Signals

&
Systems
EEE & ECE Department
BITS-Pilani, Hyderabad campus

March 9, 2019 Signals & Systems 1

 Introduction
to
Analog Filters

March 9, 2019 Signals & Systems 2

 What is a filter?

March 9, 2019 Signals & Systems 3

 What is a filter?

March 9, 2019 Signals & Systems 4

Types of Filters- Broad classification

March 9, 2019 Signals & Systems 5

March 9, 2019 Signals & Systems 6
 Digital Signal Processing

March 9, 2019 Signals & Systems 7

 Ideal Low pass Prototype
analog filter

March 9, 2019 Signals & Systems 8

 Types of Ideal filters based on frequency
selection

1 1

0
0

March 9, 2019 Signals & Systems 9

In practice such Ideal filter is impossible to
realize, we seek reasonable approximations of
this filter

March 9, 2019 Signals & Systems 10

 Filter Basics
• Stop Band – the part of the frequency spectrum that
is attenuated by a filter

• Pass Band – part of the frequency spectrum that is

unaffected by a filter

• Filters are usually described in terms of their

“frequency responses,” e.g. lowpass, highpass,
bandpass, band reject (or notch)

Signals & Systems

 Essential Terminology
• Cutoff Frequency – point in the passband where
frequencies have been attenuated by 3 dB (½-power)

• Center Frequency – mid-point of the passband in a

Bandpass filter or the stopband of a Band Reject
filter

• Band Width – distance (in Hertz) between the ½-

power points of a Bandpass or Band Reject filter

Signals & Systems

 Important Terms
• Slope – rate of attenuation within the stop band,
measured in dB/Octave
CF
• Q – the Quality of a filter. Definition: Q 
BW

Q is often a more useful parameter than BW,

because the BW needs to vary with the CF to
keep the same “interval”

The higher the Q, the narrower the Band Width

Signals & Systems

Frequency Response Curves of
Practical Filters
Low Pass High Pass

Band Pass Band Reject

Signals & Systems

 A lowpass filter circuit

March 9, 2019 Signals & Systems 15

1

March 9, 2019 Signals & Systems 16

 00

450

900

March 9, 2019 Signals & Systems 17

 A highpass filter circuit

March 9, 2019 Signals & Systems 18

 1

March 9, 2019 Signals & Systems 19

 Spectral transformation

(1) Lowpass – Lowpass

(2) Lowpass – Highpass
(3) Lowpass – Bandpass
(4) Lowpass – Bandstop

March 9, 2019 Signals & Systems 20

 Problems
(1) (i) Find the frequency response
(amplitude and phase response) of a
system whose transfer function is given
by s  0. 1
H(s) 
s5

March 9, 2019 Signals & Systems 21

 solution
s  0.1 j  0.1
H(s)   H ( j ) 
s5 j  5

find H ( j)

  0.01
2
H ( j ) 
  25
2

March 9, 2019 Signals & Systems 22

   1   
H ( j )  tan 
1
  tan  
 0.1  5

can you guess what kind of filter is this?

____________

March 9, 2019 Signals & Systems 23

 Frequency Response

Magnitude Response Phase Response

March 9, 2019 Signals & Systems 24

(ii) find the system response also
for the given input f (t )  cos 2t

For the input f (t )  cos 2t

  2

2  0.01
2
H ( j 2)   0.372
2  25
2

March 9, 2019 Signals & Systems 25

  2  1  2 
H ( j )  tan 1
  tan  
 0.1  5
 87.10  27.80
 65.3 0

Now write the output of the system, y(t)

y(t )  0.372 cos(2t  65.30 )

March 9, 2019 Signals & Systems 26

 Lowpass filter
specifications

March 9, 2019 Digital Signal Processing 27

March 9, 2019 Digital Signal Processing 28
March 9, 2019 Digital Signal Processing 29
March 9, 2019 Digital Signal Processing 30
March 9, 2019 Signals & Systems 31
 Typical Magnitude Response of a
Lowpass Filter

March 9, 2019 Digital Signal Processing 32

 Gain in dB

G  10log H a  j  in dB
2

G  20log Ha  j in dB

March 9, 2019 Digital Signal Processing 33

March 9, 2019 Digital Signal Processing 34
Magnitude Specifications in another form

March 9, 2019 Digital Signal Processing 35

March 9, 2019 Digital Signal Processing 36
 Butterworth Filter

March 9, 2019 Signals & Systems 37

 Butterworth Filter

• Amplitude response

March 9, 2019 Signals & Systems 38

March 9, 2019 Signals & Systems 39
Butterworth lowpass filter is said
to have a maximally flat magnitude at
Ω=0
March 9, 2019 Digital Signal Processing 40
 Butterworth

