Scribd
Upload
146 views

Statistic Matlab Example

This document contains a report from a student named Adang Prianto (ID: 26520338) for their ECE5883 Advanced Signal Processing tutorial and lab for week 2. The report discusses various topics related to random variables and probability distributions covered in the tutorial, including definitions of mean, covariance, correlation, Bernoulli random variables, probability distribution functions, expectation, Gaussian distributions, and transformations between different distributions. It also includes the code and results from exercises completed during the lab portion of the class related to these concepts.

Uploaded by

Adang Prianto
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
146 views

Statistic Matlab Example

This document contains a report from a student named Adang Prianto (ID: 26520338) for their ECE5883 Advanced Signal Processing tutorial and lab for week 2. The report discusses various topics related to random variables and probability distributions covered in the tutorial, including definitions of mean, covariance, correlation, Bernoulli random variables, probability distribution functions, expectation, Gaussian distributions, and transformations between different distributions. It also includes the code and results from exercises completed during the lab portion of the class related to these concepts.

Uploaded by

Adang Prianto
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 7

REPORT

SUBJECT
ACTIVITY
STUDENT

: ECE5883 (ADVANCE SIGNAL PROCESSING)


: TUTORIAL & LABORATORY WEEK 2
NAM
E
ID

: ADANG PRIANTO
: 26520338

Tutorial Question
1. X & y is Random Variable
Mean E[X] = mx
E[Y] = my
Covariance = E[(X-mx) (Y-my)
= E[X Y*] mx my

Covariance E [ X Y ] m x m y
CrossCorrelation=
=
Std Dev x , y
x y
2. Bernoulli Random Variable :Trial variable that give exact 2 possible outcome (success
or not success).
3. Probability Distribution Function, is a statistical function that describes all the
possible values and likelihoods that a random variable can take within a given range.
Probability density function (PDF), or density of a continuous random variable, is
a function that describes the relative likelihood for this random variable to take on a
given value.
4. Expectation random variable x = ak
Given P(x = ak)
E[ X ]= a k P ( x=a k )
k

Given fx()

E [ X ] = f x ( ) d

5. Importance of:
Mean Squared value: it measure average power of random variable.
Standard Deviation: it makes a difference whether the distribution is spread out over
a broad range or bunched up closely around the mean.
1

Adang Prianto | 26520338

6. Joint Distribution function of variable x & y


F( x , y)=P( X x , Y y )
Joint density function
p X , Y ( x , y)=P( X =x , Y = y )

7. Two random variable to be statistically independent


p X , Y ( x , y )= p X ( x ) p y ( y )
Two Random variable X Y said to be orthogonal if the vectors are at perpendicular
to each other.
A set of two or more random variables is called uncorrelated if each pair of
them have no linear relationship. Covariance E(XY) - E(X)E(Y), is zero.

8. Gaussian probability density function


f ( x )=

1( x )
1
exp
2
2

9. Mean and variance of Gaussian distribution function


Mean

Variance

= 2

___________________________________________________________________________
Exercise 1
Step 1 to 6
Syntax

Function Syntax

X=1.*randn(1,1000)+0;
%% Generate R.V. with variance=1, mean=0
plot(X);
title('Plot of Random Variable');
xlabel('x');
ylabel('y');
[Mean_Data,Variance_Data]=mean_var(X);

%% Function to calculate mean and


variance
function [m,v]=mean_var(x)
m = sum(x)/length(x)
v = sum((x-m).^2)/(length(x)-1)
end

Adang Prianto | 26520338

Step 7
True Mean

=0

True variance = 1

Mean of generated R.V.

= 0.0383

Variance of generated R.V.

= 1.0110

Compared to true value of mean and variance, generated value have an error or difference
0.0383 and 0.0110 for mean and variance respectively.
Step 8
Increase sample e.g. up to 100000 sample
m

= -0.0012

0.9956

The difference is much smaller, calculated value of mean and variance is closer to the true
value. Larger sample will give more accurate estimation.
___________________________________________________________________________
Exercise 2
1. Random vector Y with mean = 2 and variance = 4
E[Y] = a E[X] + b
2 = a . (0) + b
b =2
Var (Y)= a2 Var (X) + 0
3

Adang Prianto | 26520338

4 = a2 (1)
a =2
2. Matlab calculation of mean and variance
m
= 2.0188
v

= 3.9627

Sample mean of Y is higher compared to sample mean of X, shown in figure above.


Random variable of Y tend to clustered at centre value 2, while Y at 0. The variance
has similar behaviour as mean, data of random variable Y is scattered from -5 to
-9, higher than X which is scattered within interval -3 to -3.
_____________________________________________________________________
Exercise 3
Syntax
function CDFVec=mycdffunc(X,x) %% Function
N=length(X);
for x_index=1:length(x)
CDFVec(x_index)=(1/N)*sum(X<=x(x_index));
end
end

CDFVec=mycdffunc(X,x);
plot(CDFVec)
hold on
title('Cumulative Distribution Function')
X=randn(1,20); %% Sample change for N=20,200,2000
CDFVec=mycdffunc(X,x);
4

Adang Prianto | 26520338

plot(CDFVec)
hold on
legend('N=20','N=200','N=2000')

___________________________________________________________________________
Exercise 4
Syntax
Beta=3;
X=rand(1,1000);
Y=(-1/Beta)*log(X);
[Ng,n]=hist(Y,50);
Bin_width=n(3)-n(2);
Pvec=Ng/Bin_width/50;
Tpdf=3.*exp(-3*n);
plot(Pvec,n);
hold on
plot(Tpdf);
title('Transform Uniform to Exponential')
legend('Calculated Py(y)','True Value Py(y)')

Bin
1
2
3
4
5
5

Total
23
20
20
14
22

Bin
11
12
13
14
15

Total
24
18
17
23
28

Bin
21
22
23
24
25

Total
0
1
1
1
0

Bin
31
32
33
34
35

Total
1
0
0
0
1

Bin
41
42
43
44
45

Total
0
0
0
0
0

Adang Prianto | 26520338

6
7
8
9
10

25
20
15
21
18

16
17
18
19
20

17
23
27
26
26

26
27
28
29
30

1
0
0
1
1

36
37
38
39
40

0
0
0
0
0

46
47
48
49
50

0
0
0
0
1

Bin_width = 0.0199

_____________________________________________________________________

Exercise 5
sigma_R = 3;
N = 10000;
U = rand(1,N);
R = sqrt(-2*sigma_R*log(U))
[Ng,n] = hist(R,20);
size(Ng)
p_hist=Ng/10000/(n(3)-n(2));
size(p_hist)
plot(n,p_hist);
Ray_pdf = (n/sigma_R).*exp(-n.^2/2/sigma_R);
hold on

Adang Prianto | 26520338

plot(n,Ray_pdf,'r')
title('Uniform to Rayleigh Transform')
legend('Calculated PDF','True PDF')

Adang Prianto | 26520338

You might also like