You are on page 1of 28

Unit.4.

Correlation and spectral densities


1
MISRIMAL NAVAJEE MUNOTH JAIN ENGINEERING COLLEGE, CHENNAI
DEPARTMENT OF MATHEMATICS
PROBABILITY AND RANDOM PROCESSES (MA2261)
SEMESTER IV
UNIT-IV: CORRELATION AND SPECTRAL DENSITIES
QUESTION BANK ANSWERS
PART-A
Problem1. Define autocorrelation function and prove that for a WSS process ( ) { }
, X t
( ) ( )
XX XX
R R t t = 2.State any two properties of an autocorrelation function.
Solution:
Let ( ) { }
X t be a random process. Then the auto correlation function of the process
( ) { }
X t is the expected value of the product of any two members ( )
1
X t and ( )
2
X t of
the process and is given by ( ) ( ) ( )
1 2 1 2
,
XX
R t t E X t X t = (

or
( ) ( ) ( ) ,
XX
R t t E X t X t t t + = + (

For a WSS process ( ) { }
X t , ( ) ( ) ( )
XX
R E X t X t t t = (

( ) ( ) ( ) ( ) ( ) ( ) ( )
XX
R E X t X t E X t X t R t t R t t t t t = + = + = + = ( (

Problem 2. State any two properties of an autocorrelation function.
Solution:
The process ( ) { }
X t is stationary with autocorrelation function ( ) R t then
( i ) ( ) R t is an even function of t
( ii ) ( ) R t is maximum at 0 t = i.e., ( ) ( ) 0 R R t s
Problem 3. Given that the autocorrelation function for a stationary ergodic process with
no periodic components is ( )
2
4
25
1 6
R t
t
= +
+
. Find the mean and variance of the
process ( ) { }
X t .
Solution:
( )
2
2
4
25 25
1 6
x
Lt Lt
R t
t t t
= = + =
+
5
x
=
( ) ( ) ( )
2
0 25 4 29
XX
E X t R = = + =
( ) ( ) ( ) ( )
2
2
29 25 4 Var X t E X t E X t ( = = = (

Problem 4. Find the mean and variance of the stationary process ( ) { }
X t whose
autocorrelation function ( )
2
2
25 36
6.25 4
R
t
t
t
+
=
+
.
Solution:
Unit.4. Correlation and spectral densities
2
( )
2
2
2
2
36
25
25 36
4
6.25 4
6.25
R
t
t
t
t
t
+
+
= =
+
+
( )
2
25 2500
4
6.25 625
x
Lt
R t
t
= = = =

2
x
=
( ) ( )
2
36
0 9
4
XX
E X t R ( = = =

( ) ( ) { } ( )
2
2
9 4 5 Var X t E X t E X t = = = ( ( (

Problem 5. Define cross-correlation function and mention two properties.
Solution:
The cross-correlation of the two process ( ) { }
X t and ( ) { }
Y t is defined by
( ) ( ) ( )
1 2 1 2
,
XY
R t t E X t Y t = (

Properties: The process ( ) { }
X t and ( ) { }
Y t are jointly wide-sense stationary with the
cross-correlation function ( )
XY
R t then
( ) ( ) ( )
YX XY
i R R t t =
( ) ( ) ( ) ( ) 0 0
XY XX YY
ii R R R t s
Problem 6. Find the mean square value of the process ( ) { }
X t if its autocorrelation
function is given by ( )
2
2
R e
t
t

= .
Solution:
Mean-Square value ( ) ( )
2
2
2
0
0 1
XX
E X t R e
t
t

=
| |
( = = = = |

|
\ .
Problem 7. Define the power spectral density function (or spectral density or power
spectrum) of a stationary process?
Solution:
If ( ) { }
X t is a stationary process (either in the strict sense or wide sense with auto
correlation function ( ) R t , then the Fourier transform of ( ) R t is called the power
spectral density function of ( ) { }
X t and denoted by ( )
XX
S e or ( ) S e or ( )
X
S e
Thus ( ) ( )
i
S R e d
et
e t t

=
}
Problem 8. State any two properties of the power spectral density function.
Solution:
(i). The spectral density function of a real random process is an even function.
Unit.4. Correlation and spectral densities
3
(ii). The spectral density of a process ( ) { }
X t , real or complex, is a real function of e and
non-negative.
Problem 9. State Wiener-Khinchine Theorem.
Solution:
If ( )
T
X e is the Fourier transform of the truncated random process defined as
( )
( )
0
T
X t for t T
X t
for t T
s

=

>

Where ( ) { }
X t is a real WSS Process with power spectral density function ( ) S e , then
( ) ( )
{ }
2 1
2
T
Lt
S E X
T T
e e
(
=
(


Problem 10. If ( )
( ) 2
R e
t
t

= is the auto Correlation function of a random
process ( ) { }
X t , obtain the spectral density of ( ) { }
X t .
Solution:
( ) ( )
i
S R e d
et
e t t

=
}
( )
2
e cos isin d
t
et et t

=
}
2
0
2 e cos d
t
et t

=
}
( )
2
2 2
0
2
2
4
e
cos sin
t
et e et
e

(
= +
(
+

( )
2 2
4
4
S

e
e
=
+
Problem 11. The Power spectral density of a random process
( ) { }
X t is given by
( )
, 1
0,
XX
S
elsewhere
t e
e
<

Find its autocorrelation function.


