Digital Signal Processing Reference
In-Depth Information
Comparing (
5.23
) and (
5.1
), it follows that the random variable
X
from (
5.23
)is
a lognormal variable which itself represents the product of a number of independent
variables, as can be seen in (
5.19
).
As a result, as the sum of a large number of independent random variables
approaches a normal random variable (central limit theorem)
the product of a large
number of independent random variables gives rise to a lognormal variable
.
The lognormal PDF can be used to describe the life distribution of many
semiconductor components, the failure of which is a result of cracks caused by
material fatigue [KLA89, p. 51].
5.2 Rayleigh Random Variable
5.2.1 Density Function
The
Rayleigh
density function is defined as:
f
X
ðxÞ¼kx
k
x
2
;
for
x
0
2
(5.24)
f
X
ðxÞ¼
0
;
for
x <
0
where
k
is a positive constant.
The density function is also expressed using the parameter
s
2
, which is related
with the constant
k
as follows:
s
2
¼
1
=k;
(5.25)
resulting in:
1
s
2
x
x
2
f
X
ðxÞ¼
;
for
x
0
2
s
2
(5.26)
f
X
ðxÞ¼
0
;
for
x <
0
The Rayleigh random variable and the corresponding density for the parameter
s
2
¼
4 are shown in Fig.
5.3
.
The Rayleigh density is useful in describing the envelope of a narrowband normal
noise. It is also important in the analysis of measurement errors [PEE93, p. 56].
Note that the random variable has only positive values and that the PDF is
asymmetrical and has a maximum value for
p
k
x ¼ s ¼
1
=
:
(5.27)
r
k
e
¼ f
X
ðsÞ¼
1
1
s
p ¼
f
X
p
k
:
(5.28)
Search WWH ::
Custom Search