EE341 Final

Page1/4
EE341, 08ECE FIAL EXAM Jul 11

th
2011
ame SOLUTIOS
- Please do your own work. Explain what you are doing, and what assumptions you are making.
- If you use extra sheets of paper, please put your name on them also.
Problem 1 [20 points] Let X() and Y
d
() represent the DTFTs of x[n] and y
d
[n], respectively. In
the case y
d
[n] = x[2n], prove the relationship

If X() is as shown below, sketch Y
d
() in the range -2 2

)
`
\
|
+
+
|
\
|
=
2 2 2
1
) ( X X Y
d
( ) ( )
( ) ( )
side left Y k y k x
e n x e n x
e n x e e n x
e n x e n x side Right
d
k
k j
d
k
k j
odd n
n j
n
even n
n j
n
n
n j
n
n
n j
n j
n
n j
n
n j
= = = + =
+ + + =
+ = + =
+ =

=
|
\
|
+
) ( ) ( 0 2 ) 2 (
2
1
) 1 ( 1 ) (
2
1
) 1 ( 1 ) (
2
1
) 1 ( 1 ) (
2
1
1 ) (
2
1
) ( ) (
2
1
2
2
2 2
2 2
2
2
Y
d
()
12 points
8 points
Page2/4
Problem 2 [20 points]
a) The following five samples of the 9-point DFT X(k) of a real length-9 sequence are given: X(0) =
23, X(1) = 2.2426 j, X(4) = -6.3794 + j4.1212, X(6) = 6.5 +j2.5981, and X(7) = -4.1527 + j0.2645.
Determine the remaining four samples.

b) A length-7 sequence x(n) is given by x(0) = 5, x(1) = 3, x(2) = -2, x(3) = -4, x(4) = 1, x(5) = 2,
x(6) = 0. Let X(k) denote the 7-point DFT of x(n). Without computing the IDFT, determine the
sequence y(n) whose 7-point DFT is given by Y(k) = W
7
4k
X(k).

Problem 3 [15 points] The linear convolution of a lengh-60 sequence with a length-1200 sequence is
to be computed using 128-point DFTs and IDFTs. Determine the smallest number of DFTs and
IDFTs needed to compute the above linear convolution using the overlap-add approach.
To compute the linear convolution of the lengh-60 x[n] with the length-1200 h[n] using the overlap-
add method, we first need to divide h[n] into the segments h
i
[n] of length (128-60+1=) 69 points so
that the linear convolution of x[n] and h
i
[n] has the length of 60+69-1 = 128 points.
Second, we compute the linear convolution of x[n] and h
i
[n] by computing the circular convolution of
length-128 x[n] and length-128 h
i
[n] and get the result y
i
[n] of the same lengh 128.
Finally, we sum up y
i
[n] to get y[n].
Need to divide h[n] into (ceil(1200/69=)18 segments, and need 18x2 = 36 DFTs and 18 IDFTs.

5 points
j X X
j X X
j X X
j X X Therefore
k X k X or
k X W n x W W n x W n x k X
real n x For
n
n k
kn
n
n
n
kn
+ = =
= =
= =
= =
=
= = = =

=

2426 . 2 ] 1 [ * ] 8 [
1212 . 4 3794 . 6 ] 4 [ * ] 5 [
5981 . 2 5 . 6 ] 6 [ * ] 3 [
2645 . 0 1527 . 4 ] 7 [ * ] 2 [ :
] [ * ] [
) ( ] [ . ] [ ] [ ] [ *
: ] [
1
0
) (
1
0
1
0
2 points
2 pts/ea
] 2 3 5 0 2 1 4 [ ] [
) ]( 4 [ ] [ ] 4 [ ] [
: 4
] [ ] [ ] [ ] [
1
0
1
0
) 4 (
1
0
4
7
4
7
=
= =
= +
= = =
=
+

n y
shift circular n x n y W m x k Y
m n Let
W n x W n x W k X W k Y
k
mk
k
k n
k
kn
k k
3 points
7 points

