Signals and Systems Analysis Using

Transform Methods and MATLAB 3rd

Edition Roberts Solutions Manual
Visit to download the full and correct content document: https://testbankdeal.com/dow
nload/signals-and-systems-analysis-using-transform-methods-and-matlab-3rd-edition-
roberts-solutions-manual/
 Chapter 7 Exercise Solutions
Exercises With Answers
Orthogonality

1. Without using a calculator or computer find the dot products of (a) w1 and w -1 ,
(b) w1 and w -2 (c) w11 and w 37 , where

é W40 ù
ê ú
ê W4k ú j 2 p /N
wk = ê 2k
ú and WN = e
ê W4 ú
ê W43k úú
êë û

to show that they are orthogonal.

é ù é ù
ê (e )
jp /2 0
ú é ê (e )
jp /2 0
ú é
ê ú ê 1 ùú ê ú ê 1 ùú
ê (e )
jp /2 1
ú ê j ú ê (e )
jp /2 -1
ú ê -j ú
(a) w1 = ê ú=ê w -1 = ê ú=ê
ê (e )
jp /2 2
ú ê -1 úú ê (e )
jp /2 -2
ú ê -1 úú
ê ú êë - j úû ê ú êë j úû
ê
êë (e )
jp /2 3 ú
úû
ê
êë (e )
jp /2 -3 ú
úû
é 1 ù
ê ú
-j
w1H w -1 = é 1 - j -1 j ùê ú = 1- 1+ 1- 1 = 0
ë û ê -1 ú
ê j ú
êë úû

é ù é ù
ê (e )
jp /2 0
ú é ê (e )
jp /2 0
ú
ê ú ê 1 ùú ê jp /2 -2 ú é 1 ù
ê (e )
jp /2 1
ú ê j ú ê (e ) ú ê -1
ú
(b) w1 = ê ú=ê w -2 = ê ú=ê ú
(e ) ú ê -1 úú ê ( e jp /2 ) ú ê ú
jp /2 2 -4
ê 1
ê ú êë - j úû ê ú êë -1 ú
û
ê
êë (e )
jp /2 3 ú
úû
ê ( e jp /2 )-6
êë
ú
úû

Solutions 7-1
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 é 1 ù
ê ú
-1
w1H w -2 = é 1 - j -1 j ùê ú = 1+ j - 1- j = 0
ë ûê 1 ú
ê -1 ú
ë û
é ù é
ê (e )
jp /2 0
( e jp /2 ) ùú é
0
ú é ê
ê ú ê 1 ùú ê ú ê 1 ù
ê (e ) (e ) ú ê j ú
jp /2 11 jp /2 37
ú ê -j ú ê ú
(c) w11 = ê ú=ê w 37 = ê 74 ú
=
ú ê -1 úú
ê ( e jp /2 ) ( e jp /2 ) ú êê -1 ú
22
ê ú
ê ú êë j úû ê ú êë - j úû
ê ( e jp /2 ) 33
ú ê ( e ) úû
jp /2 111 ú
êë úû êë
é 1 ù
ê ú
j
w11 w37 = ë 1 j 1 - j û ê
H é ù ú = 1- 1+ 1- 1 = 0
ê -1 ú
ê ú
êë - j úû

é 11 ù
2. Find the projection p of the vector x = ê ú in the direction of the vector
ë 4 û
é -2 ù
y=ê ú.
ë 1 û

é -2 ù
éë 11 4 ùû ê ú
x yT
ë 1 û é -2 ù -18 é -2 ù é 36 / 5 ù é 2/5 ù
p= T y= ê ú= ê ú=ê ú = 18 ê ú
y y é -2 ù ë 1 û 5 ë 1 û ë -18 / 5 û ë -1 / 5 û
éë -2 1 ùû ê ú
ë 1 û

é 2 ù
ê ú
-3 ú
3. Find the projection p of the vector x = ê in the direction of the vector
ê 1 ú
ê 5 ú
ë û
é 1 ù
ê ú
ê j ú
y= . Then find the DFT of x and compare this result with X [ 3] y / 4 .
ê -1 ú
ê -j ú
êë úû

Solutions 7-2
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 é 1 ù
ê ú
j ú
éë 2 -3 1 5 ùû ê
ê -1 ú é 1 ù é 1 ù é 1 / 4 - j2 ù
ê ê ú
xH y êë - j úú ê j ú 1 - j8 ê j ú ê 2 + j / 4 ú
p= H y= û ê ú= ê ú=
y y é 1 ù ê -1 ú 4 ê -1 ú ê -1 / 4 + j2 ú
ê úê ú ê -j ú ê ú
j ú êë - j úû êë úû êë -2 - j / 4 úû
é 1 - j -1 j ù ê
ë ûê -1 ú
ê - j úúû
êë

X = {5,1+ j8,1,1- j8} Þ X [ 3] = 1- j8

é 1 ù é 1 / 4 - j2 ù
ê ú ê ú
1 - j8 ê j ú ê 2 + j / 4 ú
X [ 3] y/ 4 = =
4 ê -1 ú ê -1 / 4 + j2 ú
ê -j ú ê ú
êë úû êë -2 - j / 4 úû

Discrete Fourier Transform

4. A periodic discrete-time signal with fundamental period N = 3 has the values

x [1] = 7 , x [ 2 ] = -3 , x [ 3] = 1 . If , find the
magnitude and angle (in radians) of X [1] .

N -1 2
X [ k ] = å x [ n ] e- j 2 p kn/N = å x [ n ] e- j 2 p kn/3
n=0 n=0

2
X [1] = å x [ n ] e- j 2 p n/3 = x [ 0 ] + x [1] e- j 2 p /3 + x [ 2 ] e- j 4 p /3
n=0

X [1] = 1+ 7e- j 2 p /3 - 3e- j 4 p /3 = -1- j8.66 = 8.7178Ð - 1.6858

5. Using the direct summation formula find DFT harmonic function of d 10 [ n ] with
N = 10 and compare it with the DFT given in the table.
4
X[ k ] = å d 10 [ n ] e- j 2 p kn/10 = å d [ n] e
10
- j 2 p kn/10
= 1 = d1 [ k ]
n= 10 n=-5

Solutions 7-3
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
6. Without using a computer, find the forward DFT of the following sequence of data
and then find the inverse DFT of that sequence and verify that you get back the
original sequence.

{x [ 0 ],x [1],x [ 2 ],x [ 3]} = { 3, 4,1,- 2}

N -1
X [ k ] = å x [ n ] e- j 2 p nk /N
n=0

3
X [ 0 ] = å x [ n ] = 3 + 4 + 1- 2 = 6
n=0
3
X [1] = å x [ n ] e- jp n/2 = 3 - j4 - 1- j2 = 2 - j6
n=0
3
X [ 2 ] = å x [ n ] e- jp n = 3 - 4 + 1+ 2 = 2
n=0
3
X [ 3] = å x [ n ] e- j 3n/2 = 3 + j4 - 1+ j2 = 2 + j6
n=0

3
x [ 0 ] = (1 / 4 ) å X [ k ] = (1 / 4 ) [ 6 + 2 - j6 + 2 + 2 + j6 ] = 3
k=0
3
x [1] = (1 / 4 ) å X [ k ] e jp k/2 = (1 / 4 ) éë 6 + j ( 2 - j6 ) - 2 - j ( 2 + j6 ) ùû = 4
k=0
3
.
x [ 2 ] = (1 / 4 ) å X [ k ] e jp k
= (1 / 4 ) éë 6 - ( 2 - j6 ) + 2 - ( 2 + j6 ) ùû = 1
k=0
3
x [ 3] = (1 / 4 ) å X [ k ] e j 3p k/2 = (1 / 4 ) éë 6 - j ( 2 - j6 ) - 2 + j ( 2 + j6 ) ùû = -2
k=0

7. A signal x is sampled 8 times. The samples are

These samples are sent to a DFT algorithm and the output from that
algorithm is X, a set of 8 numbers .

