Digital Signal Processing Reference
In-Depth Information
x ¼
2
2
e
ð
x
m
X
Þ
y
b
a
m
X
x ¼
ð
y
b
Þ
a
1
2
p
2
s
X
p
e
f
X
ð
x
Þ
d
y
d
x
1
2
p
s
X
2
s
X
p
f
Y
ðyÞ¼
¼
¼
jj
jjs
X
yb
a
2
e
ð
y
ð
b
þ
am
X
ÞÞ
1
2
p
2
a
2
s
X
p
¼
:
jjs
X
(4.71)
Comparing the result (
4.71
) with the expression for the normal PDF, we can
write:
2
e
ð
y
m
Y
Þ
1
2
p
2
s
Y
f
Y
ðyÞ¼
p
; 1<y<1;
(4.72)
s
Y
where
m
Y
¼ am
X
þ b;
s
Y
¼ðas
X
Þ
2
¼ a
2
s
X
:
(4.73)
Observing (
4.70
), we can easily conclude that the parameters (
4.73
) can be
m
Y
¼ EfYg¼EfaX þ bg¼aEfXgþb ¼ am
X
þ b;
s
aXþb
¼ a
2
s
X
:
(4.74)
The following rule stands:
The linear transformation of the normal random variable results in a normal
variable with the mean value and variance given in
(
4.73
).
Example 4.3.1
Find the PDF of the random variable
Y ¼
3
X
+ 2, where
X
is the
normal random variable with a mean value and variance equal to
1 and 4,
respectively.
Solution
The transformed random variable
Y
is also a normal random variable
with the parameters obtained from (
4.73
):
m
Y
¼
3
m
X
þ
2
¼
3
ð
1
Þþ
2
¼
5
;
s
2
2
s
2
3
Xþ
2
¼ð
3
Þ
X
¼
9
4
¼
36
:
(4.75)
From here, the PDF of the random variable
Y
is as follows:
2
2
p
6
e
ð
y
5
Þ
1
72
f
Y
ðyÞ¼
for
1<y<1:
(4.76)
Search WWH ::
Custom Search