SPECTRAL

REPRESENTATION
By- Saurabh Shukla
R.No-ICE2012005
M.Tech
TOPICS COVERED SO FAR
1. Factorization and Innovations
2. Finite Order System and Variables

1. Series representation of Stochastic Process
2. Spectral representation of Random Process
TOPICS TO BE COVERED
1. SERIES REPRESENTATION OF
STOCHASTIC PROCESS
a) Fourier series

b) Mean Square Periodicity

c) Representation of ACF in Fourier series

d) Fourier series representation of periodic stochastic
process

e) Karhunen-Loeve expansion

2. FOURIER SERIES
a) What is Fourier Series?

b) How it decomposes the periodic signals?

c) What is Periodicity?

d) What is Orthogonality relationship?

e) What are the existence condition of Fourier
Series?
3. MEAN SQAURE PERIODICITY

a) Let we have a stochastic process X(t) in

b) can this continuous information be represented in terms of a
countable set of random variables whose relative importance
decrease under some arrangement?

c) To get to know this it is best to start with Mean-Square periodic
process.

d) A stochastic process X(t) is said to be mean square (M.S) periodic,
if for some T > 0

i.e

e) Proof: suppose X(t) is M.S. periodic. Then

, 0 T t s s
. all for 0 ] ) ( ) ( [
2
t t X T t X E = ÷ +
( ) ( ) with 1 for all . X t X t T probability t = +
--(1)

But from Schwarz’ inequality

Thus the left side equals

or

i.e.,

i.e., X(t) is mean square periodic.
. period with periodic is ) ( T R t
2
2 2
*
1 2 2 1 2 2
0
[ ( ){ ( ) ( )} ] [ ( ) ] [ ( ) ( ) ] E X t X t T X t E X t E X t T X t + ÷ s + ÷
* *
1 2 1 2 2 1 2 1
[ ( ) ( )] [ ( ) ( )] ( ) ( ) E X t X t T E X t X t R t t T R t t + = ¬ ÷ + = ÷
Then periodic. is ) ( Suppose ) ( t R :
0 ) ( ) ( ) 0 ( 2 ] | ) ( ) ( [|
* 2
= ÷ ÷ = ÷ + t t t R R R t X t X E
( ) ( ) for any R T R t t t ¬ + =
. 0 ] ) ( ) ( [
2
= ÷ + t X T t X E --(2)
--(3)
*
1 2 2
[ ( ){ ( ) ( )} ] 0 E X t X t T X t + ÷ =
4. REPRESENTATION OF ACF IN FOURIER
SERIES
Thus if X(t) is mean square periodic, then is periodic and let

represent its Fourier series expansion. Here

In a similar manner define

Notice that are random variables, and

0
0
1
( ) .
T
jn
n
R e d
T
e t
¸ t t
÷
=
}
0
0
1
( )
T
jk t
k
c X t e dt
T
e
=
}
+· ÷ ÷· = k c
k

,
--(5)
--(6)
) (t R
0
0
2
( ) ,
jn
n
R e
T
e t
t
t ¸ e

÷·
= =
¿
--(4)
0 1 0 2
0 1 0 2
0 2 1 0 1

* *
1 1 2 2
2
0 0

2 1 1 2
2
0 0

( ) ( )
2 1 2 1 1
0 0
1
[ ] [ ( ) ( ) ]
1
( )
1 1
[ ( ) ( )]
m
T T
jk t jm t
k m
T T
jk t jm t
T T
jm t t j m k t
E c c E X t e dt X t e dt
T
R t t e e dt dt
T
R t t e d t t e dt
T T
t
e e
e e
e e
t t
¸
÷
÷
÷ ÷ ÷ ÷
=
= ÷
= ÷ ÷
} }
} }
} }

i.e., form a sequence of uncorrelated random variables.

0 1
,

1 ( ) *
1
0
0,
[ ] { }
0 .
m k
T
m j m k t
k m m
T
k m
E c c e dt
k m
e
o
¸
¸
÷ ÷
> =
¦
= =
´
=
¹
}
--(7)
+· =
÷· =
n
n n
c } {
5. REPRESENTATION OF M.S. PERIODIC
PROCESS IN FOURIER SERIES

Now on considering the partial sum-

Theorem- We shall show that in the mean square
sense as
i.e.,

Proof:

0
( ) .
N
jk t
N k
K N
X t c e
e ÷

=
¿
) ( ) (
~
t X t X
N
=
--(8)
. · ÷ N
. as 0 ] ) (
~
) ( [
2
· ÷ ÷ ÷ N t X t X E
N
--(9)
2
2
*
2
[ ( ) ( ) ] [ ( ) ] 2Re[ ( ( ) ( )]
[ ( ) ].
N N
N
E X t X t E X t E X t X t
E X t
÷ = ÷
+
--(10)
0
0
0
2
* *

