DSP First, 2/e

Lecture 13
Frequency Response
of FIR Filters
READING ASSIGNMENTS

▪ This Lecture:
▪ Chapter 6, Sections 6-1 to 6-5

▪ Other Reading:
▪ Recitation
▪ FREQUENCY RESPONSE EXAMPLES
▪ Next Lecture: Chap. 6, Sects. 6-6, 6-7 & 6-8

Aug 2016 © 2003-2016, JH McClellan & RW Schafer 3

LECTURE OBJECTIVES

▪ SINUSOIDAL INPUT SIGNAL

▪ DETERMINE the FIR FILTER OUTPUT

▪ FREQUENCY RESPONSE of FIR MAG

▪ PLOTTING vs. Frequency
PHASE
▪ MAGNITUDE vs. Freq
▪ PHASE vs. Freq jˆ jˆ jH ( e jˆ )
H (e ) = H (e ) e
Aug 2016 © 2003-2016, JH McClellan & RW Schafer 4
DOMAINS: Time & Frequency
▪ Time-Domain: “n” = time
▪ x[n] discrete-time signal
▪ x(t) continuous-time signal

▪ Frequency Domain (sum of sinusoids)

▪ Spectrum vs. f (Hz)
• ANALOG vs. DIGITAL
▪ Spectrum vs. omega-hat
▪ Move back and forth QUICKLY

Aug 2016 © 2003-2016, JH McClellan & RW Schafer 5

DIGITAL “FILTERING”

x(t) x[n] y[n] y(t)

A-to-D FILTER D-to-A

̂ ̂
▪ CONCENTRATE on the SPECTRUM
▪ SINUSOIDAL INPUT
▪ INPUT x[n] = SUM of SINUSOIDS
▪ Then, OUTPUT y[n] = SUM of SINUSOIDS

Aug 2016 © 2003-2016, JH McClellan & RW Schafer 7

SINUSOIDAL RESPONSE

▪ INPUT: x[n] = SINUSOID

▪ OUTPUT: y[n] will also be a SINUSOID
▪ Different Amplitude and Phase
▪ SAME Frequency

▪ AMPLITUDE & PHASE CHANGE

▪ Called the FREQUENCY RESPONSE

Aug 2016 © 2003-2016, JH McClellan & RW Schafer 8

DCONVDEMO: MATLAB GUI

Aug 2016 © 2003-2016, JH McClellan & RW Schafer 9

COMPLEX EXPONENTIAL
j jˆ n
x[n ] = Ae e −n
x[n] is the input signal—a complex exponential

M M
y[n ] =  bk x[n − k ] =  h[k ] x[n − k ]
k =0 k =0
FIR DIFFERENCE EQUATION

Aug 2016 © 2003-2016, JH McClellan & RW Schafer 10

COMPLEX EXP OUTPUT
▪ Use the FIR “Difference Equation”
M M
y[n ] =  bk x[n − k ] =  bk Ae e j jˆ ( n −k )

k =0 k =0

M jˆ ( − k ) 
=   bk e j jˆ n
 Ae e
 k =0 
j jˆ n
= H (ˆ ) Ae e
Aug 2016 © 2003-2016, JH McClellan & RW Schafer 11
FREQUENCY RESPONSE
▪ At each frequency, we can DEFINE
M
M
jˆ
bbkkee
H (He (ˆ)) ==  −−jj
ˆˆkk FREQUENCY
RESPONSE
kk==00

▪ Complex-valued formula
▪ Has MAGNITUDE vs. frequency
▪ And PHASE vs. frequency
▪ Notation: H ( e jˆ ) in place of H (ˆ )
Aug 2016 © 2003-2016, JH McClellan & RW Schafer 12
EXAMPLE 6.1
{bk } = { 1, 2, 1 }

H ( e jˆ ) = 1 + 2e − jˆ + e − j 2ˆ EXPLOIT

− jˆ jˆ − jˆ SYMMETRY
=e (e +2+e )
= e − jˆ ( 2 + 2 cos ˆ )

Since ( 2 + 2 cos ˆ )  0
jˆ
Magnitude is H ( e ) = ( 2 + 2 cos ˆ )
and Phase is  H ( e jˆ ) = −ˆ
Aug 2016 © 2003-2016, JH McClellan & RW Schafer 13
 PLOT of FREQ RESPONSE
{bk } = {1,2,1} jˆ
H (e ) = (2 + 2 cos  )e
ˆ − jˆ

RESPONSE at /3

− ̂ (radians)

Aug 2016 © 2003-2016, JH McClellan & RW Schafer 14

EXAMPLE 6.2
jˆ
Find y[n ] when H ( e ) is known
j / 4 j ( / 3) n
and x[n ] = 2e e
x[n] j̂ y[n]
H (e )

jˆ − jˆ
H (e ) = (2 + 2 cos ˆ )e
Aug 2016 © 2003-2016, JH McClellan & RW Schafer 15
 EXAMPLE 6.2 (answer)
j / 4 j ( / 3) n
Find y[n ] when x[n ] = 2e e
jˆ
Evaluate H (e ) at ˆ =  / 3
jˆ − jˆ
H (e ) = (2 + 2 cos ˆ )e
jˆ − j / 3
H (e ) = 3e @ ˆ =  / 3
y[n] = (3e − j / 3
) 2e j / 4
e j(  / 3)n
= 6e − j /12
e j(  / 3)n