(a) In terms of a,b,c,d,e, f , g and h what is X [ 0 ] ?

X[ 0 ] = a + b + c + d + e + f + g + h

Solutions 7-4
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 (b) In terms of a,b,c,d,e, f , g and h what is X [ 4 ] ?

X[ 4 ] = a - b + c - d + e - f + g - h

(c) If X [ 3] = 2 - j5 , what is the numerical value of X [ -3] ?

X [ -3] = X* [ 3] Þ X [ -3] = 2 + j5

(d) If X [ 5 ] = 3e- jp /3 , what is the numerical value of X [ -3] ?

X [ -3] = X [ -3 + 8 ] = X [ 5 ] = 3e- jp /3

(e) If X [ 5 ] = 9e j 3p /4 , what is the numerical value of X [ 3] ?

X [ 3] = X [ 3 - 8 ] = X [ -5 ] = X* [ 5 ] = 9e- j 3p /4

8. A discrete-time periodic signal with fundamental period N 0 = 6 has the values

x [ 4 ] = 3 , x [ 9 ] = -2 , x [ -1] = 1 , x [14 ] = 5 , x [ 24 ] = -3 , x [ 7 ] = 9 .

Also, .

(a) Find x [ -5 ] . x [ -5 ] = x [ -5 + 2 ´ 6 ] = x [ 7 ] = 9

(b) Find x [ 322 ] . x [ 322 ] = x [ 322 - 53 ´ 6 ] = x [ 4 ] = 3

(c) Find X [ 2 ] .

N -1 5 5
X [ k ] = å x [ n ] e- j 2 p kn/N = å x [ n ] e- jp kn/3 Þ X [ 2 ] = å x [ n ] e- j 2 p n/3
n=0 n=0 n=0

ìïx [ 0 + 4 ´ 6 ] + x [1+ 6 ] e- j 2 p /3 + x [ 2 + 2 ´ 6 ] e- j 4 p /3 üï
X[ 2 ] = í ý
ïî+ x [ 3 + 6 ] + x [ 4 ] e + x [ 5 - 6 ] e- j10 p /3
- j 8 p /3
ïþ

X [ 2 ] = -3 + 9e- j 2 p /3 + 5e- j 4 p /3 - 2 + 3e- j 8 p /3 + e- j10 p /3

Solutions 7-5
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 X [ 2 ] = 14.9332e-2.7862

9. Find the numerical values of the literal constants in

(a) .

Using

with m = 1 , n0 = -3 , n1 = 2 , N = 12

A = 40 , b = jp / 6 , c = 1 /12 , D = 5

(b)

X [ k ] = 5e- j 2 p k( 2 )/8 = 5e- jp k/2

A=5 a = -1/2

(c)

X [ k ] = jA (d 4 [ k + a ] - d 4 [ k - a ])

X [ k ] = e- j 2p k( -1)/4 - e- j 2p k(1)/4

X [ k ] = e jp k/2 - e- jp k/2 = j2sin (p k / 2 )

For even k X [ k ] = 0 , for X [ k ] = j2 and for

X [ k ] = - j2

Therefore X [ k ] = j2 (d 4 [ k + 1] - d 4 [ k - 1])

Solutions 7-6
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 A=2 a=1

10. The signal x [ n ] = 1 has a fundamental period N 0 = 1 .

(a) Find its DFT harmonic function using that fundamental period as the
representation time.

ìx [ n / 4 ] , n / 4 an integer
(b) Now let z [ n ] = í . Find the DFT harmonic
î0 , otherwise
function for z [ n ] using its fundamental period as the representation time.

ìï x é n / m ùû , n / m an integer
z éë n ùû = í ë , N ® mN Þ Z éë k ùû = X éë k ùû
îï0 , otherwise

\ Z [ k ] = X [ k ] = Nd N [ k ]

(c) Verify that z [ 0 ] = 1 and that z [1] = 0 by using the DFT representation of
z[ n] .
1
z [ n ] = å Z [ k ] e j 2p kn/N
N k= N
1 1+ 1+ 1+ 1
z [ 0 ] = å 1e0 = =1
4 k= 4 4
2p k
1 1 æ 0 j p2 3p
ö 1
z [1] = å ÷ø = 4 (1+ j - 1- j ) = 0
j j
jp
e 4
= ç e + e + e + e 2
N k= 4 4è

11. If x [ n ] = tri ( n / 3) and x p [ n ] = x [ n ] * d10 [ n ] and ,

what is the numerical value of X p [ 0 ] ?

X p [ 0 ] is the average value of x p [ n ] . The average value of any periodic

signal is the sum of the impulse strengths in one period, divided by the
period. In this case

1 4 1 3
X p [0] = å
10 n=-5
tri ( n / 3 ) = å tri( n / 3)
10 n=-3

Solutions 7-7
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 0 + 1 / 3 + 2 / 3 + 1+ 2 / 3 + 1 / 3 + 0
X p [0] = = 0.3
10
12. Find the DFT harmonic function of x [ n ] = ( u [ n ] - u [ n - 20 ]) * d 20 [ n ] using its
fundamental period as the representation time. There are at least two ways of
computing X [ k ] and one of them is much easier than the other. Find the easy
way.

x [ n ] is just the constant one. Therefore X [ k ] = 20d 20 [ k ] .

13. For each of these signals find the DFT over one fundamental period and show that
X [ N 0 / 2 ] is real.

(a) x [ n ] = ( u [ n + 2 ] - u [ n - 3]) * d 12 [ n ]
Using

X [ k ] = 5 drcl ( k /12,5 )
æ 6 ö sin ( 5p / 2 )
X [ 6 ] = X [ k ] = 5 drcl ç ,5 ÷ = =1 , Real.
è 12 ø sin ( p / 2 )

(b) x [ n ] = ( u [ n + 3] - u [ n - 2 ]) * d 12 [ n ]

X [ k ] = 5 drcl ( k /12,5 ) e jp k/6

æ 6 ö sin ( 5p / 2 ) jp
X [ 6 ] = 5 drcl ç ,5 ÷ e jp 6/6 = e = -1 , Real
è 12 ø sin (p / 2 )

(c) x [ n ] = cos (14p n /16 ) cos ( 2p n /16 )

x [ n ] = (1 / 2 ) éë cos (12p n /16 ) + cos (16p n /16 ) ùû

x [ n ] = (1 / 2 ) éë cos ( 2p n ( 3 / 8 ) ) + cos ( p n ) ùû

é ù
x [ n ] = (1 / 2 ) cos ( 2p n ( 3 / 8 )) + cos ( p n ) ú
ê
ê 1442443 12 4 4 3ú
ë14444 N 01 =8
4244444
N 02 =2
3û
N 0 =8

Solutions 7-8
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 The fundamental periods of the two individual functions that are multiplied
are both 16 but the fundamental period of the product is 8.

Alternate Solution:

From the table of DFT pairs

From the table of DFT properties

Using N = 16 and m = 1

X [ k ] = 4 (d 16 [ k - 7 ] + d 16 [ k + 7 ]) * (d [ k - 1] + d [ k + 1])

X [ k ] = 4 (d 16 [ k - 8 ] + d 16 [ k - 6 ] + d 16 [ k + 6 ] + d 16 [ k + 8 ])

This is the DFT based on N = 16 . If we now convert it to a DFT

based on N = 8 , we get

X [ k ] = 2 ( d 8 [ k - 4 ] + d 8 [ k - 3] + d 8 [ k + 3] + d 8 [ k + 4 ] )

because when we halve the fundamental period what was at the

second harmonic (k = 2) is now at the fundamental (k = 1) and all the
other even harmonics move to a new harmonic number that is half the
old one. (There cannot be any odd harmonics because the
fundamental period really is 8.) Also because d 8 [ k - 4 ] = d 8 [ k + 4 ] ,
this result can also be expressed as

Solutions 7-9
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 X [ k ] = 2 ( 2d 8 [ k - 4 ] + d 8 [ k - 3] + d 8 [ k + 3])
Then

X [ 4 ] = 2 ( 2d 8 [ 0 ] + d 8 [1] + d 8 [ 7 ]) = 4

Real.

