You are on page 1of 7

~o\\ No- '2\1C N0j1 ..

Du ttATtOrJ qO ITl\..,\- 1~

,. (-\ -n.-:,w er "''' q,ue~ +,


-i.., ue~+tc n"V' Ca~ C?'V-'-,.J T<lq'r lvl •

,.. )
---- ---= === === === === --= ----_-_-_-_---_,
__
t Q._,. ~c_ . oL.. .dev vaf, ~ ct...rc_
't

~nJ = A- ~:Y'\ ( rtrf,._n +cf, ) -t- w C.""'>:l


' 'Yl.__ c,, \~ , •",1 N-f

r~
w½e rt W CTIJ ..'-" W r,;iJ • LJ'n1-, Vctr~ o.,.-,{' e -r-1-,~
0-r-.-,p ~i 't"-\.de f'\ or-n cL. f 'r\ (fd' • 1o -+- +£. e ....-:j i'-n'--4~ O A,d
a...._.-e.. u ..... c d.. l~ J. k....-r-,io wv, • - ..:o e f-1!'.,... "TV'\ 1-n e... 4 e
C (<_~ F~~ -r ,
4:t,~ .~J,.,1 ch 't,') r:-b e e~+:-..,,-._1-ycJ

o ""ltC--.:,'J
h ~\io'L- ..f-1-ie.... 'd le, 3'h . f'...o F-
\~
·r C:x. £:--,."JJ o -z-_) - ,-.-~'\'Q-
s-2-
(_- .J_ -,c.~ )
2- (;;'Z..

?t.L...,.;) >1>
-

I
l
;

-i
1I ~-;~. ,... ¼~ pdf-
I
!
: \., c'JC..; >-) {
0

L
y

~_(. -r'1c: o~~ ew a+, vw , o.-, e


')1
I
,- "Jl. c~'J = A 14\- A-C
J...
-1J +wC""l "'~ 01 '1 , 1 rJ-1
w heY'¤- W[ -nJ '"
v "l,.
U\ w (r.rJ Lu ,--tt.,
o; . I\
Q ,... , a Yl re_
' '\ -the.· "1Vl~t-A~
?t--th e --rvr c de { , F ;n d
~~ ~ll
ttr'"W\·
--tt,e M\ /u ~~~ "Ma
and .. ~e )y
_\,, (
I
c:1\ H-,e. Cce.fh /, e-nf,:1 A;
(o "lJ.a'Y> a "l"'I ts
/
Q./ (~U 'M l'1"l i N- "'.~ f~l "\-
'
I
I
ROLL NO.: l2lll2ICINlo lo l~I
DEPARTMENT OF ELECTRONICS & COMMUNICATION ENGINEERING
NATIONAL INSTITUTE OF TECHNOLOGY KARNATAKA, SURATHKAL

EC734 SIGNAL DETECTION AND ESTIMATION

EVEN SEMESTER 2021-22, ENDJSEMESTER EXAMINATION

DATE: 02-06-2022 TIME: 08.30 AM - 10.30 AM


DURATION: 02 HOURS MAX. MARKS: 30

Note:

• Answer all questions.


• Subparts of a question should be answered together.

l; a) The IID observations x[n] for n = 0, 1, ... , N - 1 have the Rayleigh PDF
J V p(x[n]; a 2) = xJ~lexp(-½x:knj) x[n] > 0
2
p(x[n]; a ) = 0 x[n] < 0
2
Find a sufficient statistic for a
)) Derive expressions for the Linear Least Squares estimate and the minimum least
squares error. ,, 07

that minimize the Bayes risk for different cost functions .


.,,,- .,a) Derive the estimators
2.
9) The data x[n], n = 0, l, ... , N - 1 are observed having the PDF
2
p(x[n]Iµ) = J 2~u2 exp-~(x[n ] - µ)
The x[n]'s are independent when conditioned on µ. Themean µ has the prior
PDF
µ r-.., N(µo, a5)
Find the MMSE and MAP estimators of µ 08

3.
, a) Consider the hypothesis testing problem
' Ho : x[n] = w[n] n = 0, 1, ... , N - 1
H1 : x[n] = A + w[n] n = 0, 1, ... , N - 1
2
where A-/= 0 is unknown and w[n] l"v N(0, a ) is WGN. Determine the Generalized
Likelihood Ratio Test.
b) Consider the simple binary hypothesis testing problem. Corresponding to the two
.,,, hypotheses, the observations are
Ho : x[0] = w[0]
H1 : x[0] = s[0] + w[0]
where s[0] = 1 and w[0] r-.., N(0, 1].
Determine the LRT.
Derive expressions for Po and PF A·
07
\

(x { 4) Consider the hypothesis testing problem


- / Ho : x[n] = so[n] + w[n] n = o, I, ... , N - 1
H1 : x[n] == s1{n] + w[n] n == o, 1, ... , N -- 1 ... •
where si[n] are known deterministic signals and w[n] is WGN with varian ce 0 2.
Startin g from the LRT, derive the structu re of the minimum distan ce receiver.
Obtai n expression for the probability of error. 08

**** ***
\
\
ROLL NO. : 121212\CINI I I I
DEPARTMENT OF ELECTRONICS & COMM
NATIONAL INSTITUTE OF TECHNOLO UNICATION ENGINEERING
GY KARNATAKA, SURATHKAL
EC734 SIGNAL DETECTION AND ESTIMATION

EVEN SEMESTER 2022-23, MID-SEMESTER EXAMINATION


DATE: 02-03-2023
DURATION: l½ HOURS TIME: 08.30 AM - 10.00 AM
MAX. MARKS: 25
Note:

• A'i'lSwer all ques_tions.

