ( s) =
s2 + 400s + 2 ! 10 5

with a 1 kHz sampling rate

yields

z( z ! 0.9135 )
H( z ) =
z 2 ! 1.508z + 0.6703

7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 41 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 42

7
 Digital Filters Digital Filters
Impulse and Step Invariant Design Finite Difference Design
Step invariant approximation of Every CT transfer function implies a corresponding differential
equation. For example,
1 d
H s ( s) = ! ( y(t )) + a y( t ) = x(t )
s s + a dt
H s ( s) = 2
s + 400s + 2 ! 10 5
Derivatives can be approximated by finite differences.
with a 1 kHz sampling rate Forward Backward
( y( t )) ! [ ] [ ] ( y( t )) ! [ ] [ ]
yields d y n +1 " y n d y n " y n "1
dt Ts dt Ts
7.97 ! 10 "4 ( z " 1)
H z ( z) =
z2 " 1.509z + 0.6708 Central
d y[n + 1] " y[n " 1]
( y( t )) !
dt 2Ts
7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 43 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 44

Digital Filters Digital Filters

Finite Difference Design Finite Difference Design
Using a forward difference to approximate the derivative, Then

1 y[n + 1] " y[ n ] H s ( s) =
1 " 1 %
! H z ( z) = $ =
Ts
H s ( s) = ! + a y[n ] = x[n ]
s+a Ts s+a # s + a &' s( z )1 z ) (1) aTs )
Ts

A more systematic method is to realize that every s in a CT

transfer function corresponds to a differentiation in the time
domain which can be approximated by a finite difference.

Forward Backward Central

z "1 1" z "1 z " z "1
s! X( z) s! X( z) s! X( z)
Ts Ts 2Ts

7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 45 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 46

Digital Filters Digital Filters

Finite Difference Design Direct Substitution and Matched z-Transform Design
Finite difference approximation of Direct substitution and matched filter design use the relationship,
z = esTs to map the poles and zeros of an s-domain transfer
s function into corresponding poles and zeros of a z-domain
H s ( s) = transfer function. If there is an s-domain pole at a, the z-domain
s2 + 400s + 2 ! 10 5
pole will be at e aTs .

with a 1 kHz sampling rate Direct Substitution

yields s ! a " z ! eaTs

6.25 ! 10 "4 z( z " 1)

H( z ) =
z 2 " 1.5z + 0.625 Matched z-Transform
s ! a " 1 ! eaTs z !1
z = esTs
7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 47 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 48

8
 Digital Filters Digital Filters
Direct Substitution and Matched z-Transform Design Bilinear Transformation
Matched z-transform approximation of This method is based on trying to match the frequency response
of a digital filter to that of the CT filter. As a practical matter it
s is impossible to match exactly because a digital filter has a
H s ( s) = periodic frequency response but a good approximation can be
s2 + 400s + 2 ! 10 5
made over a range of frequencies which can include all the
expected signal power.
with a 1 kHz sampling rate
yields The basic idea is to use the transformation,

z( z ! 1) s!
1
ln( z ) e sTs ! z
H z ( z) = Ts
or
z2 ! 1.509z + 0.6708
to convert from the s to z domain.

7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 49 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 50

Digital Filters Digital Filters

Bilinear Transformation Bilinear Transformation
1
The straightforward application of the transformation, s ! ln( z )
would be the substitution, Ts
Truncating the exponential series at two terms yields the
H z ( z) = H s ( s) s! 1 ln ( z ) transformation,
Ts 1 + sTs ! z
But that yields a z-domain function that is a transcendental or
z "1
function of z with infinitely many poles. The exponential s!
function can be expressed as the infinite series, Ts
! This approximation is identical to the finite difference method
x2 x3 xk
ex = 1 + x + + +!= " using forward differences to approximate derivatives. This
2! 3! k =0 k! method has a problem. It is possible to transform a stable s-
and then approximated by truncating the series. domain function into an unstable z-domain function.

7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 51 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 52

Digital Filters Digital Filters

Bilinear Transformation Bilinear Transformation
The stability problem can be solved by a very clever
The bilinear transformation
modification of the idea of truncating the series. Express the
T has the quality that every
exponential as s s
e 2 point in the s plane maps into
e sTs = Ts " z a unique point in the z plane,
!s
e 2 and vice versa. Also, the left
Then approximate both numerator and denominator with a
half of the s plane maps into
truncated series.
the interior of the unit circle
sT
1+ s 2 z "1
in the z plane so a stable s-
2 "z s! domain system is
sT
1! s Ts z + 1 transformed into a stable z-
2 domain system.
This is called the bilinear transformation because both
numerator and denominator are linear functions of z.
7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 53 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 54

9
 Digital Filters Digital Filters
Bilinear Transformation Bilinear Transformation
The bilinear transformation is unique among the digital filter design
methods because of the unique mapping of points between the two Bilinear approximation of
complex planes. There is however a “warping” effect. It can be seen
by mapping real frequencies in the z plane (the unit circle) into s
corresponding points in the s plane. Letting z = e j! with Ω real, the H s ( s) =
s2 + 400s + 2 ! 10 5
corresponding contour in the s plane is

