0 ratings0% found this document useful (0 votes) 95 views19 pagesBec502 Passing Package
Digital signal processing
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content,
claim it here.
Available Formats
Download as PDF or read online on Scribd
CB CSS CHEN E
[Semester B.E. Degree Examinati , Jan/Feb. 2021
ital Signal Proges ing
18EC52
Max. Marks: 100
y ¢
ing ONE full question'from each module.
@ Module-t
1a. Show thet the multiplication of wo DFT’s leads to cceularégavolution of the
corresponding time sequences. aw] (08 Marks)
b. Compute the N — point OFT’s of the signals : ’
1,
D x(q ; ”
ay Nensna
it) x(n)=cos “Bion, 0Sa ats
Use the Hamming window function, obtain the frequency response of the desi
filter. (19 Marks)
b. For the System function H(z) = 1 +2.8243,427 4 1.729 + 0.4z" iain the Lattice
coefficients and sketch the Lattice structure. ” (0 Marks)
LOR
a. Find the Impulse response of an FIR filter with the following’ desired frequency response,
} x
Haw) = {
6
Use Rectangular windéw/fiinction. Draw the direct ‘ore structure for the designed filter.
(10 Marks)
b. Consider an FIR Lattice filter coefficients Ky = 0.65, Ka =0.5, Ks = 0.9. Find its impulse
response and draw the direct form structures"). (10 Marks)
Modute-4
a, Define the First order analog low pass filter prototype. How this prototype is transformed
into a different filter types. 4 (05 Marks)
b. Design'a Second order digital low pass Butterworth filter with a cutoff frequency of 3.4 kHz
at a sempling frequency of 8000Hz, Draw the direct Form It structure ofthis filter. Use
Bilinear transformation. (10 Marks)
¢. Discuss the general mapping: properties of bilinear transformatiog aad show the mapping
between the § — plane and the the Z — plane. (05 Marks)
oR.” &
a. Define the Nortalized low pass prototype function of (Bu worth filter and derive the
expression forthe filter order, (05 Marks)
b. Using Bilinear transformation, design’a digital low pass Butterworth fitler with the following
specifications : Sampling frequency : 8000Hz , 3 dB’attenuation at 1.5 KHz. 10 dB stop
band attentiation at 3kHz. 7” (10 Marks)
c. Realize the following digital filter using direct Form — II
Hq)=—27 +1 Az £0. Te? +0. Se? : . . (05 Marks)
V4 Le 40,52740.72" +0324 7
) Module-5
a, With a neat diagram, explain the Harvard architecture used in DS processors. (06 Marks)
._ IMustrate the operation of circular buffers used for address generation in DS processors.
(07 Marks)
. Convert the following decimal numbers into the floating point representation
i) 0.640492 x27 Ail). - 0.638454 x 2°.
Use 4 ~ bits to represent exponent and 12 - bits for mantissa, (07 Marks)
OR
a. With a neat diag, explain the basic architecture of TMS320C54X family DS processors.
(10 Marks)
b. Describe the IEEE single precision floating point format used in DS processors. (05 Marks)
c. Find the signéd Q— 15 representation for the decimal number 0.560123. (05 Marks)
#2 of2**SCHEME
vs "
Fifth Semester B.E. Degree Examination, Feb./Mar.2022
Digital Signal Processing
ime: 3 hes. Max. b
Note: Answer any FIVE full questions, choosing ONE fall question fro module
g Modute-1
E 1 a. Prove that the sampling of DTFT of a sequence x(n) result in N-PQMM DID) with a neat
7 diagram. QO Marks)
a b. Find the 4-point DFT of the sequence x(n) = {1,0,0,1} using{Matrix mfQthod and verity the
i answer by taking the 4-point [DFT of the result. (QO Marks)
Ee oR
= 2 a Derivethe circular Time shift property. (06 Marks)
3s b. Compute the circular convolution of the following seqages Bing DET and IDFE method
ge x(n) = 1, 2.3, 4} and x, (n)= 4,3, 2,1}. (09 Marks)
a4 2 :
Be e it welts! eof (oS) what he of the window sequence
gs 27 2°"[ 2
ee y(n) = x(n).v(n)? Relate the answer in t ). (05 Marks)
£3
34
EE 3 a. Find the output © response is h(n) = (1, 1, 1} and the input
BE signal x(n) .3,2,0, sing) overlap-add method, Assume the length of
ge cach block N is 6. (10 Marks)
i z be mmplexity? Compare the direct computation and FFT
4 $F 52-point DFT of x(a), How many
ge
be
be 7
bE (10 Marks)
Es
gS
BS 4a ignal flow graph, (10 Marks)
Eg b. 0.Sn $7. Find X(K) using DIF-FFT algorithm. QO Marks)
Ee MoJates
SF 5 a, ifferent design techniques available for the FIR filters? Explain Gibbs
adi henomenen. Explain the four window techniques for the designing of FIR filters.
£ ster: , Co Marks)
a filter is to be designed with the following desired frequency response,
E sfor—38 soe 3%
& He™= "ae A 4
sos
7 5
Determine 1(e!) for M = 7 using Hamming window GoMarks)
1of2
A18ECS2'
oR
6 a, AFIR filteris given by,
2 Ben _aat
yeaa x(n)+ Ex(n—1)+ x(n -2)+ 580-3)
Draw the lattice structure.
b, Based on the frequency-sampling method, determine the coefficients of a line
filter of length 15 which has a symmetric unit sample response and
response that satisfies the conditions.
