RESAMPLING DATA

Hi,

I am new here. I have Excell file with one column. In this column there are about 10 000 process data (let say temperatures). I need to export every 15th data to a new excell (every 15th row from first to the last one). Can someone help me?

Thanks in advance, Ivan

[Jan]

Please explain the relation to Matlab. It looks liek an Excel problem.

[Daniel Shub]

Is there a reason you are not doing this directly in Excel? What have you tried so far in MATLAB?

[Ivan Mohler]

I would like to generate this problem in Matlab, because I have a lots of problem when I have in one column more then 10 000 data... I am doing some modelling in matlab, so I wanna solve this kind of problem in it. If someone has got global answer for this kind of problems, please help me....

Answer 1

[Kye Taylor]

One way:

numData = xlsread('yourExcelFile.xlsx'); % read in data
numDataOut = numData(1:15:end); % subsample (get every 15th row)
xlswrite('yourNewExcelFile.xlsx', numDataOut); % write out data

[Ivan Mohler]

Kye Taylor --> I will try your solution and report you... Tell me is the "end" my last row in my column (real number) or? Thanks...here you soon...

[Ivan Mohler]

Kye, it works....Thanks a lot.

Can you tell me how can I do this but with rows?

Do you into (familiar) filters...The thing is I need some smoothing filter for my data....

If you are into I can explane to you much more deeper...

[Kye Taylor]

You probably realized this, but "end" inside indexing parentheses returns the largest valid index at that location.

To subsample rows, replace the second command with

numDataOut = numData(1:15:end,:);

Notice numData(1:15:end,:) referes to rows 1, 16,31,... and all the columns associated with these rows (that's the ,:). (That is, you will write out the entire first, sixteenth, thirtyfirst rows etc.)

You'll probably have more luck asking a specific question about the type of filtering you're into on the main answers page. In the meantime, you might find the following cool:

t = linspace(0,1,1024);

x = sin(2*pi*t);

noisyx = x + .1*randn(size(x));

% smooth it using local averaging

windowWidth = 20;

smoothedx = conv(noisyx, 1/windowWidth*ones(1,windowWidth),'same');

figure,plot(t,noisyx,'r',t,smoothedx,'b')

legend('original signal', 'smoothed signal');

Answer 2

[Ivan Mohler]

I would like to generate this problem in Matlab, because I have a lots of problem when I have in one column more then 10 000 data... I am doing some modelling in matlab, so I wanna solve this kind of problem in it. If someone has got global answer for this kind of problems, please help me.... Kye Taylor --> I will try your solution and report you... Tell me is the "end" my last row in my column (real number) or? Thanks...here you soon...