1. The IID observations x[n] for n = 0, 1, ... , N - 1 have the Rayleigh PDF
p(x[n); cr2 ) = ~exp(-½ x:~)) x[n] > O
p(x[n); CQ = 0 x[nL < 0 ___
Find a sufficient statistic for cr2 - - . - 05
2. We observe N IID samples from the PDF
p(x; ,\) = ,\exp(-,\x) x>0
p(x : ,\) 0 X < 0.
Find the MLE of the unknown parameter and verify that it indeed maximizes the
likelihood function. - 05
3. Derive expressions for the Linear Least Squares estimate and the minimum least
squares error. 05
4. The data x[n] for n = 0, 1, ... , N - 1 are observed, each sample having the conditional
PDF
p(x[n] 10) = exp[-(x[n] - 0)] .~[n] > 0
p(x[n] 10) = 0 x[n] < 0
and conditional on 0 the observations are independent. The prior PDF is
p(0) = exp(-0) 0> 0
p(0) = 0 0 < 0.
Find the MMSE estimator of 0 05
5. Consider the simple binary hypothesis testing problem. Corresponding to the two
hypotheses, the observations are
Ho : x[O] = w[O]
H1 : x[O] = s[O] + w[O]
where s[O] = 1 and w[O] rv N(O, 1].
For the NP test, derive expressions for Po and PFA• If PFA = 3X10-3 determine Pv
for the NP test.
For the Q(.) function, use the approximation
Q(x) = /4;,exp( -½x2) \_ 05
* * * '-* * *
ROLL NO. : l2I2I2ICINI () 11-1
DEPARTMENT OF ELECTRONICS & COMMUNICATION ENGINEERING
NATIONAL INSTITUTE OF TECHNOLOGY KARNATAKA, SURATHKAL

EC734 SIGNAL DETECTIO N AND ESTIMATIO N

EVEN SEMESTER 2022-23, END-SEMESTER EXAMINATION

DATE: 18-04-2023 TIME: 10.00 AM - 01.80 Pi~1


DURATION: 03 HOURS MAX. MARKS: 50

Note:

• Answer all questions.

• Subparts of a question should be answered together.

1. a) Consider deriving expression for the BLUE. Obtain expressions for the variance
and the other constraints that the estimator should satisfy. 04
b) The data x[n] = Arn+ w[n] for n = 0, 1, ... , N - 1 are observed, where w[n] is
WGN with variance a 2 and r > 0 is known. Find the CRLB for A. Show that an
efficient estimator exists and find its variance. What happens to the variance as
N -+ oo for vi .rious values of r ? 06
2. a) The HD observations x[n) for n = 0, 1, 2, ... , N - 1 have the exponential PDF
p(x[n]; .\) = .\exp(-.\x[n)) for x[n] > 0
p(x[n]; .\) = 0 for x[n] < 0
Find a sufficient statistic for A. 04
b) Show that a Fourier analysis is an estimation of the linear model parameters. 06
3. 6 a) Derive the estimators .that minimize the Bayes risk for different cost functions.
05
b) Derive expressions for the LMMSE estimator and the minimum Bayesian MSE.
05
4. a) Consider the hypothesis testing problem
H o x[n] = w[n]
:
n = 0, 1, ... ,
N - 1
H1 : x[n] =A+ w[n] n = 0, 1, ... , N - 1
where A-/= 0 is unknown and w[n] rv N(0, a 2 ). Determine the Generalized Like-
lihood Ratio Test. 04
b) Consider the detection problem to distinguish between the two hypotheses
H0 : x[n] = w('.'l] n = 0, 1, ... , N - 1
H 1 : x[n) = s[n) + w[n] n = 0, 1, ... , N - 1
where s[n] is assumed to be known and w[n] is WGN with variance a 2 . Starting
from the NP criterion, obtain the two implementations of the detector. 06
5. a) Apply the LMMSE estimator for the filtering problem and obtain the Wiener-
Hopf filtering equations. 04

L
b) Consider the hypothesis testing problem :
Ho : xin] = so[nJ + w[nJ n = o, 1, ... , N - 1
H1 : x[nJ = s1 [nJ + w[nJ n = o, 1, ... , N - 1
where si[nJ are known deterministic signals and w(n] is WGN with variance a 2 .
. I

Starting from the minimum probability of error criterion, derive the structure of
the minimum distance receiver and obtain expression for the probability of error.
06

*******

You might also like