æ 2p ( n - 3) ö
(d) x [ n ] = cos (12p n /14 ) cos ç ÷ø
è 14

The multiplication-convolution property is

{ }
X [ k ] = ( 49 /14 ) (d 14 [ k - 6 ] + d 14 [ k + 6 ]) * (d [ k - 1] + d [ k + 1]) e- j 3p k/7

X [ k ] = ( 49 /14 ) (d 14 [ k - 7 ] + d 14 [ k - 5 ] + d 14 [ k + 5 ] + d 14 [ k + 7 ]) e- j 3p k/7

X [ 7 ] = ( 49 /14 ) (d 14 [ 0 ] + d 14 [ 2 ] + d 14 [12 ] + d 14 [14 ]) e- j 3p = -98 /14 = -7

Real

Discrete-Time Fourier Transform Definition

14. From the summation definition, find the DTFT of

x [ n ] = 10 ( u [ n + 4 ] - u [ n - 5 ])

and compare with the DTFT table.

¥ ¥
X ( e jW ) =
4

å x [ n ] e- jWn = å 10 ( u [ n + 4 ] - u [ n - 5 ]) e- jWn = 10 å e- jWn

n=-¥ n=-¥ n=-4

Let m = n + 4 . Then

Solutions 7-10
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 1- e- j 9W
X( e ) = 10 å e
8 8
jW - jW( m-4 )
= 10e j 4W
åe - jWm
= 10e j 4W

1- e- jW
m=0 m=0

e- j 9W/2 e j 9W/2 - e- j 9W/2 sin ( 9W / 2 )

X ( e jW ) = 10e j 4 W = 10 = 90 drcl ( W / 2p ,9 )
e - jW/2
e - jW/2
-e - jW/2
sin ( W / 2 )

From the table,

15. From the definition, derive a general expression for the W form of the DTFT of
functions of the form

x [ n ] = a n sin ( W0 n ) u [ n ] , a < 1 .

Compare with the DTFT table.

¥ ¥
e jW0n - e- jW0n - jWn
X ( e jW ) = å a n sin ( W0 n ) u [ n ] e- jWn = åa n e
n=-¥ n=0 j2

) = (1 / j2 ) å { éëa e ( } , a <1
¥
X( e j W 0 -W )
ù - éa e- j( W0 +W ) ù
n n
jW
û ë û
n=0

é ù
X ( e jW ) = ( - j / 2 ) ê
1 1
j ( W 0 -W )
- - j ( W 0 +W ) ú
, a <1
ë 1- a e 1- a e û

é j2a e- jW sin ( W 0 ) ù
X( e ) = (- j / 2) ê
jW
2 - j 2W ú
, a <1
ë 1- 2a e cos ( W 0 ) + a e
- jW
û

a e jW sin ( W 0 )
X ( e ) = j 2W
jW
, a <1
e - 2a e jW cos ( W 0 ) + a 2

Solutions 7-11
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
16. Given the DTFT pairs below convert them from the radian frequency form to the
cyclic frequency form using W = 2p F without doing any inverse DTFT's.

(a)

(b)

Using the scaling property of the periodic impulse,

( ) ( )
d T a ( t - t0 ) = 1/ a d T / a ( t - t0 ) , we get

or

17. If x [ n ] = n 2 ( u [ n ] - u [ n - 3]) and ( )

, what is X e jW
W=0
?

¥ ¥ ¥
X ( e jW ) = å x [ n ] e- jWn Þ X ( e jW )W=0 = å x[n] = å n ( u [ n ] - u [ n - 3] )
2

n=-¥ n=-¥ n=-¥

X ( e jW )W=0 = å n 2 = 0 + 1+ 4 = 5
2

n=0

Forward and Inverse Discrete-Time Fourier Transforms

18. A signal is defined by

Solutions 7-12
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 x [ n ] = sinc ( n / 8 ) .

Graph the magnitude and phase of the DTFT of x [ n - 2 ] .

From the table of transform pairs,

or

19. If X ( F ) = 3 éëd 1 ( F - 1 / 4 ) + d 1 ( F +1 / 4 ) ùû - j4 éëd 1 ( F + 1 / 9 ) - d ( F - 1 / 9 ) ùû and

, what is the fundamental period of x [ n ]?

x [ n ] = 6 cos ( 2p n / 4 ) - 8 sin ( 2p n / 9 )

The fundamental period of x [ n ] is the least common multiple of the

fundamental periods of the sine and cosine which are added to form it.
The least common multiple of 4 and 9 is 36.

Solutions 7-13
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
20. If X ( F ) = d 1 ( F - 1 /10 ) + d 1 ( F + 1 /10 ) + d 1/16 ( F ) and , what is
the fundamental period of x [ n ]?

or

Fundamental period is the least common multiple of 10 and 16 which is

80.

21. Graph the magnitude and phase of the DTFT of

x [ n ] = ( u [ n + 4 ] - u [ n - 5 ]) * cos ( 2p n / 6 ) .
Then graph x [ n ] .

From the table,

and

X ( F ) = 9 drcl ( F,9 ) ´ (1 / 2 ) éëd 1 ( F -1 / 6 ) + d 1 ( F + 1 / 6 ) ùû

Solutions 7-14
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 Since both functions are periodic with period, one, at every impulse in the periodic
impulse the value of the Dirichlet function will be the same.

X ( F ) = ( 9 / 2 ) éë drcl (1 / 6,9 )d 1 ( F - 1 / 6 ) + drcl ( -1 / 6,9 )d 1 ( F + 1 / 6 ) ùû

X ( F ) = ( 9 / 2 ) drcl (1 / 6, 9 ) éëd 1 ( F - 1 / 6 ) + d 1 ( F + 1 / 6 ) ùû
14243
sin( 3p /2 )
=-2/9
9 sin( p /6 )

X ( F ) = - éëd 1 ( F - 1 / 6 ) + d 1 ( F + 1 / 6 ) ùû
Then, using

and, therefore,
x [ n ] = -2 cos ( 2p n / 6 )

x[n] |X( F )|
1
2

-1 1
F
-12 12
n Phase of X( F )|
p
-2 -1 1
F
-p

22. Graph the inverse DTFT of .

From the table,

and

Therefore

and

Therefore using multiplication-convolution duality,

Solutions 7-15
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 x [ n ] = (1 / 4 ) sinc ( n / 4 ) d 2 [ n ] .

x[n]
0.25

n
-16 16
-0.1

23. Let X e ( ) = 4p - j6p sin ( 2W ) .

jW
Its inverse DTFT is x [ n ]. Find the numerical value

of x éë n ùû for -3 £ n < 3 .

e j 2 W - e- j 2 W
X ( e ) = 4p - j6p
jW
= 4p - 3p ( e j 2 W - e- j 2 W )
j2

x [ n ] = 2d [ n ] - ( 3 / 2 ) (d [ n + 2 ] - d [ n - 2 ])

n -3 -2 -1 0 1 2
x [ n] 0 -3 / 2 0 2 0 3/2

24. ( ) ( )
A signal x éë n ùû has a DTFT , X F = 5drcl F,5 . What is its signal energy?

Using

(
x éë n ùû = 2 u éë n + 2 ùû - u éë n - 3ùû . )
¥

å 2 ( u éë n + 2 ùû - u éë n - 3ùû )
2
= 4 å 1 = 20
2
Ex =
n=-¥ n=-2

25. Find the numerical values of the literal constants.

(a)

Find A, W and B.

Solutions 7-16
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 n0 + n1 = 1 and n1 - n0 = 5 Þ n0 = -2 and n1 = 3

A = 10 , W = 2 , B = -2

(b) Find A and B.

