TE3223
SISTEM KOMUNIKASI 2
REVIEW:
REVIEW
Transformasi
T
f
i Fourier
F
i &
Probabilitas Statistik
Program Studi S1 Teknik Telekomunikasi
Departemen
p
Teknik Elektro - Sekolah Tinggi
gg Teknologi
g Telkom
Bandung 2007
Transformasi Fourier
S( f ) =
j 2ft
s
(
t
).
e
dt
s (t ) =
S ( f ).e
j 2ft
df
S( f ) =
j 2ft
(
t
).
e
dt = 1
s(t)
AT
- T/2
+ T/2
- 1/T
|S(f)|
+1/T
harga modulus
AT
- 1/T 0
+1/T
(f)|
f
harga fasa
- 1/T 0
+1/T
A j 2f ct
j 2f c t
e
+e
Contoh : s(t) = A Cos 2fct =
2
maka
A
A
S ( f ) = ( f + fc ) + ( f fc )
2
2
1
X p( f )=
T0
m. f
X
m =
To
+
m
. f
T0
1
s( t ).dt j 2f . S ( f )
d
s( t ) j 2f . S ( f )
dt
Modul +an - Siskom 2 - Transformasi Fourier & Probstat
10
s ( t ).s (t )d S ( f ). S ( f )
1
s1 ( t ). s2 ( t )
S ( ). S ( f )d
1
11
Likelihood Principle
E
Experiment:
i
t
Pick
12
13
Bayes manipulations:
B
i l ti
P(Box A | Black) =
P(Box A and Black) /P(Black)
Decompose the numerator and denomenator in terms of the probabilities we
know.
P(Box A and Black) = P(Black | Box A)*P(Box A)
P(Black) = P(Black| Box A)*P(Box A) + P(Black| BoxB)*P(BoxB)
We know all these values (see prev page)! Plug in and crank.
P(Box A and Black) = 1/3 * 1/2
P(Black) = 1/3 * 1/2 + 2/3 * 1/2 = 1/2
P(Box A and Black) /P(Black) = 1/3 = 0.333
Notice that it matches our gut feeling that Box A is less likely,
likely once we have
seen black.
The information that the ball is black has CHANGED !
From P(Box A) = 0.5 to P(Box A | Black) = 0.333
Modul +an - Siskom 2 - Transformasi Fourier & Probstat
14
Likelihood Principle
Way of thinking
Hypotheses: Box A or Box B ?
Observation: Black
Prior probabilities: P(Box A) and P(Box B)
Likelihood of Black given choice of Box: {aka forward probability}
P(Black
P(Bl k | Box
B A) and
d P(Bl
P(Black
k | Box
B B)
Posterior Probability: of each hypothesis given evidence
P(Box A | Black) {aka inverse probability}
Likelihood Principle (informal): All inferences depend ONLY on
The likelihoods P(Black | Box A) and P(Black | Box B), and
The priors P(Box A) and P(Box B)
Result is a probability (or distribution) model over the space of
possible hypotheses.
Modul +an - Siskom 2 - Transformasi Fourier & Probstat
15
Recall:
P(Box A | Black) = P(Box A and Black) /P(Black) =
P(Black
(
| BoxA)*P(Box
) (
A)) / P(Black)
(
)
16
17
18
Note: this does not give the posterior probability P(Box A | Black),
but quickly helps us choose the best hypothesis (Box B) that would
explain the data
More examples: (biased coin etc)
p
p
g
_likelihood
http://en.wikipedia.org/wiki/Maximum
http://www.inference.phy.cam.ac.uk/mackay/itprnn/book.html
(chap 3)
19
y
1
pN ( y ) =
exp
2
2
2
pN(y)
0.8
N1(=0,1)
0.7
0.6
0.5
2 > 1
0.4
0.3
N2(=0,2)
0.2
0.1
0
-3
-2
-1
20
P
R
1
P=
2
1
R=
2
( y )2
exp
dy
2
( y )2
exp
dy
2
+ 2
21
G
Gaussian/Normal
i /N
l
0.4
0.35
N(=0,=1)
0.3
0.25
0.2
0.15
0.1
0 05
0.05
0
-4
-3
-2
-1
p(y>z)=
22
Normal
Distribution
Standardized Normal
Distribution
=1
= 0
One table!
Modul +an - Siskom 2 - Transformasi Fourier & Probstat
23
Example P(X 8)
X 85
Z=
=
= .30
10
N
Normal
l
Distribution
= 10
St d di d Normal
Standardized
N
l
Distribution
=1
.5000
.3821
3821
.1179
=5
=0
.30
30 Z
24
Tabel
Q function
Q-function
25