OP01 Random Variables
-
Upload
nguyen-ngoc-thai -
Category
Documents
-
view
215 -
download
0
Transcript of OP01 Random Variables
-
8/12/2019 OP01 Random Variables
1/57
Lectured by Ha Hoang Kha, Ph.D.Ho Chi Minh City University of Technology
Email: [email protected]
Random variables
-
8/12/2019 OP01 Random Variables
2/57
Content
Random variables Probability Histogram or probability density function Cumulative function Mean Variance Moments Some representations of random variables
Bi-dimensional random variables Marginal distributions
Independence Correlations Gaussian expression of multiple random variables
Changing random variables
2
-
8/12/2019 OP01 Random Variables
3/57
Examples : Acoustic waves Music,speech,
...
Light waves Light source(star, )
...
Electric current given bya microphone
Current given bya spectrometer
Number series Physical measurements
Photography ...
Introduction
s igna l = every entity which contains some physicali n fo rma t ion
3
-
8/12/2019 OP01 Random Variables
4/57
SpeechBiomedicalSound and MusicVideo and ImageRadar
4
Typical Signals
-
8/12/2019 OP01 Random Variables
5/57
Signa l p roc ess in g = procedure used to:
extract the information (filtering, detection,estimation, spectral analysis...)
Adapt the signal (modulation, sampling.) (to transmit it or save it)
pattern recognition
5
Signal Processing
-
8/12/2019 OP01 Random Variables
6/57
Dimension al c lass i f ica t ion : Number of free variables.
Examples :
Electrical potential V(t) = Unidimensional signal
Statistic image black and white brightness B(x,y) = bi-dimensional signal Black and white film B(x,y,t) = tri-dimensional signal...
Phenom enolog ical Class i f ica t ion Random or deterministic evolution
Deterministic signal : temporal evolution can be predicted or modeled by an
appropriate mathematical mode
Random signal : the signal cannot be predicted statistical description
The signal theory is independent on the physic phenomenon and the types ofvariables.
Every signal has a random component ( external perturbation, )
6
Classification of Signals
-
8/12/2019 OP01 Random Variables
7/57
Morph olog ica l c lass i f icat ion :
7
-
8/12/2019 OP01 Random Variables
8/57
Probability
8
-
8/12/2019 OP01 Random Variables
9/57
Probability
If two events A and B occurs,
P(B/A) is the conditional probability
If A and B are independent, P(A,B)=P(A).P(B)
9
, lim
lim lim lim /
AB
N
df AB A AB A
N N N A A
n P A B
N
n n n n P B A P A
n N n N
, / / P A B P B A P A P A B P B
-
8/12/2019 OP01 Random Variables
10/57
Let us consider the random process : measure the
temperature in a room
Many measurements can be taken simultaneously usingdifferent sensors (same sensors, same environments) andgive different signals
t z 1
z2
z 3
t 1 t 2
z , t x
Signals obtained when
measuring temperature
using many sensors
10
Random variable and random process
-
8/12/2019 OP01 Random Variables
11/57
The random process is represented as a functionEach signal x(t), for each sensor, is a random signal.
At an instant t, all values at this time define a random variable
z ,
t x
11
Random variable and random process
t z 1
z 2
z 3
t 1 t 2
z ,
t x
Signals obtained whenmeasuring temperature
using many sensors
-
8/12/2019 OP01 Random Variables
12/57
N(m, t i ) = number of events: " x i = x + D x"
x
m D x (m+ 1) D x
N(m)
Precision of measurment
1
,Prob m 1 lim
,Prob lim
mes
mes
ii N
mes
p
ik mi N
mes
N m t m x x x
N
N k t m x x p x
N
D D
D D
N mes = total number of measurments
Probability density function PDF)
The characteristics of a random process or a randomvariable can be interpreted from the histogram
12
-
8/12/2019 OP01 Random Variables
13/57
PDF properties
x=dx, the histogram becomes continuous. In thiscase we can write:
0Prob 1 x x
2
1
1 2Prob , x
i i x
x x x f x t dx mes
i
N i N t xn
t x f mes
,lim, where
13
-
8/12/2019 OP01 Random Variables
14/57
Histogram or PDF
Random signal Sine wave :
