UNIVERSITY OF ILLINOIS AT URBANACHAMPAIGN
Department of Electrical and Computer Engineering
ECE 410
Digital Signal Processing
Homework 4 Solution
Prof. Bresler, Prof. Jones
September 24, 2008
Problem 1
(10 points)
Complete the following signal ﬂow diagram for 8point, radix2 decimationintime FFT algorithm,
and use it to compute the DFT of the 8pt sequence
x
[
n
] =
{
1
,

4
,

2
,

3
,
1
,
4
,
2
,
3
}
by hand
.
Specify the value of the input sequence, output sequence, and the signal values after each stage,
i.e.
, at all the points marked by the solid circles in the diagram.
Note:
W
0
2
= 1;
W
0
4
= 1;
W
1
4
=

j
;
W
0
8
= 1;
W
1
8
=
1
√
2

j
1
√
2
;
W
2
8
=

j
;
W
3
8
=

1
√
2

j
1
√
2
Refer Figure 1.
Problem 2
(10 points)
The sequence
x
[
n
] = cos
±
π
8
n
²
,
∞
< n <
∞
was obtained by sampling a continuoustime signal
x
c
(
t
) = cos(Ω
0
t
)
,
∞
< t <
∞
at a sampling rate of 1024 samples/sec. What are three possible values of Ω
0
that can result in the
same sequence of
x
[
n
]?
Since
w
= Ω
T
and
T
=
1
1024
secs, the signal frequency is
Ω
0
=
±
π
8
+ 2
πn
²
1
T
=
±
π
8
+ 2
πn
²
1024
= 128
π
+ 2048
πn
1
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
where n is any integer.
Since cos function is symmetric, the following can be answers:
Ω
0
=

128
π
+ 2048
πn.
Therefore, the possible three signal frequencies could be
Ω
0
= 128
π,
2176
π,
4224
π,
when
n
= 0
,
1
,
2.
Problem 3
(20 points)
The continuoustime signal
x
a
(
t
) = cos(600
πt
)
is sampled with a sampling period
T
to obtain a discretetime signal
x
d
[
n
] =
x
a
(
nT
)
.
(a) Compute and sketch the magnitude of the continuoustime Fourier transform of
x
a
(
t
) and the
discretetime Fourier transform of
x
d
[
n
] for
T
= 1
ms
.
Continuoustime Fourier transform(CTFT) of
This is the end of the preview.
Sign up
to
access the rest of the document.
 Fall '09
 Digital Signal Processing, Signal Processing, sampling rate, complex multiplications

Click to edit the document details