Solution:
( ) ( )
1
2
i
XX XX
R S e d
et
t e e
t

=
}
1
1
1
2
i
e d
et
t e
t

=
}
1
1
1
2
i
e
i
et
t

(
=
(

1
2
i i
e e
i
t t
t

(
=
(

Unit.4. Correlation and spectral densities
4
1
2
i i
e e
i
t t
t

(
=
(

sint
t
=
Problem 12. Define cross-Spectral density.
Solution:
The process ( ) { }
X t and ( ) { }
Y t are jointly wide-sense stationary with the cross-
correlation function ( )
XY
R t , then the Fourier transform of ( )
XY
R T is called the cross
spectral density function of ( ) { }
X t and ( ) { }
Y t denoted as ( )
XY
S e
Thus ( ) ( )
i
XY XY
S R e d
et
e t t

=
}
Problem 13. Find the auto correlation function of a stationary process whose power
spectral density function is given by ( )
2
1
0 1
for
s
for
e e
e
e
s

=

>

.
Solution:
( ) ( )
1
2
i
R S e d
et
t e e
t

=
}
( )
1
2
1
1
cos sin
2
i d e et et e
t

= +
}
1
1
2 2
2 3
1 0
1 sin cos sin
.cos 2 2 d
et et et
e et e e e
t t t t

( | | | | | |
= = +
| | | (
\ . \ . \ .
}
( )
2 3
1 sin 2cos 2sin
R
t t t
t
t t t t
(
= +
(

Problem 14. Given the power spectral density : ( )
2
1
4
xx
S e
e
=
+
, find the average power
of the process.
Solution:
( ) ( )
1
2
i
R S e d
et
t e e
t

=
}
2
1 1
2 4
i
e d
et
e
t e

(
=
(
+

}
Hence the average power of the processes is given by
( ) ( )
2
0 E X t R ( =

2
1
2 4
de
t e

=
+
}
2 2
0
1
2
2 2
de
t e

=
+
}
Unit.4. Correlation and spectral densities
5
1
0
1 1
tan
2 2
e
t

( | |
=
| (
\ .
1 1
0
4 4
t
t
(
= =
(

.
Problem 15. Find the power spectral density of a random signal with autocorrelation
function e
t
.
Solution:
( ) ( )
i
S R e d
et
e t t

=
}
( ) cos sin e i d
t
et et t

=
}
0
2 cos e d
t
et t

=
}
( )
2 2
0
2 cos sin
e
t
et e et
e

(
= +
(
+

( )
2 2 2 2
1 2
2 0

e e
(
= =
(
+ +

PART-B
Problem 16. a). If ( ) { }
X t is a W.S.S. process with autocorrelation function ( )
XX
R t and
if ( ) ( ) ( ) Y t X t a X t a = + .Show that ( ) ( ) ( ) ( ) 2 2 2
YY XX XX XX
R R R a R a t t t t = + .
Solution:
( ) ( ) ( )
yy
R E y t y t t t = + (

( ) ( ) ( ) ( ) { }
E X t a X t a X t a X t a t t = + + + + ( (

( ) ( ) ( ) ( )
( ) ( ) ( ) ( )
E X t a X t a E X t a X t a
E X t a X t a E X t a X t a
t t
t t
= + + + + + ( (

+ + + + ( (

( ) ( ) ( )
( ) ( ) ( )
2
2
xx
xx
R E X t a X t a a
E X t a X t a a R
t t
t t
= + + + (

+ + + (

( ) ( ) ( ) 2 2 2
xx xx xx
R R a R a t t t = +
b). Assume a random signal ( ) ( ) ( ) Y t X t X t a = + where ( ) X t is a random signal and
' ' a is a constant. Find ( )
YY
R t .
Solution:
( ) ( ) ( )
YY
R E Y t Y t t t = + (

( ) ( ) ( ) ( ) { }
E X t X t a X t X t a t t = + + + + ( (

Unit.4. Correlation and spectral densities
6
( ) ( ) ( ) ( )
( ) ( ) ( ) ( )
E X t X t E X t X t a
E X t a X t E X t a X t a
t t
t t
= + + + ( (

+ + + + ( (

( ) ( ) ( ) ( )
xx xx xx xx
R R a R a R t t t t = + + + +
( ) ( ) ( ) ( ) 2
yy xx xx xx
R R R a R a t t t t = + + +
Problem 17. a). If ( ) { }
X t and ( ) { }
Y t are independent WSS Processes with zero means,
find the autocorrelation function of ( ) { }
Z t , when ( ) ( ) ( ) ( ) i Z t a bX t CY t = + + ,
( ) ( ) ( ) ( ) ii Z t aX t Y t = .
Solution:
Given ( ) 0 E X t = (

, ( ) 0 E Y t = (

---------(1)
{ ( )} X t and { ( )} Y t are independent
( ) ( ) { } ( ) ( ) 0 E X t Y t E X t Y t = = (

-------(2)
(i). ( ) ( ) ( )
ZZ
R E Z t Z t t t = + (

( ) ( ) ( ) ( ) {
E a bX t cY t a bX t cY t t t = + + + + + + ( (

( ) ( ) ( ) ( ) ( ) {
( ) ( ) ( ) ( ) ( ) ( ) ( )}
2 2
2
E a ab X t acY t baX t b X t X t
bc X t Y t caY t cbY t X t c y t y t
t t t
t t t
= + + + + + + +
+ + + + + + +
( ) ( ) ( ) ( ) ( ) ( )
( ) ( ) ( ) ( ) ( ) ( ) ( )
2 2
2
E a abE X t acE Y t baE X t b E X t X t
bc E X t y t caE Y t cbE Y t X t c E Y t Y t
t t t
t t t
= + + + + + + + ( ( ( (

+ + + + + + + ( ( ( (

( ) ( )
2 2 2
XX YY
a b R c R t t = + +
( ) ( ) ( )
ZZ
R E Z t Z t t t = + (

( ) ( ) ( ) ( ) E aX t Y t aX t Y t t t = + + (

( ) ( ) ( ) ( )
2
E a X t X t Y t Y t t t ( = + +

( ) ( ) ( ) ( )
2
a E X t X t E Y t Y t t t = + + ( (

( ) ( ) ( )
2
ZZ XX YY
R a R R t t t =
b). If ( ) { }
X t is a random process with mean 3 and autocorrelation ( )
0.2
9 4
xx
R e
t
t

= + .
Determine the mean, variance and covariance of the random variables
( ) 5 Y X = and
( ) 8 Z X = .
Solution:
Given ( ) ( ) ( ) 5 & 8 , 3 Y X Z X E X t = = = (

--------- (1)
Mean of
| | ( ) 5 3 Y E Y E X = = = (

Mean of
| | ( ) 8 3 Z E Z E X = = = (

We know that
Unit.4. Correlation and spectral densities
7
( ) ( ) ( )
2
2
Var Y E Y E Y = (

( ) ( ) ( )
2 2
5 E Y E X =
But ( ) ( )
2
0
XX
E X t R ( =

0.2101
9 4e

= +
9 4 13 = + =
Thus ( ) ( )
2
13 3 13 9 4 Var Y = = =
( ) ( ) ( )
2
2
Var Z E Z E Z = (