2d 15 [ n - 3]( u [ n + 3] - u [ n - 4 ]) = 2d [ n - 3]

A = 2 , B = -3

(c) Find A, B and a .

A = 9 / 4 , B = 2, a = 2 / 3

Solutions 7-17
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 (d) Find A and B.

A = 40 , B = 10

(e)

A = 5 , a = 5 / 28 = 0.1786

(f)

A = - j8 , a = 6p = 18.85

(g)

Solutions 7-18
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 Using with
n0 = -3 , n1 = 2 and W = 2p F

A = 40 , b = j , c = 1 / 2p , D = 5

(h)

A = j196p , a = 1/ 2p , b = 7 , c = p / 6

(i)

éd 1 ( F + 1/ 16 ) ù ìïd 1 ( F + 1/ 16 ) - d 1 ( F - 1/ 16 ) üï
j42drcl ( F,5) ê ú e j 4p F = A í ý
êë -d 1 ( F - 1/ 16 ) úû ïî- j éëd 1 ( F + 1/ 16 ) + d 1 ( F - 1/ 16 ) ùû ïþ

éd 1 ( F + 1/ 16 ) ù é drcl ( -1/ 16,5) e- jp /4d 1 ( F + 1/ 16 ) ù

j42drcl ( F,5) ê úe j 4p F
= j42 ê ú
êë -d 1 ( F - 1/ 16 ) úû êë - drcl (1/ 16,5) e d 1 ( F - 1/ 16 ) úû
jp /4

Solutions 7-19
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 é 1- j ù
éd 1 ( F + 1/ 16 ) ù ê d 1 ( F + 1/ 16 ) ú
j42drcl ( F,5) ê ú e j 4p F = j42drcl (1/ 16,5) ê
2 ú
êë -d 1 ( F - 1/ 16 ) úû ê 1+ j ú
d 1 ( F - 1/ 16 ) ú
ê-
ë 2 û
éd 1 F +1/ 16 ù ( ) (
35.8006 éd 1 F + 1/ 16 - d 1 F - 1/ 16 ) ( ù )
(
j42drcl F,5 ê ) ú e j 4p F = j ê ú
êë -d 1 F -1/ 16 úû ( ) (
2 êë - jd 1 F + 1/ 16 - jd 1 F -1/ 16 úû ) ( )
éd 1 ( F + 1/ 16 ) ù ìïd 1 ( F + 1/ 16 ) - d 1 ( F - 1/ 16 ) üï
j42drcl ( F,5) ê úe j 4p F
= j25.3148 í ý
êë -d 1 ( F - 1/ 16 ) úû ïî- j éëd 1 ( F + 1/ 16 ) + d 1 ( F - 1/ 16 ) ùû ïþ
A = j25.3148
(j)

N 0 = 16

36 ìï(1- j ) d 1 ( F + 1/ 16 ) üï é e- j 2p n0 /16d 1 ( F - 1/ 16 ) ù
j í ý = ( A / 2 ) ê j 2p n0 /16 ú
2 ïî- (1+ j ) d 1 ( F - 1/ 16 ) ïþ êë +e d 1 ( F + 1/ 16 ) úû

j
36
(1- j ) = ( A / 2) e j 2 p n0 /16
, - j
36
(1+ j ) = ( A / 2) e - j 2 p n0 /16

2 2

Dividing the first equation by the second,

1- j e j 2p n0 /16 1- j j 4 p n /16
- = - j 2p n /16 Þ - =e 0
1+ j e 0
1+ j

e- jp /4
- jp /4
= e j 4p n0 /16 Þ -e- jp /2 = e j 4p n0 /16 Þ e jp /2 = e j 4p n0 /16 Þ n0 = 2
e

Solutions 7-20
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 j
36
(1- j ) = ( A / 2) e jp /4
= ( A / 2)
1+ j
2 2

26. Given the DTFT pair and

ìx [ n / 2 ] , n / 2 an integer
y[n] = í , find the magnitude and phase of Y e jW ( ) W=p /4
.
î0 , otherwise

Using

Y ( e jW )W=p /4 =
10 10
- jp /2
= = 8.575Ð - 0.5404 radians
1- 0.6e 1+ j0.6

27. Let , a phase-shifted triangle in the

range -1 / 2 < F < 1 / 2 that repeats that pattern periodically, with fundamental
period one. Also let

ìx [ n / 3] , if n / 3 is an integer
y[ n] = í
î0 , if n / 3 is not an integer

and let .

(a) Find the magnitude and angle (in radians) of X 0.3 . ( )

X ( 0.3) = 8 tri ( 0.6 ) e- j 2 p ( 0.3) = 3.2Ð - 1.885

(b) Find the magnitude and angle (in radians) of X ( 2.2 ) .

Taking advantage of the periodicity of X ( F )

X ( 2.2 ) = X ( 2.2 - 2 ) = X ( 0.2 ) = 8 tri ( 0.4 ) e- j 2 p ( 0.2 ) = 4.8Ð - 1.2566 .

Solutions 7-21
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 (c) What is the fundamental period of Y ( F ) ?
Use the property

Y ( F ) is a compressed version of X ( F ) and the compression factor is 3.

So the value of X ( F ) at F = 1 / 2 is the same as the value of Y ( F ) at
F = 1 / 6 and the fundamental period of Y ( F ) is therefore 1/3.

Alternate Solution: Y ( F ) = X ( 3F ) = 8 tri ( 6F ) e- j 6 p F * d 1 ( 3F )

Using the scaling property of the periodic impulse and the scaling property
of convolution

Y ( F ) = 3 ´ 8 tri ( 6F ) e- j 6 p F * (1 / 3) d 1/3 ( F ) = 8 tri ( 6F ) e- j 6 p F * d 1/3 ( F )

(d) Find the magnitude and angle (in radians) of Y ( 0.55 ) .

Using the fact that the fundamental period is 1/3,

Y ( 0.55 ) = Y ( 0.55 - 2 / 3) = Y ( -0.1167 )

Y ( 0.55 ) = 8 tri ( -6 ´ 0.1167 ) e j 6 p ´0.1167 = 2.3984Ð2.1997

28. A signal x [ n ] has a DTFT X ( F ) . Some of the values of x [ n ] are

n -2 -1 0 1 2 3 4 5 6
.
x [ n ] -8 2 1 -5 7 9 8 2 3

Let Y ( F ) = X ( 2F ) with . Find the numerical values of

y [ n ] for -2 £ n < 4 .

Using

Solutions 7-22
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 n -2 -1 0 1 2 3
y[ n] 2 0 1 0 -5 0

29. Using the differencing property of the DTFT and the transform pair,

find the DTFT of (1 / 2 ) (d [ n + 1] + d [ n ] - d [ n - 1] - d ( n - 2 ) ) . Compare it with

Fourier transform found using the table.

The first backward difference of tri ( n / 2 ) is

(1 / 2 ) (d [ n + 1] + d [ n ] - d [ n - 1] - d [ n - 2 ]) . Applying the differencing
property,

Other route to the DTFT:

Check.

30. A signal is described by

ìln ( n + 1) , 0 £ n < 10
ï
x [ n ] = í- ln ( -n + 1) , - 10 < n < 0
ï0 , otherwise
î

Solutions 7-23
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 Graph the magnitude and phase of its DTFT over the range -p £ W < p .

