why the plot doesn't work ,in the code i'm finding the estimation value of the phase using montecarlo simulation

%phase astimation using monte carlo
clear all 
close all
clc

%signal model x(n)=Acos(2pif0n+thetha)+w(n)
phase=0.75;
A=1;
sigma=0.1;
f0=0.035;
SNR=A^2/2*(sigma^2)

SNR = 0.0050

for m=1:length(SNR)
    
    
    M=5;
    for i=1:M;
        s=A*cos(2*pi*f0*i+phase);
        
        
        w=randn(size(s));
        x(i)=s+sigma*w
        
        
        
        y1=sum(x*sin(2*pi*f0*i))
        
        y2=sum(x*cos(2*pi*f0*i))
        
        phasehat(i)=-atan(y1+y2)
        %        
        %      mat(i,:)=phasehat
        
    end   
    
    %compute bias
    b(m)=(1/M)*sum(mean(phasehat(i))-phase)
    %mean square error
    mse=mean((phasehat(i)-phase)^2)
    var=1/M*SNR
    
end

x = 0.7006

y1 = 0.1528

y2 = 0.6837

phasehat = -0.6966

x = 1×2

0.7006 0.4581

y1 = 0.4933

y2 = 1.0484

phasehat = 1×2

-0.6966 -0.9954

x = 1×3

0.7006 0.4581 0.4058

y1 = 0.9588

y2 = 1.2361

phasehat = 1×3

-0.6966 -0.9954 -1.1433

x = 1×4

0.7006 0.4581 0.4058 0.0162

y1 = 1.2178

y2 = 1.0075

phasehat = 1×4

-0.6966 -0.9954 -1.1433 -1.1485

x = 1×5

0.7006 0.4581 0.4058 0.0162 -0.1706

y1 = 1.2563

y2 = 0.6401

phasehat = 1×5

-0.6966 -0.9954 -1.1433 -1.1485 -1.0855

b = -0.3671

mse = 3.3692

var = 1.0000e-03

figure(1)
plot(10*log(SNR),b)

figure(2)
plot(10*log(SNR),mse)

figure(3)
plot(10*log(SNR),10*log(var))

Answer 1

[Chris]

since length(SNR) is 1, you are running one loop iteration and generating one point. plot() doesn't show one point unless you specify a marker:

figure(1)
plot(10*log(SNR),b,'o')
figure(2)
plot(10*log(SNR),mse,'x')
figure(3)
plot(10*log(SNR),10*log(var),'^')

or equivalently, use scatter:

figure(1)
scatter(10*log(SNR),b)
% ...etc

[Cris LaPierre]

Moving my answer here since Chris and I said the same thing (and used the same markerstyles!)

SNR and b, mse, and var are all scalars, meaning single numbers. The default plot setting in MATLAB only include a linestyle, not a markerstyle. When you have a single point, there is no line, so that is why your plots appear to be empty.

Add a marker style to you linespec and the points will be visible.A=1;

A=1;
sigma=0.1;
SNR=A^2/2*(sigma^2)

SNR = 0.0050

b = -0.3671;
mse = 3.3692;
var = 1.0000e-03;

figure(1)
plot(10*log(SNR),b,'o')

figure(2)
plot(10*log(SNR),mse,'x')

figure(3)
plot(10*log(SNR),10*log(var),'^')

[Cris LaPierre]

I'd start by making SNR a vector of numbers rather than a single number.

I'd recommend going through MATLAB Onramp if you do not know what that means. I'd recommend at least the following chapters:

• Ch 4 - vectors and matrices
• Ch 5 - indexing into and modifying arrays
• Ch 6 - array calculations
• Ch 9 - plotting
• Ch 12 - programming (section 2 covers for loops)

[mhamad Yaacoub]

Thank you a lot! Now i understand why the answers in my program are already wrong sorry i’m begginer in matlab how i can change from scalar to vector?

[mhamad Yaacoub]

Thank you a lot, and sorry for disturbing