( ) ( )
2 2
8 8 E Z E X Z X ( ( = = (


( ) 0
XX
R =
9 4 13 = + =
Hence ( ) ( )
2
13 3 13 9 4 Var Z = = =
| | ( )
0.25 8
5,8 9 4 E YZ R e

= = +
( )
( )
1 2
0.2
1 2
, 9 4
t t
R t t e

= +
0.6
9 4e

= +
Covariance ( ) ( ) ( )
1 2 1 2
, R t t E X t E X t = ( (

( ) | | | | 5, 8 5 8 R E E =
( )
0.6 0.6
9 4 3 3 4 2.195 e e

= + = =
Problem 18. a). The autocorrelation function for a stationary process is given by
( ) 9 2
xx
R e
t
t

= + . Find the mean value of the random variable ( )
2
0
Y X t dt =
}
and
variance of ( ) X t .
Solution:
Given ( ) 9 2
xx
R e
t
t = +
Mean of ( ) X t is given by
( ) ( )
2 2
xx
Lt
X E X t R t
t
= = (


( )
9 2
Lt
e
t
t

= +

2
9 X =
3 X =
Also ( ) ( )
0 2
0 9 2 9 2 11
XX
E X t R e ( = = + = + =

( ) { } ( ) ( ) ( )
2
2
Var X t E X t E X t ( ( =

2
11 3 11 9 2 = = =
Mean of ( ) ( ) Y t E Y t = (

Unit.4. Correlation and spectral densities
8
( )
2
0
E X t dt
(
=
(

}
( )
2
0
E X t dt = (
}
2
0
3dt =
}
( )
2
0
3 6 t = =
( ) 6 E Y t = (

b). Find the mean and autocorrelation function of a semi random telegraph signal process.
Solution:
Semi random telegraph signal process.
If ( ) N t represents the number of occurrences of a specified event in ( ) 0, t
and ( ) ( )
( )
1
N t
X t = , then ( ) { }
X t is called the semi random signal process and ( ) N t is a
poisson process with rate .
By the above definition ( ) X t can take the values 1 = and 1 only
( ) ( ) 1 P X t P N t is even = = ( (

( )
!
k
t
K even
e t
K

=
=

( )
2
1 ...
2
t
t
e

(
= + + (
(

( ) 1
t
P X t e cosh t

= = (

( ) ( ) 1 P X t P N t is odd = = ( (

( )
!
k
t
K odd
e t
K

=
=

( )
3
...
3!
t
t
e t

(
= + + (
(

t
e sinh t

=
( ) { } ( ) ( )
1,1 K
Mean X t K P X t K
=
= =

1 ( 1)
t t
e cosh t e sinh t



= +
| |
t
e cosh t sinh t

=
2 t t
e cosh t sinh t e



( = =

( ) ( ) ( ) R E X t X t t t = + (

( ) ( ) ( ) ( ) 1 1 1 1 P X t X t P X t X t t t = + = + + = ( (

Unit.4. Correlation and spectral densities
9
( ) ( )
! !
n n
t t
n even n odd
e e
n n

t t

= =
=

e cosh e sinh
t t
t t

=
| | e cosh sinh
t
t t

=
e e
t t
=
( )
2
R e
t
t

=
Problem 19. a). Find Given the power spectral density of a continuous process as
( )
2
4 2
9
5 4
XX
S
e
e
e e
+
=
+ +
find the mean square value of the process.
Solution:
We know that mean square value of ( ) { }
X t
( ) { } ( )
2
1
2
XX
E X t S d e e
t

= =
}
( )
2
4 2
1 9
2 5 4
d
e
e
t e e

+
=
+ +
}
2
4 2
0
1 9
.2
2 5 4
d
e
e
t e e

+
=
+ +
}
2
4 2 2
0
1 9
4 4
d
e
e
t e e e

+
=
+ + +
}
( ) ( )
2
2 2 2
0
1 9
1 4 1
d
e
e
t e e e

+
=
+ + +
}
i.e., ( ) { }
( )( )
2
2
2 2
0
1 9
4 1
E X t d
e
e
t e e

+
=
+ +
}
let
2
u e =
We have
( )( )
( )( )
2
2 2
9 9
4 1 4 1
u
u u
e
e e
+ +
=
+ + + +
( ) ( )
4 9 1 9
5 8
4 1 1 4
4 1 3 4 3 1 u u u u
+ +
+ +
= + = +
+ + + +
.Partial fractions
i.e.,
( )( ) ( ) ( )
2
2 2 2 2
9 5 8
4 1 3 4 3 1
e
e e e e
+
= +
+ + + +
From (1),
( ) { }
( ) ( )
2
2 2
0
1 1 5 8
3 4 1
E X t de
t e e

( = +
+ +
(

}
Unit.4. Correlation and spectral densities
10
1 1
0
1 1
5. 8
3 2 2
tan tan
e
e
t


(
= +
(

1 5
8 0
3 2 2 2
t t
t
( | | | |
= +
| | (
\ . \ .
1 5 1 11
. 8
3 2 2 6 2
t
t
| | | |
= + =
| |
\ . \ .
( ) { }
2
11
12
E X t = .
b). If the 2n random variables
r
A and
r
B are uncorrelated with zero mean and
( ) ( )
2 2 2
r r r
E A E B o = = . Find the mean and autocorrelation of the process
( )
1
cos sin
n
r r r r
r
X t A t B t e e
=
= +

.
Solution:
Given ( ) ( ) ( ) ( )
2 2 2
0 &
r r r r r
E A E B E A E B o = = = =
Mean: ( )
1
n
r r r r
r
E X t A cos t B sin t e e
=
(
= + (
(

( ) ( )
1
n
r r r r
r
E A cos t E B sin t e e
=
= + (

( ) 0 E X t = (

( ) ( ) 0
r r
E A E B = =
Autocorrelation function:
( ) ( ) ( ) R E X t X t t t = + (

( ) ( ) ( ) ( )
1 1
n n
r r r r s s s s
r s
E A cos t B sin t A cos t B sin t e e e t e t
= =