2 e j! " 1 2 # !% with a 1 kHz sampling rate

s= = j tan
Ts e j! + 1 Ts $ 2& yields
or
z2 ! 1
$ #Ts & H( z ) =
! = 2 tan"1 z 2 ! 1.52z + 0.68
% 2 '

7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 55 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 56

Digital Filters Digital Filters

FIR Filters FIR Filters

FIR digital filters are based on the FIR digital filters can also
idea of approximating an ideal approximate non-causal filters
impulse response. Practical CT by truncating the impulse
filters have infinite-duration impulse response both before time t = 0
response. The FIR filter and after some later time which
approximates this impulse by includes most of the signal
sampling it and then truncating it to energy of the ideal impulse
a finite time (N impulses in the response.
illustration).

7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 57 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 58

Digital Filters Digital Filters

FIR Filters FIR Filters
The design of an FIR
filter is the essence of Since this filter has no feedback paths its transfer function is of the
simplicity. It consists of form,
multiple feedforward
N !1
paths, each with a
H N ( z ) = " a m z !m
different delay and m =0
weighting factor and all
of which are summed to and it is guaranteed stable because it has N - 1 poles, all of which
form the response. are located at z = 0.
N "1
h N [ n ] = # am! [ n " m]
m=0

7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 59 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 60

10
 Digital Filters Digital Filters
FIR Filters FIR Filters
The effect of truncating an impulse response can be modeled by
multiplying the ideal impulse response by a “window” function. If The frequency-domain effect of truncating an impulse response is
a CT filter’s impulse response is truncated between t = 0 and t = T, to convolve the ideal frequency response with the transform of the
the truncated impulse response is window function.

!h(t ) , 0 < t < T $ HT ( f ) = H( f ) !W ( f )

h T (t ) = " % = h( t ) w( t )
#0 , otherwise &
where, in this case, If the window is a rectangle,
"t !T %
w(t ) = rect $$ 2'
'
$ T ' W( f ) = T sinc(Tf )e ! j"fT
# &

7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 61 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 62

Digital Filters Digital Filters

FIR Filters FIR Filters
! f # % j&fT
Let the ideal transfer function be H( f ) = rect e
" 2B $
The corresponding impulse response is
" " T $$
h( t ) = 2Bsinc& 2B t ! '
# # 2%%
The truncated impulse response is "t ! T $
" " T $$
h T (t ) = 2Bsinc& 2B t ! ' rect && 2'
# # 2 %% '
& T '
# %
The transfer function for the truncated impulse response is
! f # % j&fT
HT ( f ) = rect e 'T sinc(Tf )e % j&fT
" 2B $

7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 63 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 64

Digital Filters Digital Filters

FIR Filters FIR Filters

7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 65 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 66

11
 Digital Filters Digital Filters
FIR Filters FIR Filters
The effects of windowing a digital filter’s impulse response are
similar to the windowing effects on a CT filter.

"h[n ] , 0 ! n < N %
h N [n] = # & = h[ n ] w[n ]
$0 , otherwise '

H N ( j! ) = H( j! ) W( j!)

7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 67 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 68

Digital Filters Digital Filters

FIR Filters FIR Filters

7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 69 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 70

Digital Filters Digital Filters

FIR Filters FIR Filters
(windows continued)
The “ripple” effect in the frequency domain can be reduced # 2"n %
by windows of different shapes. The shapes are chosen to 3. Hamming w[n ] = 0.54 ! 0.46 cos , 0'n<N
$ N ! 1&
have DTFT’s which are more confined to a narrow range of
frequencies. Some commonly used windows are 4. Blackman
# 2"n % # 4 "n %
1' # 2"n % * w[n ] = 0.42 ! 0.5 cos + 0.08 cos , 0'n< N
1. von Hann w[n ] = )1 ! cos , 0-n<N $ N ! 1& $ N ! 1&
2( $ N ! 1& ,+
2. Bartlett 5. Kaiser
# 2n N !1
, 0"n" #
%N ! 1 N " 1% # N " 1% %
2 2

w[n ] = $
2 I0 ' ! a # " n"
%2 ! 2n , N ! 1 " n < N $ $ 2 & $ 2 & (&
& w[n ] =
N !1 N " 1%
I0#! a
2
$ 2 &
7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 71 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 72

12
 Digital Filters Digital Filters
FIR Filters FIR Filters
Windows Window Transforms Windows Window Transforms

7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 73 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 74

Standard Realizations Standard Realizations

• Realization of a DT system closely parallels Canonical
the realization of a CT system Summer
• The basic forms, canonical, cascade and
parallel have the same structure
• A CT system can be realized with integrators,
summers and multipliers
• A DT system can be realized with delays, Delay
summers and multipliers
Multiplier
7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 75 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 76

Standard Realizations Standard Realization

Parallel

Cascade

7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 77 7/21/06 M. J. Roberts - All Rights Reserved. Edited by Dr. Robert Akl 78

13