5 points

5 points
Page3/4
Problem 4 [20 points]
a) Determine the periodic signal x(n), with fundamental period N = 8, if their Fourier coefficients are
given by

We have:
Therefore:

The periodic signal x[n] is as below:

b) Determine and sketch the magnitude spectrum of the following periodic signal

We start with one period x
0
(n), then take the DTFT X
0
(), then find X
0
(k2/4)

We obtain the magnitude spectrum

{ }
{ } j j e n x k X
n x n x
n
n
k
j
k
n
3 2 , 3 , 3 2 , 1 ) ( ) 4 / 2 (
2 , 2 , 1 , 0 ) ( ) (
3
0
4
2
3 0
0
3 0
0
+ = =
= =
4
3
sin
4
cos
k k
c
k
+ =
{ } ... , 1 , 0 , 2 , 2 , 1 , 0 , 2 , 2 , 1 , 0 , 2 , 2 ..., ) ( =

n x
{ }
{ } 5 . 0 0 5 . 0 0 5 . 0 0 5 . 0 0 ] [
8
1
] [
7 0 ], [ 0 1 2 1 0 1 2 1
7
0
8
7 0
j j W k X n x
k k X C
k
nk
n
k
= =
=

{ } ... 0 5 . 0 0 5 . 0 0 5 . 0 0 5 . 0 0 5 . 0 0 5 . 0 0 5 . 0 0 5 . 0 0 5 . 0 0 ... ] [ j j j j n x =
4 points
4 points
2 points
{ } ... , 6636 . 5 , 7124 . 4 , 6636 . 5 , 5708 . 1 , 6636 . 5 , 7124 . 4 , 6636 . 5 , 5708 . 1 , ... | )
4
2
( |
4
2
=

k X
4 points
4 points
2 points
Page4/4
Problem 5 [25 points] Determine the transient
and steady-state responses of the FIR filter shown in
the figure below to the input signal x(n) = 10e
jn/2
u(n).
Let y(-1) = y(-2) = y(-3) = y(-4) = 0
The given FIR filter has the input-output equation as below

From the input-output equation above, by substituting x(n), we can determine the system response
y(n) of this filter:

The system response can be considered as a sum of 2 terms, the 1
st
term is the steady-state response
and the 2
nd
term is the transient response which decays toward zero as n goes to infinity:
y(n) = y
ss
(n) + y
tr
(n)
Therefore we have the following transient response of this filter:

This transient response only exists within the very short duration, so that in practical applications we
can ignore it if the accuracy is not so important.
2
x(n)
z
-1
z
-1
z
-1
z
-1

y(n)
) 4 ( 40 ) 3 ( 40 ) 2 ( 40 ) 1 ( 40 ) ( 40
) ( 40
2
) ( ) (
: ) ( 10 ) (
4 2 1
2
2 1 ) (
) 4 ( ) 2 ( 2 ) ( ) (
2
2
2
2
4
2
2
4 2
+ + =
=
|
\
|
=
=
= + =
|
\
|
+ =
+ =

n u e n j n n j n
n u e H n x n y
response state steady the have we n u e n x given the With
e e H
e e H
n x n x n x n y
n
j
n
j
ss
n
j
j j
j j
) 4 ( 40 ) 3 ( 30 ) 2 ( 30 ) 1 ( 10 ) ( 10
) 4 ( 10 ) 2 ( 20 ) ( 10 ) (
2
2
4
2
2
2
+ + =
+ =

n u e n j n n j n
n u e n u e n u e n y
n
j
n
j
n
j
n
j

) 3 ( 10 ) 2 ( 10 ) 1 ( 30 ) ( 30
) ( ) ( ) (
+ + =
=
n j n n j n
n y n y n y
ss tr

5 points
5 points
5 points
5 points
5 points

EE341 Final

Uploaded by

Copyright:

Available Formats

EE341 Final

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

EE341 Final

Uploaded by

Copyright:

Available Formats

Page1/4

EE341, 08ECE FIAL EXAM Jul 11

You might also like