= + + + +

Given 2n random variables


r
A and
r
B are uncorrelated
| | | | | | | | , , , ,
r s r s r s r s
E A A E A B E B A E B B are all zero for r s =
( ) ( ) ( ) ( )
2 2
1
n
r r r r r r
r
E A cos t cos t t E B sin tsin t e e t e e t
=
= + + +

( )
2
1
n
r r
r
cos t t o e t
=
=

( )
2
1
n
r r
r
cos o e t
=
=

( )
2
1
n
r r
r
R cos t o e t
=
=

Problem 20. a). If ( ) { }


X t is a WSS process with autocorrelation ( ) R Ae
o t
t

= ,
determine the second order moment of the random variable
( ) ( ) 8 5 X X .
Solution:
Unit.4. Correlation and spectral densities
11
Given ( ) R Ae
o t
t

=
( )
1 2
1 2
,
t t
R t t Ae
o
=
( ) ( )
0 2
0
XX
E X t R Ae A
o
( = = =

( ) ( ) ( ) ( ) ( ) ( )
2
2 2
8 5 8 5 2 8 5 E X X E X E X E X X ( ( = + ( (

( ) ( )
2 2
8 5 E X E X A ( ( = =

Also ( ) ( ) ( )
8 5 3
8 5 8, 5 E X X R Ae Ae
o o
= = = (

Substituting (2), (3) in (1) we obtain
( ) ( )
2
3
8 5 2 E X X A A Ae
o
= + (

3
2 2 A Ae
o
=
( )
3
2 1 A e
o
=
.
b). Two random process ( ) { }
X t and ( ) { }
Y t are given by
( ) ( ) ( ) ( ) cos t , sin X t A Y t A t e u e u = + = + where A & e are constants and u is a
uniform random variable over 0 to 2t . Find the cross-correlation function.
Solution:
( ) ( ) ( )
XY
R E X t Y t t t = + (

( ) ( ) ( )
E Acos t Asin t e u e t u ( = + + +

( ) ( )
2
A E sin t cos t e et u e u = + + + (

u is a uniform random variables ( )
1
, 0 2
2
f
u
u u t
t
= s s
( ) ( ) ( )
2 2
0
2
XY
A
R sin t cos t d
t
t e et u e u u
t
= + + +
}
( ) ( )
2 2
0
2 2
2 2
sin t sin t A
d
t
e et u e
u
t
+ + + | |
=
|
\ .
}
( )
( )
2
2
0
2 2
4 2
cos t A
sin t
t
e et u
u e
t
+ + (
= +
(

| |
2
0 2
4
A
sin t et
t
= +
2
2
A
sinet =
Problem 21. a). Find the cross-correlation function of
( ) ( ) ( ) ( ) ( ) ( ) & W t A t B t Z t A t B t = + = where ( ) A t and ( ) B t are statistically
independent random variables with zero means and autocorrelation function
, , 3
AA BB
R e R e
t t
t t t t

= < < = < < respectively.
Solution:
Unit.4. Correlation and spectral densities
12
Given ( ) ( ) 0, 0 E A t E B t = = ( (

( ) ( ) & A t B t are independent
( ) ( ) ( ) ( ) ( ) 0
AB
R E A t B t E A t E B t t t t = + = + = ( ( (

Similarly ( ) ( ) ( ) ( ) ( ) 0
BA
R E B t A t E B t E A t t t t = + = + = ( ( (

( ) ( ) ( )
WZ
R E W t Z t t t = + (

( ) ( ) ( ) ( ) { }
E A t B t A t B t t t = + + + ( (

( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) E A t A t A t B t B t A t B t B t t t t t = + + + + + (

( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) E A t A t E A t B t E B t A t E B t B t t t t t = + + + + + ( ( ( (

( ) ( ) ( ) ( )
AA AB BA BB
R R R R t t t t = +
( ) ( )
AA BB
R R t t =
3 e e
t t
=
( ) 2
WZ
R e
t
t

=
b). The random processes ( ) { }
X t and ( ) { }
Y t defined by
( ) ( ) cos sin , cos sin X t A t B t Y t B t A t e e e e = + = where A & B are uncorrelated zero
mean random variables with same variance find its autocorrelation function.
Solution:
Given ( ) ( ) ( ) ( )
2 2 2
0 & E A E B E A E B o = = = =
Also A and B are uncorrelated i.e., ( ) 0 E AB =
( ) ( ) ( )
XY
R E X t Y t t t = + (

| | ( ) ( ) { }
E Acos t Bsin t Bcos t Asin t e e e t e t = + + + (

( ) ( )
( ) ( )
2
2
ABcos tcos t A cos tsin t
E
B sin tcos t BAsin tsin t
e e t e e t
e e t e e t
( + +
= (
+ + +
(

( ) ( ) ( ) ( )
( ) ( ) ( ) ( )
2
2
E AB cos tcos t E A cos tsin t
E B sin tcos t E BA sin tsin t
e e t e e t
e e t e e t
= + +
+ + +
( ) ( ) ( ) ( ) ( ) ( )
2 2
0 E B sin tcos t E A cos tsin t E AB e e t e e t = + + =
( ) ( ) ( ) ( )
2 2 2 2
sin tcos t cos tsin t E A E B o e e t e e t o
(
= + + = = (

( )
2
sin t t o e t =
( )
2
sin o e t =
( ) ( )
2
XY
R sin t sin sin t o e u u = = (

Unit.4. Correlation and spectral densities


13
Problem 22. a). Consider two random processes ( ) ( ) 3cos X t t e u = + and
( ) 2cos
2
Y t t
t
e u
| |
= +
|
\ .
where u is a random variable uniformly distributed in
( ) , 2 u t . Prove that ( ) ( ) ( ) 0 0
XX YY XY
R R R t > .
Solution:
Given ( ) ( ) 3 X t cos t e u = +
( ) 2
2
Y t cos t
t
e u
| |
= +
|
\ .
( ) ( ) ( )
XX
R E X t X t t t = + (

