I want to do the correlation of the following two plots

clc; clear all; Ts=1/200e6; t=Ts:Ts:0.00005; y=[ones(1,2000) zeros(1,8000)]; x=sin(1256e6*t+2198e10*t.^2); z=x.*y; subplot(2,1,1) plot(t,z) N=0; while N<4 t1=Ts+0.00002+0.00005*N:Ts:0.00002+0.00005*(N+1); j=@(t)x.*y; w=@(t)j(t-0.00002); subplot(2,1,2) plot(t1,w(t)) hold on N=N+1; end

[Star Strider]

This line makes no sense and throws an error:

t1=Ts+0.00002+0.00005*N:Ts:0.00002+0.00 005*(N+1); 

Please learn how to format your code. Hightlght the code, then use the left-most button in the CODE tab in the toolstrip at the top to format it so it is readable.

[Devesh Yadav]

There was a typing mistake from my side. Please check it once more

Answer 1

[Star Strider]

Going back to using my code (because it is easier to work with):

Ts=1/200e6; 
t=Ts:Ts:0.00005; 
y=[ones(1,2000) zeros(1,8000)]; 
x=sin(1256e6*t + 2198e10*t.^2); 
z=x.*y; 
NR = 4;                                                 % Number Of Repeats Desired
z_extended = reshape(repmat(z(:), NR, 1), [], 1).';     % Duplicate & Convert To Vector
t_extended = linspace(Ts, max(t)*NR, numel(t)*NR);      % Time Vector
figure
plot(t,z)
figure
plot(t_extended, z_extended)                            % Plot Original
xlim([0 max(t_extended)*1E-1])

and adding the cross-correlation:

[xc,lags] = xcorr(z, z_extended);

figure
plot(lags, xc)
grid

produces:

[Devesh Yadav]

I want correlation of z and z_extended(delayed waveform).

[Devesh Yadav]

And also it would be so kind of you,if you could help me find correlation(of z and z_extended_delayed_waveform(which is w in my code)) using my code.It would surely mean a lot to me.

[Star Strider]

That was not what your code produced. There was no initial delay.

Use this instead, then:

delay_z_extended = [zeros(1, fix(2E-5/Ts)) z_extended];     % With 20 µs Lag
[xc,lags] = xcorr(z, delay_z_extended);

figure
plot(delay_z_extended)                                      % Plot With 20 µs Lag
grid

figure
plot(lags, xc)
grid

The rest of the code is unchanged.

[Devesh Yadav]

Thank you

[Devesh Yadav]

Thank you.I got the desired result(correlation plot) what I wanted but the code you suggested me is a bit complex and quite difficult for me to digest as I am a beginner in MATLAB. If there is any easy way around of doing the same,please let me know. Once again thank you so much

[Star Strider]

As always, my pleasure!

My code uses the repmat function and then the reshape function to create and concatenate copies of the original waveform. It is quite useful in that respect, and so I use it frequently. The linspace function is a convenient way to create vectors of specific lengths. It then calculates the appropriate interval lengths. I encourage you to experiment with them with random vectors and matrices to see how they work and what they can do.

[Devesh Yadav]

delay_z_extended=[zeros(1,fix(2E-5/Ts)) z_extended]; What does this do?

[Star Strider]

It pre-pads ‘z_extended’ with 20 µs of zeros to produce the requested delay.

[Devesh Yadav]

Ok

[Star Strider]

Noted.

[Devesh Yadav]

Can you please tell me the no. of computations done in this correlation :- [xc,lags] = xcorr(z, z_extended);

[Star Strider]

I cannot.

I doubt that would be possible to determine.