This can only be done numerically using the DFT to approximate the DTFT. The
approximation is

N -1
X ( e jW )W®2 p k/N = X ( e j 2 p k/N ) @ å x [ n ] e- j 2 p nk/N
n=0

n0 = -9 ;
n1 = 9 ;
n = [n0:n1]' ; % Vector of discrete-time indices for x[n]
N = length(n) ; % Number of points used to represent x[n]
% Compute x[n]
x = [-log(-n(1:9)+1);log(n(10:end)+1)] ;
% Zero-pad the time domain function to increase frequency domain resolution
padFac = 32 ;
Npad = N*padFac ;
nextra = [n1+1:n1+(Npad-N)]' ;
npad = [n;nextra] ;
xpad = [x;zeros(Npad-N,1)] ;
Npad = length(npad) ; % Number of values used to represent x[n]
Xpad = fft(xpad) ; % DFT of x is the approximation to DTFT of x
kpad = [0:Npad-1]' ; % Harmonic numbers in the range 0 through N-1
Xpad = Xpad.*exp(-j*2*pi*kpad*n0/Npad) % Compensate for the fact that x starts
at n=-9, not n=0
% Rearrange k and X for a range of frequencies centered at zero
kpad0 = -floor(Npad/2) ;
kpad = kpad0 + [0:Npad-1]' ;
Wpad = 2*pi*kpad/Npad ; % Vector of discrete-time radian frequencies
Xpad = fftshift(Xpad) ; % Rearrange X for frequencies centered at zero

% Graph the results

figure('Position',[20,20,1500,1500],'PaperPosition',[0.5,0.5,10,10]) ;
subplot(3,1,1) ;
ptr = stem(n,x,'k','filled') ;
set(ptr,'LineWidth',2,'MarkerSize',4) ; grid on ;
xlabel('\itn','FontName','Times','FontSize',24) ;
ylabel('x[{\itn}]','FontName','Times','FontSize',24) ;
set(gca,'FontName','Times','FontSize',18) ;
subplot(3,1,2)
ptr = plot(Wpad,abs(Xpad),'k') ;
set(ptr,'LineWidth',2,'MarkerSize',4) ; grid on ;
xlabel('\Omega','FontName','Times','FontSize',24) ;
ylabel('|X({\ite}^{{\itj}\Omega})|','FontName','Times','FontSize',24) ;
set(gca,'FontName','Times','FontSize',18) ;
subplot(3,1,3)
ptr = plot(Wpad,angle(Xpad),'k') ;
set(ptr,'LineWidth',2,'MarkerSize',4) ; grid on ;
xlabel('\Omega','FontName','Times','FontSize',24) ;
ylabel('Phase of X({\ite}^{{\itj}\Omega})','FontName','Times','FontSize',24) ;
set(gca,'FontName','Times','FontSize',18) ;

Solutions 7-24
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 4

x[n]
0

-2

-4
-10 -8 -6 -4 -2 0 2 4 6 8 10
n
30
|X(ej )|

20
W

10

0
-4 -3 -2 -1 0 1 2 3 4
W
2
Phase of X(ej )
W

-1

-2
-4 -3 -2 -1 0 1 2 3 4
W

Exercises Without Answers in Text

Discrete Fourier Transform

é 3 ù é 9 ù
ê ú ê ú
ê -j ú ê -4 + j ú
31. If x = and y = find x H y.
ê 3+ j ú ê -5 ú
ê ú ê 2 + j7 ú
ë 7 û ë û
é 9 ù
ê ú
-4 + j ú
x H y = éë 3 j 3- j 7 ùû ê
ê -5 ú
ê 2 + j7 ú
ë û

x H y = 27 - j4 - 1- 15 + j5 + 14 + j49 = 25 + j50 = 5.5902e j1.107

32. Fill in the blanks with correct numbers for this DFT harmonic function of a real-
valued signal with N = 8 .
Solutions 7-25
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 k 0 1 2 3 4 5 6 7
X [ k ] 5 ________ 2 - j7 4 + j2 -3 ________ ________ 9 + j4

k 11 -9 26 -47
X [ k ] ________ ________ ________ ________

k 0 1 2 3 4 5 6 7
X [ k ] 5 9 - j4 2 - j7 4 + j2 -3 4 - j2 2 + j7 9 + j4

k 11 -9 26 -47
X [ k ] 4 + j2 9 + j4 2 - j7 9 - j4

33. A discrete-time signal x [ n ] is periodic with period 8. One period of its DFT
harmonic function is

{ X [ 0 ],L, X [ 7 ]} = {3, 4 + j5,-4 - j3,1+ j5,-4,1- j5,-4 + j3, 4 - j5} .

(a) What is the average value of x [ n ]?

The average value of x [ n ] is

é N -1
- j 2 p kn/N ù
N -1
X [ 0 ] / N = ê(1 / N ) å x [ n ] e ú = (1 / N ) å x [ n ] .
ë n=0 û k=0 n=0

In this case 3/8.

(b) What is the signal power of x [ n ]?

From Parseval's theorem,

(1 / N ) å x [ n ] = (1 / N 2 ) å X[ k ]
2 2

n= N k= N

(1 / N ) å x [ n ] = (1 / 64 )[ 9 + 41+ 25 + 26 + 16 + 26 + 25 + 41] = 3.626

2

n= N

Solutions 7-26
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 (c) Is x [ n ] even, odd or neither?

This signal is neither even nor odd because its DFT harmonic function is
neither purely real-valued nor purely imaginary-valued.

34. A set of samples x [ n ] from a signal is converted to a set of numbers X [ k ]

by using the DFT.

(a) If { x [ 0 ] ,x [1] ,x [ 2 ] ,x [ 3] ,x [ 4 ]} = {2,8,-3,1,9} , find the numerical value

of X [ 0 ] .
4
X [ 0 ] = å x [ n ] = 2 + 8 - 3 +1+ 9 = 17
n=0

(b) If x [ n ] consists of 24 samples taken from exactly 3 periods of a sinusoid

at a sampling rate which is exactly 8 times the frequency of the sinusoid,
two values of X [ k ] in the range 0 £ k < 24 are not zero. Which ones are
they?

If exactly three periods of a sinusoid are sampled, the sinusoid must be at

the third harmonic of the fundamental frequency representable by the
DFT. Therefore X [ 3] is not zero. For any DFT X [ k ] = X* [ -k ].
Therefore X [ -3] is not zero. The DFT is periodic. In this case the period
is 24. Therefore X [ -3] = X [ -3 + 24 ] = X [ 21] and X [ 21] is not zero.

(c) If x [ n ] consists of 7 samples, all of which are the same, -5, in the range
0 £ k < 7 which X [ k ] values are zero?

If the set of values of x [ n ] is repeated periodically we can say that

x [ n ] = -5 for all n. Then using

we see that, in the range 0 £ k < 7 , the only non-zero value is X [ 0 ] .

Therefore the values of X [ k ] that are zero are .

(d) If { x [ 0 ],x [1],x [ 2 ] ,x [ 3]} = { a,b,0,b} and

{ X [ 0 ], X [1], X[ 2 ], X[ 3]} = { A, B,0, B }

*

Solutions 7-27
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 how are a and b related to each other? Express B in terms of a and b.

X éë 2 ùû = 0 = å x éë n ùû e- jp n = å x éë n ùû ( -1) = a - b + 0 - b Þ a = 2b
3 3
n

n=0 n=0

X éë1ùû = B = å x éë n ùû e- jp n/2 = å x éë n ùû ( - j ) = a - jb + 0 + jb = a
3 3
n

n=0 n=0

35. If and , find the

numerical values of x 2 [ 2 ],x 2 [ 4 ],x 2 [ 8 ] and x 2 [ 204 ] .

According to the time-scaling property

In this case m = 4 and

ì10 cos ( 2p n / 32 ) , n / 4 an integer ü

x2 [ n] = í ý
î0 , otherwise þ

ì10 cos ( 4p / 32 ) , 2 / 4 an integer ü

x2 [2] = í ý=0
î0 , otherwise þ

ì10 cos ( 8p / 32 ) , 4 / 4 an integer ü

x2 [ 4 ] = í ý = 7.07
î0 , otherwise þ

ì10 cos (16p / 32 ) , 8 / 4 an integer ü

x2 [8] = í ý=0
î0 , otherwise þ

ì10 cos ( 408p / 32 ) , 204 / 4 an integer ü

x 2 [ 204 ] = í ý = -7.07
î0 , otherwise þ

Solutions 7-28
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
36. A periodic discrete-time signal x [ n ] is exactly described for all discrete time by its
DFT

X [ k ] = 8 (d 8 [ k - 1] + d 8 [ k + 1] + j2d 8 [ k + 2 ] - j16d 8 [ k - 2 ]) e- jp k/4 .