( ) ( ) 3 3 E cos t cos t e u e et u = + + + (

( ) ( )
9
2 2
2
E cos t cos e et u et = + + + (

u is uniformly distributed in ( ) 0, 2t , ( )
1
2
f u
t
=
( )
2
0
9 1
2 2
2 2
cos t cos d
t
e et u et u
t
= + + + (
}
( )
2
0
2 2 9 1
2 2 2
sin t
cos
t
e et u
u et
t
+ + (
= +
(

| |
( )
2
0
2 2 9
0 2 0
4 2
sin t
cos t
t
e et u
t e
t
(
+ + (
( = + =
(
(

( )
9
2
XX
R cos t et =
( )
9
0
2
XX
R =
( ) ( ) ( )
YY
R E Y t Y t t t = + (

2 2
2 2
E cos t cos t
t t
e u e et u
( | | | |
= + + +
| | (
\ . \ .
( )
4
2 2
2
E cos t cos e et u t et = + + + (

( )
2
0
1
2 2 2
2
cos cos d
t
et et u t et u
t
= + + + (
}
( )
2
0
2 2 1
2
sin t
cos
t
e et u t
u et
t
+ + (
= +
(

2coset =
Unit.4. Correlation and spectral densities
14
Similarly, ( ) 3
2
XY
R cos
t
t et
| |
=
|
\ .
( )
max
3
XY
R t =
By property (2) i.e., ( ) ( ) ( ) 0 0
XX YY XY
R R R t >
9
.2 3 ,
2 2
cos
t
et t
| |
>
|
\ .
In this care ( )
XY
R t takes on its maximum possible value of 3 at
2
,
2
n
n
t t
t
e e
= +
Since it is periodic
( ) ( ) ( )
2
max
0 0
XX YY XY
R R R t =
b). State and prove Wiener Khinchine Theorem
Solution:
Statement:
If ( )
T
X e is the Fourier transform of the truncated random process defined as
( )
( ),
0 ,
T
X t T t T
X t
otherwise
s s

Where ( ) { }
X t is a real WSS process with power spectral function ( ) ,
XX
S e then
( ) ( )
{ }
2 lim
1
2
XX T
S E X
T T
e e
(
=
(


Proof:
Given ( ) ( )
i t
T T
X X t e dt
e
e

=
}
( )
T
i t
T
X t e dt
e

=
}
Now
( ) ( ) ( )
2
T T T
X X X e e e =
( ) ( )
1 2
1 1 2 2
T T
i t i t
T T
X t e dt X t e dt
e e

=
} }
( ) ( )
( )
1 2
1 2 1 2
T T
i t t
T T
X t X t e dt dt
e

=
} }
( )
{ }
( ) ( )
( )
1 2
2
1 2 1 2
T T
i t t
T
T T
E X E X t X t e dt dt
e
e


= (
} }
( )
( )
1 2
1 2 1 2
T T
i t t
T T
R t t e dt dt
e

=
} }
Unit.4. Correlation and spectral densities
15
( )
1 2 1 2
0
T T
T T
t t dt dt

=
} }
----------- (1)
{ ( )} X t is WSS its autocorrelation is a function of time difference ie.,
1 2
t t =
T
( ) , A T T B
T T
Q
1 2
t t = d
S
P
D R ( ) , C T T
T
The double integral (1) is evaluated over the area of the square ABCD bounded by the
1
, t T T = &
2
, t T T = as shown in the figure
We divide the area of the square into a number of strips like PQRS, where PQ is given by
1 2
t t t = and RS is given by
1 2
t t d t t = + .
When PQRS is at A,
1 2
2 t t T T T = = and
1 2
2 t t T = and at C
t varies from 2T to 2T such that the area ABCD is covered.
Now
1 2
dt dt = elemental area of the
1 2
, t t plane
=Area of the strip PQRS------- (2)
At P ,
2
t T = ,
1 2
t t T t t = + =
At Q,
1
t T = ,
2 1
t t T = =
2 , 0 PC CQ T t t = = >
2 , 0 PC CQ T t t = = + <
2 , 0 RC SC T d t t t = = >
When 0 t > ,
Area of PQRS PCQ RCS = A A
( ) ( )
2 2 1 1
2 2 2
2 2
T T t t t =
( ) 2T d t t = Omitting ( )
2
dt ------------- (3)
From (2) & (3)
Unit.4. Correlation and spectral densities
16
( )
1 2
2 (4) dt dt T d t t =
Using (4) in (1)
( )
{ }
( )( )
2
2
2
0 2
T
T
T
E X T d e t t t

=
}
( )
{ }
( )
2
2
2
1
0 1
2 2
T
T
T
E X d
T T
t
e t t

| |
=
|
\ .
}
( )
{ }
( )
2
2
2
lim lim lim
1
0
2 2
T
T
T
E X w d d
T T T T T
t
t t t

=

} }
( ) 0 d t t

=
}
( )
i
R e d
et
t t

=
}
( ) S e =
This theorem provides an alternative method for finding ( ) S e for a WSS Process.
Problem 23. a). Define the spectral density ( ) S e of a real valued stochastic
processes ( ) { }
: 0 X t t > , obtain the spectral density of ( ) { } ( ) ( ) : 0 , Y t t Y t X t o > = when
o is increment of
( ) X t such that ( ) ( )
1
1 1
2
P P o o = = = = .
Solution:
If ( ) X t is a stationary process {either in the strict serve or wide sense with
autocorrelation function R t , then the Fourier transform of R t is called the Power
spectral density function of ( ) X t denoted by ( ) S
Thus ( ) ( )
i
S R e d
et
e t t

=
}
Given { } { }
1
1 1
2
P P o o = = = =
( ) ( )
1 1
1. 1 . 0
2 2
E o = + =
( ) ( ) ( )
2 2
2
1 1
1 . 1 . 1
2 2
E o = + =
( ) ( ) E Y t E X t o = ( (