Uniform PDF 14
-1 1
f(x)
x
Uniform PDF
-
8/12/2019 OP01 Random Variables
15/57
Cumulative density function
15
-
8/12/2019 OP01 Random Variables
16/57
Examples
16
-
8/12/2019 OP01 Random Variables
17/57
,i i E g g x f x t dx
Statistical parameters :
Average value :
Mean quadratic value:
, x i i it E x x f x t dx
Variance : 22 22 x i i x i x i xt E x t m t
Standard deviation : 2 2
x i x xt m
Expectation, variance
Every function of a random variable is a random variable. If we know the
probability distribution of a RV, we can deduce the expectation value of
the function of a random variable:
17
2
2 2 ,i i i x
m t E x x f x t dx
-
8/12/2019 OP01 Random Variables
18/57
Exponential random variable
18
-
8/12/2019 OP01 Random Variables
19/57
-
8/12/2019 OP01 Random Variables
20/57
20
Examples
-
8/12/2019 OP01 Random Variables
21/57
21
Examples
-
8/12/2019 OP01 Random Variables
22/57
22
Examples
-
8/12/2019 OP01 Random Variables
23/57
Gaussian random variable
23
22
121
( ) 2
x m
f x e
2( ) ( ) E X m V X
( ; ) (0;1) X m If X N m N
-
8/12/2019 OP01 Random Variables
24/57
24
-
8/12/2019 OP01 Random Variables
25/57
25
Examples
-
8/12/2019 OP01 Random Variables
26/57
26
-
8/12/2019 OP01 Random Variables
27/57
27
-
8/12/2019 OP01 Random Variables
28/57
-
8/12/2019 OP01 Random Variables
29/57
Bi-dimensional random variable
Two random variablesX and Y have acommon probabilitydensity functions as :
(X,Y) f XY (x,y) is the probability densityfunction of the couple
(X,Y)
2 2( )( , ) x y XY f x y ce
29
-
8/12/2019 OP01 Random Variables
30/57
30
-
8/12/2019 OP01 Random Variables
31/57
Bi-dimensional Random variables
Cumulative functions:
Marginal cumulative distribution functions
Marginal probability density functions
dxdy y x f yY x X P y x F x y
Y X Y X
),(),(),( ,,
1),(, dxdy y x f Y X y x
y x F y x f Y X
),(),(
2
,
dxdy y x f y F y F
dxdy y x f x F x F
y
Y X XY Y
x
Y X XY X
),(),()(
),(),()(
,
,
duvu f v f
dvvu f u f
XY X
XY X
),()(
),()(
31
-
8/12/2019 OP01 Random Variables
32/57
32
-
8/12/2019 OP01 Random Variables
33/57
33
Examples
-
8/12/2019 OP01 Random Variables
34/57
Bi-dimensional Random variables
Moments of a random variable X
,
, ,
[ ] ( , )
[ ] ( , ) ( , ) ( )
[ ] ( )
[ ] [ ]. [ ]
n n n n X Y
X Y X Y X
y
E x y x y f x y dxdy
E X xf x y dxdy x f x y dydx xf x dx
E Y yf y dy
E XY E X E Y
, ( , ) ( ). ( ) X Y X Y f x y f x f y
If X and Y are independents and in this case
34
-
8/12/2019 OP01 Random Variables
35/57
-
8/12/2019 OP01 Random Variables
36/57
36
-
8/12/2019 OP01 Random Variables
37/57
37
-
8/12/2019 OP01 Random Variables
38/57
Covariance
38
-
8/12/2019 OP01 Random Variables
39/57
Covariance
39
-
8/12/2019 OP01 Random Variables
40/57
Correlation coefficient
40
-
8/12/2019 OP01 Random Variables
41/57
Correlation coefficient
41
-
8/12/2019 OP01 Random Variables
42/57
Correlation coefficient
42
-
8/12/2019 OP01 Random Variables
43/57
43
-
8/12/2019 OP01 Random Variables
44/57
44
-
8/12/2019 OP01 Random Variables
45/57
45
-
8/12/2019 OP01 Random Variables
46/57
46
-
8/12/2019 OP01 Random Variables
47/57
47
-
8/12/2019 OP01 Random Variables
48/57
48
-
8/12/2019 OP01 Random Variables
49/57
49
PDF of a transformed RV
-
8/12/2019 OP01 Random Variables
50/57
PDF of a transformed RV
Suppose X is a continuous RV with known PDF
Y=h(X) a function of the RV XWhat is the PDF of Y?
50
-
8/12/2019 OP01 Random Variables
51/57
51
-
8/12/2019 OP01 Random Variables
52/57
52
PDF of a transformed RV: exercises
-
8/12/2019 OP01 Random Variables
53/57
PDF of a transformed RV: exercises
X is a uniform random variable between -2 and 2. Write the expression on pdf of X Find the PDF of Y=5X+9
53
Sum of 2 RVs
-
8/12/2019 OP01 Random Variables
54/57
Sum of 2 RVs
54
Sum of 2 RVs
-
8/12/2019 OP01 Random Variables
55/57
Sum of 2 RVs
55
Sum of 2 RVs
-
8/12/2019 OP01 Random Variables
56/57
Sum of 2 RVs
56
Sum of 2 RVs
-
8/12/2019 OP01 Random Variables
57/57
Sum of 2 RVs