Error using + Matrix dimensions must agree.

by using white noise effect in the three level discrete F T:

clear all; close all; clc

fs=200;     %sampling freq.
dt =1/fs;
N0=fs/50;	%number of samples/cycle
m=3;        %no. of cycles

t = dt*(0:200); %data window
fi=50; %Frequency test

ww=wgn(201,1,-40);
size(transpose(ww))

v=@(t) sin(2*pi*fi*t + 0.3)+transpose(ww);
tmax=1;
% v      :  as function of time
% fs     : sampling frequency (Hz)
% tmax   : time of final estimation
% to test: [t,f]=ZC(@(t)(220*sin(2*pi*50.1*t+pi/2)),50*512,1)
% to test: [t,f]=ZC(@(t)(220*sin(2*pi*50.1*t+pi/2)+randn(1)*.1),50*512,1)


n=N0-1:-1:0;
f0=50;
f=50.88;
Hc=2/N0*cos(2*pi*n/N0+pi/N0);
Hs=-2/N0*sin(2*pi*n/N0+pi/N0);
t_est=[];
f_est=[];
j_max=tmax*fs;

for j=1:j_max+1
	x=v((200:j+N0-2).*dt);
    c(j)=x*Hc';
    s(j)=x*Hs';    
    if(j>N0)
        Ac(j-N0)=sqrt(sum(c(end-N0+1:end).^2)/N0);   
        As(j-N0)=sqrt(sum(s(end-N0+1:end).^2)/N0);    
        cc(j-N0)=c(end-N0+1:end)*Hc';
        ss(j-N0)=c(end-N0+1:end)*Hs';
        if(j>2*N0)
            Acc(j-2*N0)=sqrt(sum(cc(end-N0+1:end).^2)/N0);   
            Ass(j-2*N0)=sqrt(sum(ss(end-N0+1:end).^2)/N0);   
            ccc(j-2*N0)=cc(end-N0+1:end)*Hc';
            ccs(j-2*N0)=cc(end-N0+1:end)*Hs';
            ssc(j-2*N0)=ss(end-N0+1:end)*Hc';
            sss(j-2*N0)=ss(end-N0+1:end)*Hs';
            ff=f0*N0/pi*atan(tan(pi/N0)*((ccc(j-2*N0).^2+ccs(j-2*N0).^2)./(ssc(j-2*N0).^2+sss(j-2*N0).^2)).^.25);
            t_est=[t_est;(j-1)*dt];
            f_est=[f_est;ff];
        end
    end
end


t_est
f_est
plot(t_est,f_est,'red')
o=rms(fi)
c=rms(f_est)
RMSE = sqrt(mean(c - o).^2)
t_est;
f_est

plot(t_est, f_est,'red')
hold on

RMSE = sqrt(mean((f_est-fi).^2))

xlabel('time')
ylabel('frequency')
title('three LDFT white noise')

   plot (t_est,fi*ones(size(t_est)))
hold off

the error is :

Error using +

Matrix dimensions must agree.

Error in @(t)sin(2*pi*fi*t+0.3)+transpose(ww)

Error in threeDFT (line 33)

x=v((200:j+N0-2).*dt);

[Rik]

Please write your comments in the comments, not in flags. You can use links to the comment itself if you want an unambiguous reference.

[common fernando]

ok

Answer 1

[VBBV]

%t = dt*(0:200); %data window   COMMENT THIS LINE IN YOUR PROGRAM
fi=50; %Frequency test
ww=wgn(201,1,-40);
size(transpose(ww))
v=@(t) sin(2*pi*fi*t + 0.3)+transpose(ww); % using function handle

since you are using fuinction handle to define v in terms of t and calling inside the for loop with varying length of t

for j=1:j_max+1
	x=v((200:j+N0-2).*dt); % ....

you dont have to define data window at beginning, comment that line

[VBBV]

Declare variable t as symbolic variable befroe using it inside loop

syms t
v=@(t) sin(2*pi*fi*t + 0.3)+transpose(ww); % using function handle

[Walter Roberson]

There is no point in defining t as a symbolic variable. Any initilization you do to t is ignored when you create an anonymous function with named parameter t: inside the body of the anonymous function, any named parameter will be replaced by the value passed into the function handle.

[common fernando]

clear all; close all; clc

fs=200; %sampling freq.

dt =1/fs;

N0=fs/50; %number of samples/cycle

m=3; %no. of cycles

t = dt*(0:200); %data window

fi=50; %Frequency test

ww=wgn(201,1,-40);

size(transpose(ww))

x= sin(2*pi*fi*t + 0.3);

%v=@(t) sin(2*pi*fi*t + 0.3)+transpose(ww);

v = bsxfun( @plus, x , ww );

tmax=1;

% v : as function of time

% fs : sampling frequency (Hz)

% tmax : time of final estimation

% to test: [t,f]=ZC(@(t)(220*sin(2*pi*50.1*t+pi/2)),50*512,1)

% to test: [t,f]=ZC(@(t)(220*sin(2*pi*50.1*t+pi/2)+randn(1)*.1),50*512,1)

n=N0-1:-1:0;

f0=50;

f=50.88;

Hc=2/N0*cos(2*pi*n/N0+pi/N0);

Hs=-2/N0*sin(2*pi*n/N0+pi/N0);

t_est=[];

f_est=[];

j_max=tmax*fs;

for j=1:j_max+1

x=v((200:j+N0-2).*dt);

c(j)=x.*Hc';

s(j)=x*Hs';

if(j>N0)

Ac(j-N0)=sqrt(sum(c(end-N0+1:end).^2)/N0);

As(j-N0)=sqrt(sum(s(end-N0+1:end).^2)/N0);

cc(j-N0)=c(end-N0+1:end)*Hc';

ss(j-N0)=c(end-N0+1:end)*Hs';

if(j>2*N0)

Acc(j-2*N0)=sqrt(sum(cc(end-N0+1:end).^2)/N0);

Ass(j-2*N0)=sqrt(sum(ss(end-N0+1:end).^2)/N0);

ccc(j-2*N0)=cc(end-N0+1:end)*Hc';

ccs(j-2*N0)=cc(end-N0+1:end)*Hs';

ssc(j-2*N0)=ss(end-N0+1:end)*Hc';

sss(j-2*N0)=ss(end-N0+1:end)*Hs';

ff=f0*N0/pi*atan(tan(pi/N0)*((ccc(j-2*N0).^2+ccs(j-2*N0).^2)./(ssc(j-2*N0).^2+sss(j-2*N0).^2)).^.25);

t_est=[t_est;(j-1)*dt];

f_est=[f_est;ff];

end

end

end

t_est

f_est

plot(t_est,f_est,'red')

o=rms(fi);

c=rms(f_est)

RMSE = sqrt(mean(c - o).^2)

t_est;

f_est

plot(t_est, f_est,'red')

hold on

RMSE = sqrt(mean((f_est-fi).^2))

xlabel('time')

ylabel('frequency')

title('three LDFT white noise')

plot (t_est,fi*ones(size(t_est)))

hold off

Error

Error using *

Inner matrix dimensions must agree.

Error in threeDFT (line 39)

c(j)=x*Hc';