( ) ( ) E E X t o = (

( ) X t areindependent o (

( ) ( ) 0 0 E Y t E o = = ( (

( ) ( ) ( )
YY
R E Y t Y t t t = + (

( ) ( ) E X t X t o o t = + (

Unit.4. Correlation and spectral densities
17
( ) ( )
2
E X t X t o t ( = +

( ) ( )
2
E E X t X t o t ( = + (

( ) ( ) 1 E X t X t t = + (

( ) ( ) ( )( ) (1) 1 ( 1) 1 P X t t P X t t t t = + = + + = ( (

( ) ( )
(1) ( 1)
! !
n
T T
n even n odd
e T e T
n n



= =
= +

T T
e cosh e sinh

t t

=
| | e cosh T sinh T
t

=
2 T T T
e e e

= =
( ) ( )
i
S R e d
et
e t t

=
}
2 i
e e d
t et
t

=
}
0
2 2
0
i i
e e d e e d
t et t et
t t

= +
} }
( ) ( )
0
2 2
0
i i
e d e d
e t e t
t t

= +
} }
( ) ( )
( )
0
2 2
0
2 2
i i
e e
i i
e t e t
e e

( (
= +
( (
+

1 1
2 2 i i e e
= +
+
2 2
2 2
4
i i e e
e
+
=

2 2
4
4

e
=
+
b). Show that (i) ( ) 0 S e > & (ii) ( ) ( ) S S e e = where ( ) S e is the spectral density of a
real valued stochastic process.
Solution:
(i) ( ) 0 S e >
Proof:
If possible let
( ) 0 S e < at
0
e e =
i.e., let ( ) 0 S e < in
0 0
2 2
c c
e e e < < + , where c is very small
Let us assume that the system function of the convolution type linear system is
Unit.4. Correlation and spectral densities
18
( )
0 0
1 ,
2 2
0 ,
H
elsewhere
c c
e e e
e

< < +

Now
( ) ( ) ( )
2
YY XX
S H S e e e =
( )
0 0
,
2 2
0 ,
XX
S
elsewhere
c c
e e e e

< < +

( ) ( )
2
0
YY
E Y t R ( =

( )
1
2
YY
S d e e
t

=
}
( )
0
0
2
2
1
2
XX
S d
c
e
c
e
e e
t
+

=
}
( )
0
2
XX
S
c
e
t
=
[ ( )
XX
S e Considered a constant ( )
0 XX
S e band is
narrow] ( ) ( )
2
0
0, 0
xx
E Y t S e ( > >

which is contrary to our initial assumption.
( ) 0
XX
S e > everywhere, since
0
e e = is arbitrary
(ii) ( ) ( ) S S e e =
Proof:
Consider ( ) ( )
i
S R e d
et
e t t

=
}
Let u t = then d du t = and u varies from and
( ) ( ) ( )
i u
S R u e du
e
e

=
}
( )
i u
R u e du
e

=
}
( ) ( )
i u
R u e du R is aneven functionof
e
t t

= (
}

( ) S e =
Hence ( ) S e is even function of e
Problem 24. a). An autocorrelation function R (t ) of
( ) { }
; X t t < < in given by
, 0, 0 ce c
o t
o

> > . Obtain the spectral density of R (t ).


Solution:
Given ( ) R ce
ot
t

=
Unit.4. Correlation and spectral densities
19
By definition
( ) ( )
i
S R e d
et
e t t

=
}
i
ce e d
o t et
t

=
}
( ) ( )
0
i i
ce e d ce e d
o t o t et et
t t

= +
} }
( ) ( )
0
0
i i
c e d e d
o e t o e t
t t

(
= +
(

} }
( ) ( )
( )
0
0
i i
e e
c
i i
o e t o e t
o e o e


( (

= +
` ( (
+

)
1 1
c
i i o e o e
(
= +
(
+

2 2
i i
c
o e o e
o e
+ + (
=
(


( )
2 2
2 c
S
o
e
o e
=
+
b). Given that a process ( ) X t has the autocorrelation function
( ) ( )
0
cos
XX
R Ae
o t
t e t

= where 0 A , 0 and
0
e are real constants, find the power
spectrum of ( ) X t .
Solution:
By definition ( ) ( )
i
S R e d
et
e t t

=
}
( )
0
i
Ae cos e d
o t et
e t t

=
}
( ) | |
0
A e cos cos isin d
o t
e t et et t

=
}
( )
0 0
A e cos cos A e cos sin d
o t o t
e t et e t et t



= +
} }
0
0
2A e cos cos d
ot
e t et t

=
}
( ) ( )
0 0
2
2
cos cos
A e d
ot
e e t e e t
t

+ + (
=
(

}
Using
2 2
0
ax
a
e cosbxdx
a b

=
+
}
Unit.4. Correlation and spectral densities
20
( )
( ) ( )
2 2
2 2
0 0
S A
o o
e
o e e o e e
(
= + (
+ + +
(

( ) ( )
2 2
2 2
0 0
1 1
Ao
o e e o e e
(
= + (
+ + +
(

Problem 25. a). Find the power spectral density of the random binary transmission
process whose autocorrelation function is ( )
1 for 1
0 elsewhere
R
t t
t
s

.
Solution:
By definition ( ) ( )
i
S R e d
et
e t t

=
}
1
1
1
i
e d
et
t t

= (
}
| | | |
0 1
1 0
1 1
i i
e d e d
et et
t t t t

= + +
} }
| | | |
0 1
1 0
1 1
i i
e d e d
et et
t t t t

= + +
} }
( ) ( ) ( )
0 1
2 2 2 2
1 0
1 1 1
i i i i
e e e e
i i i i
et et et et
t t
e e e e

( (
= + +
( (


2 2
2
i i
e e
e e
e e

(
=
(

| |
2
1
2 2cose
e
=
( )
2
2
1 cose
e
=
2
2
2
1 1
2
sin
e
e
( | |
=
| (
\ .
2
2
2
2
2
sin
e
e
(
=
(

2
4
2
sin
e
e
=
( )
2
2
2
sin
S
e
e
e
(
(
=
(
(

Unit.4. Correlation and spectral densities
21
b). Show that the power spectrum of the autocorrelation function 1 e
o t
o t

+ (

is
( )
3
2
2 2
4
w
o
o +
.
Solution:
By definition, ( ) ( )
i
S R e d
et
e t t

=
}
1
i
e e d
o t et
o t t

= + (
}
( )
( )
( )
( )
0
0
1 1
i i
e d e d
o e t o e t
ot t ot t

= + +
} }
( )
( )
( )
( )
( )
( )
( )
( )
( )
( )
( )
0
2 2
0
1 1
i i i i
e e e e
i i
i i
o e t o e t o e t o e t
ot o ot o
o e o e
o e o e