H a  j  
2 1
2N

1   
 c 

H a  j  
1
2N

1   
 c 
March 9, 2019 Digital Signal Processing 41
 Gain in dB

10 log H a  j   10 log
2 1
2N

1   
 c 

20 log H a  j   20 log
1
2N

1   
 c 
March 9, 2019 Digital Signal Processing 42
 Butterworth

Case (i)   0 

Ha  j  1 (or)

20 log Ha  j  20 log 1  0 dB

March 9, 2019 Digital Signal Processing 43

 Butterworth

Case (ii)   c 

H a  j 
1
2 (or)

 1 
20 log H a  j  20 log    3.010 dB  3 dB
 2

Irrespective of order (N)

March 9, 2019 Digital Signal Processing 44
 Butterworth

Case (iii)   c ( for simplicity assume   )

N
H a  j  
1


2N  H a  j    
1     c 
 c 

(or)

March 9, 2019 Digital Signal Processing 45

 Butterworth

20 log H a  j   20 log
1
2N

1   
 c 


N

20 log H a  j   20 log  
 c 


20 log H a  j  20 N log  
 c 
March 9, 2019 Digital Signal Processing 46
 Butterworth

If   10c

March 9, 2019 Digital Signal Processing 47

 Butterworth


20 log H a  j  20 N log   
 c 

20 log Ha  j  20 N dB / decade

March 9, 2019 Digital Signal Processing 48

 Butterworth

If   2c

March 9, 2019 Digital Signal Processing 49

 Butterworth


20 log H a  j  20 N log   
 c 

20 log Ha  j  6 N dB / octave

March 9, 2019 Digital Signal Processing 50

Expression for the order (N)
of the filter

March 9, 2019 Digital Signal Processing 51

March 9, 2019 Digital Signal Processing 52
Digital Signal Processing 53
Expression for the order (N) of the filter

March 9, 2019 Digital Signal Processing 54

March 9, 2019 Digital Signal Processing 55
 (ii) Express in terms of k1 and k

March 9, 2019 Digital Signal Processing 56

March 9, 2019 Digital Signal Processing 57
 Problem

March 9, 2019 Digital Signal Processing 58

 Solution

March 9, 2019 Digital Signal Processing 59

March 9, 2019 Digital Signal Processing 60
 Butterworth Polynomial

March 9, 2019 Digital Signal Processing 61

 1
H a ( j ) 
2

  
1  C  2  2
  N
 p

March 9, 2019 Digital Signal Processing 62

where CN(Ω) is the Chebyshev Polynomial
of order N


cos( N cos 1
),   1
C N     1

cosh( N cosh )   1


where  
p

March 9, 2019 Digital Signal Processing 63

 Chebyshev Type – I

where
ε is the ripple factor

March 9, 2019 Signals & Systems 64

• Note: If ε = 0 there is no ripple but the filter
becomes All pass Filter

March 9, 2019 Signals & Systems 65

 Chebyshev Type – 2

March 9, 2019 Signals & Systems 66

 ELLIPTIC

March 9, 2019 Signals & Systems 67

March 9, 2019 Signals & Systems 68
 Standard Transfer Functions
• Butterworth
– Flat Pass-band.
– 20n dB per decade roll-off.
• Chebyshev
– Pass-band ripple.
– Sharper cut-off than Butterworth.
• Elliptic
– Pass-band and stop-band ripple.
– Even sharper cut-off.
• Bessel
– Linear phase response – i.e. no signal distortion in pass-
band.
 Filter Design by
Placement of
Poles and Zeros of H(s)

March 9, 2019 Signals & Systems 70

 Dependence of Frequency response on
Poles and Zeros Location
• A system Transfer Function expressed as

• The value of Transfer function at some

frequency s=p

March 9, 2019 Signals & Systems 71

 Dependence of Frequency response on
Poles and Zeros Location

March 9, 2019 Signals & Systems 72

 Dependence of Frequency response on
Poles and Zeros Location

March 9, 2019 Signals & Systems 73

 Dependence of Frequency response on
Poles and Zeros Location

March 9, 2019 Signals & Systems 74

March 9, 2019 Signals & Systems 75
 Gain Enhancement by a pole

March 9, 2019 Signals & Systems 76

 Gain Suppress by a Zero

March 9, 2019 Signals & Systems 77

 Gain Suppress by a Zero

March 9, 2019 Signals & Systems 78

 Lowpass Filters
• A typical lowpass filter has a maximum gain at
ω =0
• So we need to place a pole opposite the origin
(jω =0)

March 9, 2019 Signals & Systems 79

 Lowpass Filters
• The Transfer Function of this system is

• d is the distance from -ωc to a point jω then

March 9, 2019 Signals & Systems 80

 Lowpass Filters
• As ω increases, d increases and |H(jω)|
decreases monotonically

March 9, 2019 Signals & Systems 81

 Lowpass Filters

March 9, 2019 Signals & Systems 82