(a) Write a correct analytical expression for x [ n ] in which -1 , “j”, does not
appear.

Therefore
æ 2p ( n - 1) ö æ 2p ( n - 1) ö
x [ n ] = 2 cos ç ÷ + 4 sin ç ÷ø
è 8 ø è 4

(b) What is the numerical value of x [ n ] at n = -10 ?

æ 2p ( -10 - 1) ö æ 2p ( -10 - 1) ö
x éë -10 ùû = 2cos ç ÷ + 4sin ç ÷
è 8 ø è 4 ø

æ 11p ö æ 11p ö
x éë -10 ùû = 2cos ç - + 4sin çè - 2 ÷ø
è 4 ÷ø

æ 11p ö æ 11p ö
x éë -10 ùû = 2cos ç ÷ - 4sin ç
è 4 ø è 2 ÷ø

æ 3p ö æ 3p ö æ 1 ö
x éë -10 ùû = 2cos ç ÷ - 4sin ç ÷ = 2 ç -
è 4ø è 2ø è ÷ - 4 ( -1)
2ø

Solutions 7-29
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 2 4 2 - 2 3.657
x [ -10 ] = 4 - = = = 2.586
2 2 1.414

37. A discrete-time signal x [ n ] with fundamental period N 0 = 4 has a DFT X [ k ] .

k 0 1 2 3
X [ k ] 4 2 - j3 1 2 + j3

(a) Find X [ -5 ] .

X [ -5 ] = X [ -5 + 4 + 4 ] = X [ 3] = 2 + j3

(b) Find X [ 22 ] .

X [ 22 ] = X [ 22 - 4 - 4 - 4 - 4 ] = X [ 2 ] = 1

(c) Find x [ 0 ].

1 N -1
x[ n] = å
N k=0
X [ k ] e j 2 p kn/N

1 3 4 + 2 - j3 +1+ 2 + j3 9
x[ 0] = å
4 k=0
X[ k ] =
4
= = 2.25
4

(d) Find x [ 3] .

1 3 4 + ( 2 - j 3) e j 3p /2 + e j 3p + ( 2 + j3) e j 9 p /2
x [ 3] = å X [ k ] e j 3p k/2
=
4 k=0 4

4 + ( 2 - j3) ( - j ) + ( -1) + ( 2 + j3) j

x [ 3] =
4

4 - j2 - 3 - 1+ j2 - 3 3
x [ 3] = = - = -0.75
4 4

38. A signal x ( t ) is sampled 4 times and the samples are { x [ 0 ],x [1],x [ 2 ],x [ 3]} . Its
DFT harmonic function is { X [ 0 ], X[1], X[ 2 ], X[ 3]} . X[ 3] can be written as

Solutions 7-30
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 X [ 3] = a x [ 0 ] + b x [1] + c x [ 2 ] + d x [ 3] . What are the numerical values of a,b,c and
d?

N -1 3
X [ k ] = å x [ n ] e- j 2 p nk/N Þ X [ 3] = å x [ n ] e- j 3p n/2 = x [ 0 ] + j x [1] - x [ 2 ] - j x [ 3]
n=0 n=0

a = 1 , b = j , c = -1 , d = - j

39. The DFT harmonic function X éë k ùû of using

N = 4 can be written in the form

( )(
X éë k ùû = A d 4 éë k - a ùû + d 4 éë k + a ùû d 4 éë k + b ùû - d 4 éë k - b ùû . )
Find the numerical values of A, a and b.

From the tables,

Therefore

or

From the tables,

Therefore

Then, using

( ) (
X éë k ùû = 10 d 4 éë k - 2 ùû + d 4 éë k + 2 ùû ´ j6 d 4 éë k + 1ùû - d 4 éë k -1ùû )
( )(
X éë k ùû = j60 d 4 éë k - 2 ùû + d 4 éë k + 2 ùû d 4 éë k + 1ùû - d 4 éë k - 1ùû )
A = j60 , a = 2 , b = 1
Solutions 7-31
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
40. Four data points {x éë0 ùû ,x éë1ùû ,x éë2 ùû , x éë3ùû} are converted by the DFT into four
corresponding data points { X éë 0 ùû ,X éë1ùû ,X éë 2 ùû ,X éë3ùû} . If x éë 0 ùû = 2 , x éë1ùû = -3,
x éë 3ùû = 7 , X éë1ùû = -3 + j10 , X éë 2 ùû = 3, find the numerical values of the missing
data, x éë 2 ùû , X éë0 ùû and X éë 3ùû . (Be careful to observe which symbols are lower
and upper case.)

N -1
X éë k ùû = å x éë n ùû e- j2p kn/ N
n=0

3
X éë1ùû = -3+ j10 = å x éë n ùû e- jp n/2 = 2 + j3- x éë 2 ùû + j7
n=0

-3+ j10 = 2 + j3- x éë 2 ùû + j7 Þ x éë 2 ùû = 3+ 2 - j10 + j3+ j7 = 5

X éë3ùû = X éë -1ùû = X * éë1ùû = -3- j10

X éë0 ùû = 2 - 3+ 5 + 7 = 11

41. Demonstrate with a counterexample that the fundamental period of a discrete-time

function and the fundamental period of its DFT harmonic function are not
necessarily the same.

The fundamental period of x [ n ] = d N [ n ] is N. Its DFT harmonic function is

X [ k ] = 1 with fundamental period one, regardless of N.

42. In Figure E-42 is graphed exactly one period of a periodic function x éë n ùû .

Its harmonic function X éë k ùû (with N = N 0 ) can be written in the form

(
X éë k ùû = A Be jbk - e jck + De jdk )
Find the numerical values of the constants.

Solutions 7-32
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 10

x[n]
0

-2

-4

-6

-8

-10
-10 -8 -6 -4 -2 0 2 4 6 8 10
n

Figure E-42

x éë n ùû = 6d 20 éë n + 8 ùû - 2d 20 éë n - 1ùû + 8d 20 éë n - 9 ùû

Using and

we get

X éë k ùû = 6e j16p k /20 - 2e- j 2p k /20 + 8e- j18p k /20

(
X éë k ùû = 2 3e j 4p k /5 - e- jp k /10 + 4e- j9p k /10 )
A = 2, B = 3 , b = 4p / 5, c = -p / 10 , D = 4 , d = -9p / 10

43. Associate each discrete-time signal in Figure E-43-1 with its corresponding DFT
magnitude in Figure E-43-2.

1 2 3
2 4 2
1.5 3 1
x[n]

x[n]

x[n]

1 2 0
0.5 1 -1

0 0 -2
0 5 10 15 0 5 10 15 0 5 10 15
n n n

4 5 6
3 4 2

2 1.5
2
x[n]

x[n]

x[n]

0 1
1
-2 0.5

0 -4 0
0 5 10 15 0 5 10 15 0 5 10 15
n n n

Solutions 7-33
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 Figure E-43-1

1-H 2-B 3-A

4-K 5-L 6-C
A B C
15 30 30

10 20 20

|X[k]|

|X[k]|

|X[k]|
5 10 10

0 0 0
0 5 10 15 0 5 10 15 0 5 10 15
k k k
D E F
8 30 15
6
20 10
|X[k]|

|X[k]|

|X[k]|
4
2 10 5

0 0 0
0 5 10 15 0 5 10 15 0 5 10 15
k k k
G H I
8 2 20
6
|X[k]|

|X[k]|

|X[k]|
4 1 10
2
0 0 0
0 5 10 15 0 5 10 15 0 5 10 15
k k k
J K L
40 30
8
6 20
|X[k]|

|X[k]|

|X[k]|

20 4
10
2
0 0 0
0 5 10 15 0 5 10 15 0 5 10 15
k k k

Figure E-43-2

44. Associate each discrete-time signal in Figure E-44-1 with its corresponding DFT
magnitude in Figure E-44-2.

Solutions 7-34
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 Figure E-44-1

1-B 2-A 3-I 4-M 5-O

6-D 7-J 8-G 9-C 10-H

Figure E-44-2

Forward and Inverse Discrete-Time Fourier Transforms

45. Given the DTFT pairs below convert them from the cyclic frequency form to the
radian frequency form using W = 2p F without doing any inverse DTFT's.