+ +

( (
= + + ( (
+
+
( (

( ) ( )
2 2
1 1
0
i i
i i
o o
o e o e
o e o e
( (
= + + + + ( (
+
+
( (

( ) ( )
2 2
1 1 1 1
i i
i i
o
o e o e
o e o e
(
(
= + + + (
(
+
+
(

( ) ( )
( )
2 2
2 2 2
2 2
2 i i o e o e o
o
o e
o e
(
+ +
(
= +
( +
+

( )
( )
2 2
2 2 2
2 2
2
2
o o e
o
o e
o e

= +
+
+
( ) ( )
( )
2 2 2 2
2
2 2
2 2 o o e o o e
o e
+ +
=
+
( )
( )
2 2 2 2
2
2 2
2o o e o e
o e
+ +
=
+
( )
3
2
2 2
4o
o e
=

Problem 26. a). Find the spectral density function whose autocorrelation function is
given by
( )
1 ;
0 ; elsewhere
T
R
t
t
t
t

s
=

Solution:
Unit.4. Correlation and spectral densities
22
( ) ( )
i
S R e d
et
e t t

=
}
1
T
i
T
e d
et
t
t
t

(
=
(

}
| | 1
T
T
cos isin d
T
t
et et t

| |
=
|
\ .
}
1
T
T
cos d
T
t
et t

| |
=
|
\ .
}
0
2 1
T
cos d
T
t
et t
| |
=
|
\ .
}
0
2 1
T
cos d
T
t
et t
| |
=
|
\ .
}
2
0
1
2 1
T
sin cos
T
t et et
e t e
( | | | || |
=
| | | (
\ . \ .\ .
2 2
1
2
cos T
T T
e
e e
(
= +
(

| |
2
2
1 cos T
T
e
e
=
2
2
2
2
2
T
sin
T
e
e
| |
=
|
\ .
( )
2
2
4
2
T
sin
S
T
e
e
e
| |
|
\ .
=
b). Find the power spectral density function whose autocorrelation function is given by
( ) ( )
2
0
2
A
R cos t e t =
Solution:
( ) ( )
i
S R e d
et
e t t

=
}
( )
2
0
2
i
A
cos e d
et
e t t

=
}
( )| |
2
0
2
A
cos cos isin d e t et et t

=
}
( ) ( ) ( ) ( )
2
0 0 0 0
4
A
cos cos d i sin sin d e e t e e t t e e t e e t


(
= + + + + ( (
(


} }
Unit.4. Correlation and spectral densities
23
( ) ( ) ( ) ( )
2
0 0 0 0
4
A
cos isin d cos isin d e e t e e t t e e t e e t t

= + + + ( (

} }
( ) ( )
0 0
2
4
i i
A
e d e d
e e t e e t
t t

+

(
= +
(

} }
By definition of Dirac delta function
( )
1
2
i
S e d
et
e t
t

=
}
( ) 2
i
e d
et
to e t

=
}
(1) reduces to
( ) ( ) ( )
2
0 0
2 2
4
A
S e to e e to e e = + + (

( ) ( ) ( )
2
0 0
4
A
S
t
e o e e o e e = + + (

Problem 27. a). If the power spectral density of a WSS process is given by
( )
( )
;
0 ;
b
a a
a S
a
e e
e
e

>

find its autocorrelation function of the process.


Solution:
( ) ( )
1
2
i
R S e d
et
t e t
t

=
}
( )
1
2
a
i
a
b
a e d
a
et
e t
t

=
}
( )( )
1
2
a
a
b
a cos isin d
a
e et et e
t

= +
}
( )
0
2
2
a
b
a cos d
a
e et e
t
=
}
( )
2
0
a
b sin cos
a
a
et et
e
t t t
(
=
(

2 2
1 b cosa
a
t
t t t
(
= +
(

| |
2
1
b
cosa
a
t
t t
=
2
2
2
2
b a
sin
a
t
t t
| |
=
|
\ .
Unit.4. Correlation and spectral densities
24
b).The power spectral density function of a zero mean WSS process ( ) { }
X t is given
by ( )
0
1 ;
0 ; elsewhere
S
e e
e
<

. Find ( ) R t and show also that ( )


0
& X t X t
t
e
| |
+
|
\ .
are
uncorrelated
Solution:
( ) ( )
1
2
i
R S e d
et
t e e
t

=
}
0
0
1
2
i
e d
e
et
e
e
t

=
}
0
0
1
2
i
e
i
e
et
e
t t

(
=
(

0 0
1
2
i i
e e
e t e t
tt

(
=
(

0
sine t
tt
=
To show that
( ) X t &
0
X t
t
e
| |
+
|
\ .
are uncorrelated we have to show that the auto
covariance is zero.
i.e., ( )
0
0 C X t X t
t
e
( | |
+ =
( |
\ .
Consider,
( ) ( )
0 0 0
XX
C X t X t R E X t E X t
t t t
e e e
( ( | | | | | |
+ = + (
( ( | | |

\ . \ . \ .
0
0
XX
R
t
e
| |
=
|
\ .
( )
0
0
sin
C X t X t
t t
e tt
( | |
+ = =
( |
\ .
Hence ( ) X t &
0
X t
t
e
| |
+
|
\ .
are uncorrelated.
Problem 28. a). The power spectrum of a WSS process ( ) { }
X t is given
by ( )
( )
2
2
1
1
S e
e
=
+
. Find the autocorrelation function and average power of the process.
Solution:
( ) ( )
1
2
i
R S e d
et
t e e
t

=
}
Unit.4. Correlation and spectral densities
25
( )
2
2
1 1
2
1
i
e d
et
e
t
e

=
+
}
------- (1)
The integral in (1) is evaluated by contour integration technique as given below.
Consider
( )
2
,
1
iaz
C
e dz
z +
}
where C is the closed contour consisting of the real axis from R to
R and the upper half of the
le
O , Z R =
2
1 0 Z + =
2
1 Z =
Z i =
Z i = is a double pole.
[Res] Z i =
( )
( ) ( )
2 1
1 !
Lt
d
z i z
z i z dz
(
= +


( )
( )
2
2
2
1
1!
1
iaz
Lt
d e
z i
z i dz
z
(
(
=
(
+

( )
2
iaz
Lt
d e
z i dz
z i
=

+
( ) ( )
( )
2
4
2 .1
iaz iaz
Lt z i i ae e z i
z i
z i
(
+ +
= (

+
(

( )
( )
3
2
iaz iaz
Lt z i i ae e
z i
z i
(
+
= (

+
(

( )
( )
( ) ( ) 2 2 2 1 1
8 8 4
a a a
e a e a e a
i i i

+ +
= = =

By Cauchy residue theorem, ( ) ( )


