0% found this document useful (0 votes)
52 views

Ecb3113, Edb2073 - Digital Signal Processing

The document is a final exam for a Digital Signal Processing course that took place on August 26, 2018. It contains 9 pages including the cover page and appendices. The exam consists of multiple choice and written answer questions testing concepts such as linear convolution, discrete time Fourier transforms, sampling rates, and properties of linear time-invariant systems. Students are instructed to show their work and identify any assumptions made in solving the problems.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
52 views

Ecb3113, Edb2073 - Digital Signal Processing

The document is a final exam for a Digital Signal Processing course that took place on August 26, 2018. It contains 9 pages including the cover page and appendices. The exam consists of multiple choice and written answer questions testing concepts such as linear convolution, discrete time Fourier transforms, sampling rates, and properties of linear time-invariant systems. Students are instructed to show their work and identify any assumptions made in solving the problems.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 10

UNIVERSITH

TEKNOLOGI
PETRONAS

FINAL EXAMINATION
MAY 2018 SEMESTER

COURSE : ECB3113/EDB2073 - DIGITAL SIGNAL PROCESSING


DATE : 26 AUGUST 2018 (SUNDAY)
TIME : 9:00 AM - 12:00 NOON (3 HOURS)

INSTRUCTIONS TO CANDIDATES

1. Answer ALL questions in the Answer Booklet.


Begin EACH answer on a new page in the Answer Booklet.
indicate clearly answers that are cancelled, if any.

4. Where applicable, show clearly steps taken in arriving at the solutions and
indicate ALL assumptions, if any.
5. DO NOT open this Question Booklet until instructed.

Note :
i. There are NINE (9) pages in this Question Booklet including the cover
page and appendices.
ii. DOUBLE-SIDED Question Booklet.

Universiti Teknologi PETRONAS


ECB3113/EDB2073

The impulse response zn] of a LTl system and the input signal x[#] are given
respectively, as
hinl={2,-172}, n=123
xnl={2,7,3-41;}, n=01234

a. Evaluate the output of the system, y[n], using the linear convolution.

[5 marks]

b. Evaluate. the. output of the system, 3{n], using the


5-point circular convolution.
[5 marks]

C. Evaluate the output of the system using Discrete Time Fourier Transform.

[5 marks]

d. Determine whether the system described by A[n] is an IIR or FIR filter.


Justify.
[5 marks]

e. State the changes that need to be made in x[x] and A[#] in order to make
a 7-point circular convolution equal to that of a linear convolution.
[5 marks]
ECB3113/EDB2073

a. Consider an ADC-DAC system depicted in FIGURE Q2. The DAC part of


the system uses an ideal lowpass filter with a cutoff frequency of
fi , i.e. half of the sampling frequency.

Sampler
|
x (O=x(nT)
| | |
|
|
x(n ) _ - , D/A Wr)
Sampling freq. =. x(n] » Quantizer |
Converter EUS
IT =f Hz
. | |
FIGURE Q2

i. The sampling frequency of the system depicted in FIGURE Q2 is


unknown. The input to the system is a linear combination of
sinusoidal signals with frequencies 300 Hz, 400 Hz and 600 Hz. The
output however consists of frequency components 300 Hz, 350 Hz
and 400 Hz. Using frequency domain analysis, determine the
sampling frequency used in the system.
[8 marks]

ii. In part (a)(i), the output of the system has different frequency
components than those found in the input. Explain the phenomenon
and describe how to avoid it.
[5 marks]
ECB3113/EDB2073

A digital video disc (DVD) has 8 channels of audio signal, each sampled at
48 kHz. The combined bitrate of all the channels is 6.144 Mbps.

i. What is the number of bits used to encode the audio signal in each
channel?
[6 marks]

ii. If the digitized audio signal from each channel needs to achieve a
signal-to-quantization-noise-ratio (SGNR) of 80 dB, what is the
maximum peak-to-mean signal ratio in decibels?
[6 marks]
ECB3113/EDB2073

a. A digital system formed from two LT! systems is shown in FIGURE Q3.

H,(=

=n] BO
x y(n)

H,(@) :
1
1\Z
C1432 42272

Hh =ms
FIGURE Q3

i. Determine H(z), the transfer function of the overall system.


[7 marks]

il. Determine the z-transform of the output ¥(z) if the input to the
system in FIGURE Q3 is given by

x[n]= on -1]+28]n].
[6 marks]

iii. Derive the output y[rn] for the input given in part (a)(ii).

