Digital Signal Processing Reference
In-Depth Information
This result can be generalized for an arbitrary sum of
N
independent random
variables:
X ¼
X
N
X
i
¼ Nðm
i
; s
i
Þ:
X
i
;
(4.108)
i¼
1
The variable
X
is also a normal variable
Nðm; s
2
Þ
with the parameters:
m ¼
X
¼
X
N
N
s
2
s
i
:
m
i
;
(4.109)
i¼
1
i¼
1
Example 4.5.2
Find the PDF of the random variable
X
,
X ¼ X
1
þ X
2
:
(4.110)
where
i ¼
1, 2 are normal random variables:
X
1
¼ N
(1, 5),
X
2
¼ N
(2.6, 8).
Solution
According to the observations in this section, the random variable
X
is
also a normal random variable with the parameters:
s
2
m ¼
1
þ
2
:
6
¼
3
:
6
;
¼
5
þ
8
¼
13
:
(4.111)
The corresponding signals and PDFs are shown in Fig.
4.17
.
4.5.3 Sum of Linear Transformations
of Independent Normal Random Variables
random variables
X
i
(
m
i
,
s
i
2
). The
Consider
independent
normal
linear
transformation
Y
i
¼ a
i
X
i
þ b
i
(4.112)
results in independent normal random variables
Y
i
. Their sum denoted as
Y
,
Y ¼
X
Y
i
¼
X
N
N
a
i
X
i
þ b
i
(4.113)
i¼
1
i¼
1
is also normal variable with parameters
m
Y
¼
X
Y
¼
X
N
N
s
2
a
2
i
s
2
a
i
m
i
þ b
i
;
i
:
(4.114)
i¼
1
i¼
1
Search WWH ::
Custom Search