1
2
n
i i
c
Res
f z dz i f z
z a

=
=
=
_
]
and taking limits R
and using Jordans lemma
( )
( ) ( )
2
2
2 1 1
4 2
1
a a iax
ie a e a
e
dx
i
x


= =

]
----- (2)
Using (2) in (1)
( )
( ) ( ) 1 1
1
2 2 4
e e
R
t t
t t t
t
t

+ +
= =
Average power of ( ) { } ( )
0
X t R
t
t
=
= (

( )
0
1 1
0.25
4 4
e
t
t
t

=
( +
= = =
(

Unit.4. Correlation and spectral densities
26
b). Suppose we are given a cross power spectrum defined by
( )
,
0 , elsewhere
XY
jb
a W W
S W
e
e
e

+ < <

where 0 e > , a & b are real constants. Find the cross


correlation
Solution:
( ) ( )
1
2
j
XY XY
R S e d
et
t e e
t

=
}
1
2
W
j
W
jb
a e d
W
et
e
e
t

| |
= +
|
\ .
}
2 2
2 2
W W
j j j
W W
a e jb e e
j W j j
et et et
e
t t t t t

( (
= +
( (

2 2
2 2
jW jW jW jW jW jW
a e e jb We e We e
j W j j
t t t t t t
t t t t t t t

( (
= + + +
( (

( )
2
2 2
jW jW jW jW
asin W bW e e jb e e
W W
t t t t
t
tt t t t t

( (
= + +
( (

( ) ( ) ( )
2
asin W b cos W bsin W
W
t t t
tt tt t t
= +
( ) ( )
2
1 b
a sin W b cos W
W
t t t t
tt
( | |
= +
| (
\ .
Problem 29. a). The cross-power spectrum of real Random process ( ) { }
X t and ( ) { }
Y t
is given by ( )
; 1
0 ; elsewhere
XY
a jb
S
e e
e
+ <

. Find the cross correlation function.


Solution:
( ) ( )
1
2
j
XY XY
R S e d
et
t e e
t

=
}
( )
1
1
1
2
j
a jb e d
et
e e
t

= +
}
1 1
2 2
1 1
2 2
j j j
a e b e e
j j j
et et et
e
t t t t t

( (
= +
( (

2 2
2 2
j j j j j j
a e e jb e e e e
j j j
t t t t t t
t t t t t t t

( (
= + + +
( (

| |
2
2 2
j j j j
a jb e e jb e e
sin
j
t t t t
t
tt t t t t

( (
= + +
( (

2
asin bcos bsin t t t
tt tt tt
= +
Unit.4. Correlation and spectral densities
27
( )
2
1
a b sin b cos t t t t
tt
= + (

b). Find the power spectral density function of a WSS process with autocorrelation
function. ( )
2
R e
ot
t

= .
Solution:
( ) ( )
i
XX XX
S R e d
et
e t t

=
}
2
a i
e e d
t et
t

=
}
( )
2
a i
e d
t et
t

=
}
2
i
a
a
e d
et
t
t
| |
+
|
\ .

=
}
2
2. .
i
a
a
e d
e
t t
t
| |
+
|
\ .

=
}
2 2 2 2
2
2 2 2 2
2
2 2 2
i i i
a t
a a a
e d
e e e
t
t
( | |

+ + ( |
|
( \ .

=
}
2 2
2
2 4
i
a
a a
e d
e e
t
t
(
| |
+ + (
|
\ . (

=
}
2
2
2 2
4
.
i
a
a
a
e e d
e
e
t
t
(
| |
+ (
|
\ . (

=
}
( )
2
2
2
4
i
a
a
a
XX
S e e d
e
e
t
e t
| |
+
|
\ .

=
}
Now let,
2
2
. .
2 2
i i
a u i e u a
a a
e e
t t
| | | |
+ = = +
| |
\ . \ .
du a dt =
and as, , u t and as , u t
( )
2
2
4
u
a
XX
du
S e e
a
e
e

=
}
2
2
4
0
.2
a
u
e
e du
a
e

=
}
2
....
u
e

is even
2
4
2
2
a
e
a
e
t

= ...Standard Result :
2
0
2
X
x
e dx

=
}
Unit.4. Correlation and spectral densities
28
( )
2
4a
XX
S e
a
e
t
e

=
Problem 30. a). The autocorrelation function of the random telegraph signal process is
given by ( )
2 2
R a e
t
t

= . Determine the power density spectrum of the random telegraph
signal.
Solution:
( ) ( )
2 2 i i
XX XX
S R e d a e e d
t et et
e t t t



= =
} }
0
2 2 2 2
0
i i
a e e d a e e d
t t et et
t t

= +
} }
( ) ( )
0
2 2 2 2
0
.
i i
a e e d a e e d
t t et et
t t

= +
} }
( ) ( )
0
2 2 2 2
0
i i
a e d a e d
e t e t
t t

= +
} }
( )
( )
( )
( )
0
2 2
2 2
0
2 2
i i
e e
a a
i i
e t e t
e e

( (
= +
( (


( )
( )
( )
( )
2 2
0 0
2 2
a a
e e e e
i i e e

=
+
( )
( )
( )
2 2
1 0 0 1
2 2
a a
i i e e
=
+
2
1 1
2 2
a
i i e e
(
= +
(
+
( )
2
2
2
2 2
2
i i
a
e e
e
(
+ +
= (
+
(

i.e., ( )
2
2
2 2 2 2
4 4
4 4
XX
a
S a

e
e e
(
= =
(
+ +

b). Find the power spectral density of the random process ( ) { }
x t if ( ) { }
1 E x t =
and ( ) 1
xx
R e
o t
t

= + .
Solution:
( ) ( )
i
XX
S R e d
et
e t t

=
}
1
i
e e d
o t et
t

(
= +

}
0
0
i i i
e d e e d e e d
et ot et ot et
t t t



= + +
} } }
( )
1 1
i i
o e
o e o e
= + +
+
( )
2 2
2o
o e
o e
= +
+
*******