Given: aun) <>“


z—a
[6 marks]

b. Discuss the time-invariance, causality and stability of the following filter

EC GE [6 marks]
ECB3113/EDB2073

wf
A discrete-time signal, x,[n], is given as

~ dy, n=0
3 . n<0

Sketch the sequence: and clearly label the sample values for
n=-2,-1,0,1,2.
[4 marks]

Determine the DTFT of x, [x].


[6 marks]

The sequence x, [zn] is multiplied with another sequence,

x,|n]=uln+2]-u[n—3] resulting in a new sequence, w|n].


Sketch the sequence, win].
[4 marks]

Determine the output sequence if w|[r] is fed as input to an

LTI system with impulse response

hn] ={-1,2,-2,0,4}.
0

[5 marks]
ECB3113/EDB2073

The magnitude spectrum and phase spectrum of a sequence, x(n], is


shown in FIGURE Q4. Sketch the magnitude spectrum and phase spectrum
of x[-n].
[6 marks]

Magnitude spectrum of x[n] Phase spectrum of x[n}

il

4 23 3 0 13 23 4 4 wz a3 0 uz 23 1
w (x rads) © (n rads)

FIGURE Q4

— END OF PAPER —
ECB3113/EDB2073

APPENDIX |
TABLE Al.1 Common DTFT Pairs

Sequence DTFT

o[n] 1
dfn — ng] g iMow
1
> 2nd (w + 2nk)
=—0c0

aun] Ljal<1
1—ge-j®
—a"u[-n—1] ,Ja}>1 1
1 — ge JO
u[n] 1 5
1=eet > nd (w + 2mk)
k=-~co
(n+ Daufn] ,lal <1 1
(1 —ae iv)?
r*sinw,(n +1) 1
sin w, un], Ir <1 1—2rcoswy,eJ® + r2g~Jj20
sin w, jay — {0 lol < w,
mm X(e’) = 0, w. <|lwl<n
_ (1, 0<n<M sinfa(M +1)/2] _.
x[n] = Py otherwise sin(w/2) ge JoM/2
eJnwo *

> 218(0 — wg + 21k)


k=—o0

cos(wyn + @)
> [me/?8(w — wy + 21k) + me P85(w + wy + 21k)
k=—00

TABLE Al.2 Properties of DTFT


Property Sequence DTFT
Linearity ax[n] + by[n] aX(e!®) + bY(e/®) |
Time-shift x[n — ng) Ee (CU)
Time-reversal x[-n] X(e 19)
Modulation e/™@ox[n] X(e/@—wa)y
Convolution x[n] * y[n] X(e/)yY(e/)
Conjugation x" [n] X*(e /*)
Derivative nxn] dX (el?)
ET n T dw
Multiplication x[n]y[n] =| X (0) ¥ (eI @-90))gg
—Ir
ECB3113/EDB2073

APPENDIX li
TABLE All.1 Common z-Transform Pairs
Sequence z-Transform ROC
S[n] 1 Allz
a™u[n] 1 lz] > |a|
1—gz1
—a™u[-n—1] 1 lz{ < |a]
1—gz-?1
un] 1 lz] > 1
1—2z1
—u[~n — 1] 1 lz] <1
1—2z"1
nauln) az™1 |z| > la]
(1—az"1)?
—na"u[-n — 1] az™1 |z| <a]

(cos won)uln] 1 — (coswg)z™? lz] > 1


1-(2coswy)z"t +272
(sinwgn)u[n] (sinwg)z™? lz} > 1
1-(2coswg)z™! +272
(r™ cos won)u[n] 1— (rcoswg)z™? lz] >r
1—- (2rcoswg)z t+ riz 2
(r™ sin won) un) (rsinwg)z™? lz| > r
1—(2rcoswg)z t+ r2z72
x) = [1 0=n=N 1 1-aVz™N jz] >0
0, otherwise 1—z1

TABLE AIll.2 Properties of z-Transform


Property Sequence z-transform ROC
Linearity ax{n] + by[n] aX(z) + bY (2) Contains Ry NR,
Time-shift x[n — ng] z7X(2) R,
Time-reversal x[-n] Xz 1/R,
Exponentiation a™x[n] X(a™t2) la|Ry
Convolution x[n] * y[n] X@)Y(2) Contains Ry NR,
Conjugation x*[n] X'(z") R,
Derivative nx[n] —, dX(z) Ry
dz

You might also like