( ) *
0

( )
0
[ ( ) ] (0) ,
[ ( ) ( )] [ ( )]
1
[ ( ) ( ) ]
1
[ ( ) ( )] .
k
k
k
N
jk t
N k
k N
N
T
jk t
k N
N N
T
jk t
k
k N k N
E X t R
E X t X t E c e X t
E X e X t d
T
R t e d t
T
e
e o
e o
¸
¸
o o
o o ¸

=÷·
÷

÷ ÷

÷ ÷
=÷ =÷
= =
=
=
= ÷ ÷ =
¿
¿
¿
}
¿ ¿
}
--(12)
Similarly

i.e.,
0 0
2
( ) ( ) * *
2
[ ( ) ] [ [ ] .
[ ( ) ( ) ] 2( ) 0 as
N
j k m t j k m t
N k m k m k
k m k m k N
N
N k k
k k N
E X t E c c e E c c e
E X t X t N
e e
¸
¸ ¸
÷ ÷

=÷· =÷
= = =
¬ ÷ = ÷ ÷ ÷ ·
¿¿ ¿¿ ¿
¿ ¿ --(13)
0
( ) , .
jk t
k
k
X t c e t
e

÷
=÷·
÷·< < +·
¿
--(14)
and
But,
CONCLUSION DRAWN FROM
EQUATION -14
a) Thus mean square periodic processes can be represented in the form
of a series as in (14).
b) The stochastic information is contained in the random variables

c) Further these random variables are uncorrelated
and their variances

i.e

d) Thus if the power P of the stochastic process is finite, then the
positive sequence converges, and hence

e) This implies that the random variables in (14) are of relatively less
importance as and a finite approximation of the series in
(14) is indeed meaningful.

*
,
( { } )
k m k k m
E c c ¸ o =
0 as .
k
k ¸ ÷ ÷·
. , +· ÷ ÷· = k c
k
2
(0) [ ( ) ] .
k
k
R E X t P ¸

=÷·
= = = < ·
¿
k
k
¸

=÷·
¿
0 as .
k
k ¸ ÷ ÷·
, k ÷·
6-KARHUNEN-LOEVE EXPANSION
a) QUES-What about a general stochastic process, that is not
mean square periodic? Can it be represented in a similar
series fashion as in (14),

b) if not in the whole interval then take in finite
support of

c) Suppose that it is indeed possible to do so for any arbitrary
process X(t) in terms of a certain sequence of orthonormal
functions

, · < < · ÷ t
0 ? t T s s
i.e.,

where

and in the mean square sense

Further, as before, we would like the c
k
s to be uncorrelated random
variables. If that should be the case, then we must have

Now

¿
·
=
=
1
) ( ) (
~
n
k k
t c t X ¢
--(15)
--(16)
--(17)
( ) ( ) in 0 . X t X t t T s s
*
,
[ ] .
k m m k m
E c c ì o =
--(18)

* * *
1 1 1 2 2 2
0 0

* *
1 1 2 2 2 1
0 0

*
1 1 2 2 2 1
0 0
[ ] [ ( ) ( ) ( ) ( ) ]
( ) { ( ) ( )} ( )
( ){ ( , ) ( ) }
T T
k m k m
T T
k m
T T
k XX m
E c c E X t t dt X t t dt
t E X t X t t dt dt
t R t t t dt dt
¢ ¢
¢ ¢
¢ ¢
=
=
=
} }
} }
} }
--(19)

*
0

*
,
0
( ) ( )
( ) ( ) ,
T
k k
T
k n k n
c X t t dt
t t dt
¢
¢ ¢ o
=
=
}
}
A
and

Substituting (19) and (20) into (18), we get

Since (21) should be true for every we must have

or

*
1 1 2 2 2 1 1
0 0
( ){ ( , ) ( ) ( )} 0.
XX
T T
k m m m
t R t t t dt t dt ¢ ¢ ì ¢ ÷ =
} }

1 2 2 2 1
0
( , ) ( ) ( ) 0,
XX
T
m m m
R t t t dt t ¢ ì ¢ ÷ ÷
}
( ), 1 ,
k
t k ¢ = ÷·

*
, 1 1 1
0
( ) ( ) .
T
m k m m k m
t t dt ì o ì ¢ ¢ =
}
--(20)
--(21)

1 2 2 2 1 1
0
( , ) ( ) ( ), 0 , 1 .
XX
T
m m m
R t t t dt t t T m ¢ ì ¢ = < < = ÷·
}
--(22)
a) This is the desired uncorrelated condition in (18) which is
translated into the integral equation in (22) and it is known
as the Karhunen-Loeve or K-L. integral equation

b) The functions are not arbitrary and they must be
obtained by solving the integral equation in (22).They are
known as the eigenvectors of the autocorrelation function
of

c) Similarly the set represent the eigenvalues of the
autocorrelation function.

·
=1
)} ( {
k k
t ¢
1 2
( , ).
XX
R t t
1
{ }
k k
ì
·
=
END OF
“SERIES REPRESNTATION
OF
STOCHASTIC PROCESS”
THANK YOU
By- Saurabh Shukla
R.No-ICE2012005
M.Tech