(a)

Using the scaling property of the periodic impulse,

( ) ( )
d T a ( t - t0 ) = 1/ a d T / a ( t - t0 ) , we get

(b)

Solutions 7-35
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 If W = 2p F then W 0 = 2p F0 and

Using the scaling property of the periodic impulse,

( ) ( )
d T a ( t - t0 ) = 1/ a d T / a ( t - t0 ) , we get

(c)

Using the scaling property of the periodic impulse,

( ) ( )
d T a ( t - t0 ) = 1/ a d T / a ( t - t0 ) , and the scaling property of convolution,
() () () ( ) ( ) ( )
if y t = x t * h t then y at = a x at * h at , we get

or

46. Find the DTFT of each of these signals:

x [ n ] = (1 / 3) u [ n - 1]
n
(a)
¥ ¥ ¥
X ( e jW ) = å x [ n ] e- jWn = å (1 / 3) u [ n -1] e- jWn = å (1 / 3) e- jWn
n n

n=-¥ n=-¥ n=1

Let m = n - 1 . Then

Solutions 7-36
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 m+1
¥ ¥
æ e- jW ö
X(e ) = å (1 / 3) - jW( m+1)
= åç
jW m+1
e ÷
m=0 m=0 è 3 ø

m
e- jW ¥ æ e- jW ö e- jW e- jW
X( e ) =
1
jW
å
3 m=0 çè 3 ÷ø
= =
3 1- e- jW / 3 3 - e- jW
Alternate Solution:
x [ n ] = (1 / 3) u [ n - 1] = (1 / 3) u [ n ] - d [ n ]
n n

Using
and

1
x [ n ] = (1 / 3) u [ n ] - (1 / 3)d [ n ] = -1
n

e- jW
1-
3

æ e- jW ö e- jW
1- ç 1- ÷
è 3 ø e- jW
x[ n] = = 3 =
1- e- jW / 3 1- e- jW / 3 3 - e- jW

Second Alternate Solution:

x [ n ] = (1 / 3) (1 / 3) u [ n - 1]
n-1

e- jW
X ( e jW ) =
1 1 - jW
e =
3 1- e- jW / 3 3 - e- jW

x [ n ] = sin ( p n / 4 ) (1 / 4 ) u [ n - 2 ]
n
(b)

sin ( p n / 4 ) = sin ( 2p n / 8 ) = cos ( 2p ( n - 2 ) / 8 )

x [ n ] = (1 / 4 ) cos ( 2p ( n - 2 ) / 8 ) (1 / 4 ) u[n - 2]
2 n-2

1- (1 / 4 ) cos (p / 4 ) e- jW
X ( e jW ) = (1 / 4 ) e- j 2W
2

1- (1 / 2 ) cos (p / 4 ) e- jW + (1 /16 ) e- j 2 W

Solutions 7-37
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 X(e ) =
jWe- j 2W ( 2 / 8) e
1- - jW

16 1 - ( 2 / 4 ) e + (1 /16 ) e
- jW - j 2W

Alternate Solution:

x[ n] =
e jp n/4 - e- jp n/4
j2 j2
n
(
(1 / 4 )n u [ n - 2 ] = ( e jp /4 / 4 ) - ( e- jp /4 / 4 ) u [ n - 2 ]
1 n
)
x[ n] =
1
j2
( (
e j /4 / 4 ) ( e j /4 / 4 ) - ( e- jp /4 / 4 ) ( e- j/4 / 4 ) u [ n - 2 ]
2 n-2 2 n-2
)

1 æ æ e jp /4 ö ö
2 2
e- j 2W æ e- jp /4 ö e- j 2W
X( e ) = ç ç
jW
- ÷
j2 è è 4 ÷ø 1- ( e jp /4 / 4 ) e- jW çè 4 ÷ø 1- ( e- jp /4 / 4 ) e- jW ø

X( e ) =
e- j 2W ( e / 4 ) 1- ( e
jW
jp /4 2
- jp /4
( )
/ 4 ) e- jW - ( e- jp /4 / 4 ) 1- ( e jp /4 / 4 ) e- jW
2
( )
j2 ( )(
1- ( e jp /4 / 4 ) e- jW 1- ( e- jp /4 / 4 ) e- jW )
e- j 2W ( e ) - (e ) (e / 4 ) e- jW - ( e- jp /4 ) + ( e- jp /4 ) ( e jp /4 / 4 ) e- jW
jp /4 2 jp /4 2 - jp /4 2 2

X( e ) =
jW

j32 1- ( e jp /4
/ 4 + e- jp /4 / 4 ) e- jW + e- j 2W /16

jp /2
e- j 2W e - e
- jp /2
- ( e jp /4 / 4 ) e- jW + ( e- jp /4 / 4 ) e- jW
X( e ) =
jW

j 32 1- (1 / 2 ) cos ( p / 4 ) e- jW + (1 /16 ) e- j 2W

e- j 2W j2 - (1 / 4 ) ( e jp /4 - e- jp /4 ) e- jW
X ( e jW ) =
j32 1- (1 / 2 ) cos (p / 4 ) e- jW + (1 /16 ) e- j 2W

e- j 2W 1- (1 / 4 ) sin (p / 4 ) e- jW
X( e ) =
jW

16 1- (1 / 2 ) cos ( p / 4 ) e- jW + (1 /16 ) e- j 2W
Solutions 7-38
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 X(e ) =
jWe- j 2 W 1- ( 2 / 8) e - jW

16 1- ( 2 / 4 ) e + (1 /16 ) e
- jW - j 2W

(c) x [ n ] = sinc ( 2p n / 8 ) *sinc ( 2p ( n - 4 ) / 8 )

Using

X ( F ) = ( 8 / 2p ) éë rect ( 8F / 2p ) * d 1 ( F ) ùû ( 8 / 2p ) éë rect ( 8F / 2p ) * d 1 ( F ) ùû e- j 8 p F

X ( F ) = ( 4 / p ) éë rect ( 4F / p ) * d 1 ( F ) ùû e- j 8 p F
2

x [ n ] = ( 4 / p ) sinc ( 2p ( n - 4 ) / 8 )

(d) x [ n ] = sinc 2 ( 2p n / 8 )

Using

X ( F ) = ( 8 / 2p ) rect ( 8F / 2p ) * rect ( 8F / 2p ) * d 1 ( F )
2

X ( F ) = ( 8 / 2p ) ( 2p / 8 ) tri ( 8F / 2p ) * d 1 ( F )
2

X ( F ) = ( 8 / 2p ) tri ( 8F / 2p ) * d 1 ( F )

47. Graph the inverse DTFT’s of these functions.

(a) X ( F ) = d1 ( F ) - d1 ( F - 1 / 2 )

Using and

Solutions 7-39
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 x[n]
2

n
-12 12

(b) X ( e jW ) = j2p éëd 2 p ( W + p / 4 ) - d 2 p ( W - p / 4 ) ùû

Using and

Solutions 7-40
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 x[n]
2

n
-12 12

-2

(c) X ( e jW ) = 2p éëd ( W - p / 2 ) + d ( W - 3p / 8 ) + d ( W - 5p / 8 ) ùû * d 2 p ( 2W )

éd ( W - p / 2 ) ù
ê ú
X ( e ) = p ê +d ( W - 3p / 8 ) ú * éëd 2 p ( W ) + d 2 p ( W - p ) ùû
jW

ê +d ( W - 5p / 8 ) ú
ë û

éd 2 p ( W - 4p / 8 ) + d 2 p ( W - 3p / 8 ) ù
ê ú
X ( e ) = p ê +d 2 p ( W - 5p / 8 ) + d 2 p ( W - 4p / 8 - p ) ú
jW

ê +d ( W - 3p / 8 - p ) + d ( W - 5p / 8 - p ) ú
ë 2p 2p û

Using and we get

or

Solutions 7-41
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 x[n]
3

n
-16 16

-3

48. ( )
A signal x [ n ] has a DTFT , X e jW = 10 drcl ( W / 2p ,5 ) . What is its signal
energy?