Aug 2016 © 2003-2016, JH McClellan & RW Schafer 16

EXAMPLE: COSINE INPUT
jˆ
Find y[n ] when H ( e ) is known
 
and x[n ] = 2 cos( 3 n + 4 )
x[n] j̂ y[n]
H (e )
̂ ̂
jˆ − jˆ
H (e ) = (2 + 2 cos ˆ )e
Aug 2016 © 2003-2016, JH McClellan & RW Schafer 17
 EX: COSINE INPUT

Find y[n ] when x[n ] = 2 cos(  n + )

3 4
j ( n / 3+ / 4 ) − j ( n / 3+ / 4 )
2 cos( n+ 4 =e
3
) +e
 x[n ] = x1[n ] + x2 [n ]
j / 3 j ( n / 3+ / 4 )
Use
Linearity
y1[n ] = H ( e )e
− j / 3 − j ( n / 3+ / 4 )
y2 [ n ] = H ( e )e
Aug 2016
 y[n ] = y1[n ] + y2 [n ]
© 2003-2016, JH McClellan & RW Schafer 18
 EX: COSINE INPUT (ans-2)

Find y[n ] when x[n ] = 2 cos(  n + )

3 4
jˆ − jˆ
H (e ) = (2 + 2 cos ˆ )e
j / 3 j ( n / 3+ / 4 ) − j ( / 3) j ( n / 3+ / 4 )
y1[n ] = H ( e )e = 3e e
− j / 3 − j ( n / 3+ / 4 ) j ( / 3) − j ( n / 3+ / 4 )
y2 [ n ] = H ( e )e = 3e e
j ( n / 3− / 12 ) − j ( n / 3− / 12 )
y[n ] = 3e + 3e
 
 y[n ] = 6 cos( 3 n − 12 )
Aug 2016 © 2003-2016, JH McClellan & RW Schafer 19
MATLAB:
FREQUENCY RESPONSE

▪ HH = freqz(bb,1,ww)
▪ VECTOR bb contains Filter Coefficients
▪ SP-First: HH = freekz(bb,1,ww)

▪ FILTER COEFFICIENTS {bk}

M
H ( e ) =  bk e
jˆ − jˆ k

k =0
Aug 2016 © 2003-2016, JH McClellan & RW Schafer 20
LTI SYSTEMS

▪ LTI: Linear & Time-Invariant

▪ COMPLETELY CHARACTERIZED by:

▪ FREQUENCY RESPONSE, or
▪ IMPULSE RESPONSE h[n]

▪ Sinusoid IN -----> Sinusoid OUT

▪ At the SAME Frequency

Aug 2016 © 2003-2016, JH McClellan & RW Schafer 21

 Time & Frequency Relation

▪ Get Frequency Response from h[n]

▪ Here is the FIR case:

M M
H ( e ) =  bk e
jˆ − jˆ k
=  h[k ] e − jˆ k

k =0 k =0

IMPULSE RESPONSE

Aug 2016 © 2003-2016, JH McClellan & RW Schafer 22

BLOCK DIAGRAMS

▪ Equivalent Representations
x[n] y[n]
h[n ]

x[n] j̂ y[n]

H (e )
̂ ̂
Aug 2016 © 2003-2016, JH McClellan & RW Schafer 23
 UNIT-DELAY SYSTEM
jˆ
Find h[n ] and H (e ) for y[n ] = x[n − 1]
x[n]
 [n − 1] y[n]

{bk } = { 0, 1 }
j̂
H (e )
x[n]
− j̂ y[n]
e
̂ ̂
Aug 2016 © 2003-2016, JH McClellan & RW Schafer 24
 FIRST DIFFERENCE SYSTEM
jˆ
Find h[n ] and H ( e ) for the Difference
Equation : y[n ] = x[n ] − x[n − 1]
x[n] y[n]
 [n ] −  [n − 1]
j̂
H (e )
− j̂ y[n]
1− e
x[n]

Aug 2016 © 2003-2016, JH McClellan & RW Schafer 25

DLTI Demo with Sinusoids

FILTER = y[n]
x[n]

Aug 2016 © 2003-2016, JH McClellan & RW Schafer 26

 CASCADE SYSTEMS
▪ Does the order of S1 & S2 matter?
▪ NO, LTI SYSTEMS can be rearranged !!!
▪ WHAT ARE THE FILTER COEFFS? {bk}
▪ WHAT is the overall FREQUENCY
RESPONSE ?

 [n ] h1[n ]
S1 h1[n ] h2 [n ]
S2 h1[n ]  h2 [n ]

Aug 2016 © 2003-2016, JH McClellan & RW Schafer 27

CASCADE EQUIVALENT

jˆ jˆ jˆ

Aug 2016 H ( e
© 2003-2016, JH McClellan & RW Schafer
) = H 1 ( e ) H 2 ( e ) 28
CASCADE EQUIVALENT

▪ MULTIPLY the Frequency Responses

x[n] jˆ jˆ y[n]
H1 ( e ) H 2 (e )
x[n] j̂ y[n]
H (e )
EQUIVALENT
jˆ jˆ jˆ
SYSTEM H ( e ) = H1 ( e ) H 2 ( e )
Aug 2016 © 2003-2016, JH McClellan & RW Schafer 29
PLOT of FREQ RESPONSE
{bk } = {1,2,1}

̂
H (e jˆ ) = (2 + 2 cos ˆ )e − jˆ RESPONSE at /3

−Aug2016 
̂ (radians)
© 2003-2016, JH McClellan & RW Schafer 30
DCONVDEMO: MATLAB GUI

Aug 2016 © 2003-2016, JH McClellan & RW Schafer 31

DLTI Demo with Sinusoids

x[n] FILTER y[n]

Aug 2016 © 2003-2016, JH McClellan & RW Schafer 32