ECE 464/564: Digital Signal Processing - Winter 2016

Homework 1
Due Tue. Jan 19, 2016

1. The continuous-time signal

xc (t) = cos(4000πt)
is sampled with a sampling period T to obtain the discrete-time signal
πn
x[n] = cos( )
3
(a) Determine a choice for T consistent with this information.
(b) Is your choice for T in part (a) is unique? If so, explain why. If not, specify another choice of
T consistent with the information given.

2. (a) Continuous signal

xc (t) = sin(20πt) + cos(40πt) + sin(60πt)

is used as the input for an ideal C/D converter as shown in Fig. 1 with the sampling period
1
T = 100 sec.

Figure 1: system for problem 2

Find the resulting discrete-time signal x[n].

(b) The continuous-time signal
sin(10πt)
xc (t) =
10πt
is sampled with a sampling period T to obtain the discrete-time signal

sin( πn
2 )
x[n] = πn
2

i. Determine a choice for T consistent with this information.

ii. Is your choice for T in part (i) is unique? If so, explain why. If not, specify another choice
of T consistent with the information given.

1
3. For the system shown in Fig. 2, X(ejω ), the Fourier Transform of the input signal x[n] is shown in
Fig. 3.

Figure 2: system for problem 3

Figure 3: X(ejω ), the Fourier Transform of the input signal x[n]

For each of the following choices of L and M, specify the maximum possible value of ω0 such that
Mω
X̃d (ejω ) = aX(ej L ) for some constant a.

(a) M=3, L=2

(b) M=2, L=3

2
4. Fig. 4 shows a continuous-time filter that is implemented using an LTI discrete-time filter ideal low
pass filter with frequency response over −π ≤ ω ≤ π as
(
1 |ω| < ωc ,
H(ejω ) =
0 ωc < |ω| ≤ π

Figure 4: system for problem 4

(a) If the continuous-time Fourier transform of xc (t), namely Xc (jΩ), is as shown in Fig. 5 and
ωc = π5 , sketch and label X(ejω ), Y (ejω ), Yc (jΩ) for T1 = 2 × 104

Figure 5: X(ejω ), the Fourier Transform of the input signal x[n]

(b) For T1 = 6×103 and for input signals xc (t), whose spectra are bandlimited to |Ω| < 2π ×5×103
(but otherwise unconstrained), what is the maximum choice of the cutoff frequency ωc of the
filter H(ejω ) for which no aliasing occurs. For the maximum choice of ωc , specify Hc (jΩ)

5. Consider the discrete-time system shown in Fig. 6

Figure 6: discrete-time system in problem 5

where

i L is an integer
(
x[ Ln ] n = kL, k is integer,
ii xe [n] =
0 otherwise

3
 iii y[n]=ye [nL]
(
L |ω| < π4 ,
iv H(ejω ) =
0 π4 < |ω| ≤ π

(a) Assume that L=2 and that X(ejω ), the DTFT of x[n], is real and is shown in Fig. 7. Make an
appropriately labeled sketch of Xe (ejω ), Y (ejω ), Ye (ejω ). Be sure to clearly label amplitudes
and frequencies.

Figure 7: X(ejω ), the Fourier Transform of the input signal x[n]

(b) For L=2, the overall system is LTI. Determine and sketch the magnitude of the frequency
response of the overall system |Hef f (ejω )|.
(c) For L=6, the overall system is still LTI. Determine and sketch the magnitude of the frequency
response of the overall system |Hef f (ejω )|

6. For the system shown in Fig. 8 find an expression for y[n] in terms of x[n]. Simplify the expression
as much as possible. Each of the three blocks (upsampling, interp., and downsampling) correspond
to the system in Fig. 2 and the intepolator is the lowpass filter in Fig. 2.

Figure 8: Discrete-time system for problem 6