Using

(n 0
+ n1 ) / 2 = 1/ 2 and n1 - n0 = 5 Þ n0 = -2 and n1 = 3

\x [ n ] = 2 ( u [ n + 2 ] - u [ n - 3])

¥ 2

å 22 ( u [ n + 2 ] - u [ n - 3]) = 4 å 1 = 20
2
Ex =
n=-¥ n=-2

49. Given this DTFT pair

and the related pair find X 2 e jW ( ) W=p /4

.

Using
,

and the linearity of the DTFT,

X 2 ( e jp /4 ) = 3( e- jp /4 + 2e- j 2 p /4 + 3e- j 3p /4 ) e- j 2 p /4 = - j3( e- jp /4 - j2 + 3e- j 3p /4 )

Solutions 7-42
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 X 2 ( e jp /4 ) = 15.0938e j 2.8567

50. A signal x [ n ] has a DTFT,

X ( e jW ) = 2p éëd 2p ( W - p / 2 ) + d 2 p ( W + p / 2 ) + jd 2 p ( W + 2p / 3) - jd 2 p ( W - 2p / 3) ùû .

What is the fundamental period, N 0 , of x [ n ]?

Using and
,

x [ n ] = 2 cos ( 2p n / 4 ) + 2sin ( 2p n / 3) .

The least common multiple of the two periods of the two sinusoids, 4 and 3, is 12.
Therefore N 0 = 12 .

51. ( )
Let x éë n ùû = tri n / 4 * d 6 éë n ùû and let .

(a) What is the numerical value of x éë 21ùû ?

¥

(
x éë n ùû = tri n / 4 * ) å d éë n - 6k ùû
k = -¥
¥
x éë n ùû = å tri (( n - 6k ) / 4)
k=-¥

All the terms in this series are zero except two.

( ) (
x éë 21ùû = tri ( 21- 18) / 4 + tri ( 21- 24 ) / 4 )
x éë 21ùû = tri ( 3 / 4 ) + tri ( -3 / 4 ) = 1/ 4 + 1/ 4 = 1/ 2

(b) What is the lowest positive frequency W at which X e jW ¹ 0 ? ( )

Solutions 7-43
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 The fundamental period of x éë n ùû is 6. Therefore X e jW has ( )
impulses at integer multiples of 2p / 6 and the lowest frequency at
( )
which X e jW ¹ 0 is then W = 2p / 6 .

52. Find an expression for the inverse DTFT x éë n ùû of

and evaluate it at the discrete time n = 2 .

( )
Using multiplication-convolution duality, x éë n ùû = 1/ 10 sinc n / 10 cos 2p n / 4 . ( ) ( )
sin ( p / 5)
x éë 2 ùû = (1/ 10 ) sinc ( 2 / 10 ) cos ( p ) = (1/ 10 ) cos ( p ) = -0.09355
p /5

53. If then what is the

numerical value of X (11 / 8 ) ?

X ( F ) = tri ( 4F ) * d ( F ) * éëd 1 ( F - 1 / 4 ) + d 1 ( F + 1 / 4 ) ùû

X ( F ) = éë tri ( 4F ) * d 1 ( F - 1 / 4 ) + tri ( 4F ) * d 1 ( F + 1 / 4 ) ùû

¥
X( F ) = å éë tri ( 4 ( F - k - 1 / 4 )) + tri( 4 ( F - k + 1 / 4 )) ùû
k=-¥

¥
X (11 / 8 ) = å éë tri( 4 (11 / 8 - k - 1 / 4 )) + tri( 4 (11 / 8 - k + 1 / 4 ))ùû
k=-¥

¥
X (11 / 8 ) = å éë tri ( 9 / 2 - 4k ) + tri (13 / 2 - 4k ) ùû
k=-¥

54. Find the numerical values of the literal constants.

Solutions 7-44
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 (a)

X ( e jW ) = 2e+ j 3W - 3e- j 3W = 2e+ j 3W - 2e- j 3W - e- j 3W

X ( e jW ) = j4 sin ( 3W ) - e- j 3W

A = j4 , b =3 , C = -1 , d = −j3

(b)

A = 1/ 2 = 0.5 A = 1/ 2 = 0.5
a = 7 / 24 = 0.2917 or a = 5 / 24 = 0.2083
b = 5 / 24 = 0.2083 b = 7 / 24 = 0.2917

(c)

Solutions 7-45
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 because the rectangle width and the period are the same, 9.

A = 8p b = 2p

Alternate Solution: (Much more difficult, but instructive.)

drcl ( W / 2p ,9 ) is zero at every W that is an integer multiple of 2p / 9

unless W is also an integer multiple of 2p . When W is an integer
multiple of 2p , drcl ( W / 2p ,9 ) = 1 . d 2 p /9 ( W ) is a periodic sequence of
impulses spaced apart by 2p / 9 , each having a strength of 1. Therefore

8p drcl ( W / 2p ,9 )d 2p /9 ( W ) = 8pd 2 p ( W ) .

(d)

A = 5p @ 15.71 b = p / 7 @ 0.449

(e)

A = - j , b = 1 / 9 @ 0.111 , c = - j6p @ - j18.85

(f)

Solutions 7-46
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 A = 140 b = 20 c=1

(g)

A = 2 B = 2 c = 1 / p @ 0.3183

Alternate Solution:

4d 2 [ n ] alternates between 4 and 0 with 4 occurring at even values of n

and 0 occurring at odd values of n. The form, A + B cos ( n / c ) can be
made to alternate between 4 and 0 by making A and B both 2 and by
making n / c = p n which makes the cosine alternate between +1 and -1.

(h)

A = j24 / 5 @ j4.8 b = 5 c = 5 / 2p @ 0.796

Solutions 7-47
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.
 (i)

A=1 b=6 c=2

55. Let
n
( )
x [ n ] be a signal and let y [ n ] = å m=-¥ x [ m ] . If Y e jW = cos ( 2W ) , x [ n ]
consists of exactly four discrete-time impulses. What are their strengths and
locations?

X ( e jW )
Y( e jW
) = cos ( 2W) = 1- e - jW
Þ X ( e jW ) = cos ( 2W ) (1- e- jW )
¥
y[ n] = å x [ m ] Þ y [ n ] - y [ n - 1] = x [ n ] Þ X ( e ) = (1- e ) Y ( e )
jW - jW jW

m=-¥

X ( e jW ) = ( ) (1- e- jW ) = ( e j 2W + e- j 2W - e jW - e- j 3W )
1 j 2W - j 2W 1
e +e
2 2
x[ n] =
1
2
(d [ n + 2 ] + d [ n - 2 ] - d [ n + 1] - d [ n - 3])
Impulse #1. Strength = 1/2 Located at n = -2

Impulse #2. Strength = 1/2 Located at n = 2

Impulse #3. Strength = -1/2 Located at n = 3

Impulse #4. Strength = -1/2 Located at n = -1

Solutions 7-48
Copyright © McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior
written consent of McGraw-Hill Education.