K=0,1,2,3
04; Kad
0 5 K=5,6,7
iIkehis givgp by,
S?414148+1
Convert the analog filter into digital filter will
bilinear transformation. Assume T = 1 ses.
H,@)=
ey of O.5x md’see using
(QO Marks)
dyed sin=2)
b. A filter is given by the difference equation y(n)
Draw direet form ~ I and direet form — 1] 1s TAlso obtain the transfer function of
the filter, (QO Marks)
8 a, Derive mapping function used ir
transformation, preserves the freq
ing analog filter to digital filter by bilinear
ivity and stability properties of analog filter.
(Marks)
b. When used in the analog to digital with digital to
ent specification,
Alfbut off 100 z radsee,
$5 dB at 1000 x rad'sce.
lop band and pass band.
Go starks)
90. ignal processors based on the Harvard architecture.
Go Marks)
b. ignal processor hardware u
jer and Accumulator (MAC) unit.
ers.
88 Generators. Go Marks)
oR
oating-point formats:
ingle precision format.
Double precision format, (Qo Marks)
ith the diagram, explain the basic architecture of TMS320CS4X family processor.
GO Marks)
2o0f2=
»
CBES SCHEME
TUTTI ise!
Degree Examination, Jan./Feb. 2023
Digital Signal Processing
Max. Marks
FIVE full questions, choosing ONE full question from engh wodute
Mosuect °
State and prove circular time shift property, 6 Markey
Find te nin others tn) = Cos (En) rcs
Comiedintngenestn)= Co (S200
hn= 2"; 050
‘Compute circular convolution. Using concentric circle Red (os Marka
on
tate and prove Pareval’s theorem, (06 Marksy
DIT of the sequ
: Osnss on tar
=0; otherwise (nMarial
Find the IDFT of the DET X(K) =
-2 coo Marky)
Consider a FIR with titer whose
xO) = 12.33.21 1
method assuming the length
Develop Radix. 2 DIT-FFT
rose Bem) = 13. 24 Ua HL the input is
find the output using over lap aed
Co atarks)
8
aon
rey
di the omput yin) &
ter whose impulse response in h(n) = {1.1.1} and the input
1,0, 1,3,2,0, 1,2, 1). Using overlap save method,
{10 Marhsy
Joint DET ofa real valued sequence is given by
I sequence x(n) using Decimation in
(10 Marks)
Mowute-3
ht types of windowing techniques used in the design of ers. Waite the
analytical equations, draw the magnitude response and show the largest side lobe value
slow tpg de magnitude.
108 Marks
Freeney response ofan FIR filter is given by
eP(1 4 LXeos309 + 1.200820 10.5 €0s00)
DefEnnine the coetficient of the impulse response hin) of the FIR filter. Ma
dot2system funet
ion (7) = 1-422" + 27 7. And also draw the L:
or
ISECS2
Detennine the coefficient Ky of the lattice filter corresponding to FIR filter described by the
(06 Magy
Determine the Biter coefficient hy(n) for the desired frequency response of a Lowpatyilter
is given by
ec
How)
Find h(n) an
He
Discuss the general procedure for IIR filter design using,
An analog filter is given by H,(s)
transformation method. Digital filter is to hy
my Bewe®
3 4
: Zswen
0: Eswse
also frequency response H(w) using Hamming window.
Obtain the cascade form realization of system funtion :
Se) 4B De!
Realize the following function in Direct form.
fy. Va ytay
Ways [1s 5rt=2 te a” )
Madute-4
ar (as formation.
S01
one
MIR filter ws
frequency ,
adians,
Compare FIR and IIR filter.
Design a But
4) 3dB attenuation at the passbar
ii) 1008
iti) Sami
A system is represented by| ction 11(z) is given by H(z)
Convert the
iterworth digital law pi the following speci
FL SKIz
of SKI
topband attenuation at
ing frequency of
ration by showing «id
fr and Accumulator unit in Digital signal processor hardy
fam to TMS320C3X floating point digital
oR
I processor based on Harv
ignal processor.
jagram explain Digital
wd architecture,
(06 Macks)
Q-15 sigiied number to decima
numbers,
i) QJ10101 110000010 ii) 01000111101 10010
xplaih the
ie architecture of TMS320CSS4X used in fixed point Digi
seers
@
Marks)
(05 Marks)
(05 Marks)
(06 Marky,
bilinear
(08 Marks)
(06 Marks)
tiflerences
(10 Maeks)
(Ox Marks)
ware units
(04 Marks)
(08 Marks)
(01 Marks)
tal signal
(10 Markaeae
GAGS SCHEME
usw [ Tr
Fifth Semester B.E. Degree Examination, July/August 20:
Digital Signal Processing
188C52
x. Marks: 100
Time: 3 hrs.
Note: Answer any FIVE full questions.
1 a. Describe the process of frequency domain samplingsaR@ reconsgiction of discrete, tine
signal. (03 Marks)
using Matrix method. (04 Marks)
Find the 4-point DFT of the sequence x(n) = {1, 2,
©. Using graphical method (concentric method) obl
signal defined 2s,
circular convolution of two DFT
x(a) =(L.5)"; Ot wet (08 Marks)
~~ 3 of 3