How do I linearly resample data?

I would like to linearly resample a matrix. For example if A = [0,0;5,10;10,20] I would like to resample it such that B = [0,0;1,2;2,4;3,6;4,8;5,10;6,12;7,14;8,16;9,18;10,20]

B =

     0     0
     1     2
     2     4
     3     6
     4     8
     5    10
     6    12
     7    14
     8    16
     9    18
    10    20

I originally thought that this is what the resample function did, but upon further analysis I was wrong. Please keep in mind that I am doing this for large sensor files, files which contain data for weeks at 5 minute timesteps.

Thank you for any help

Answer 1

[Chad Greene]

A = [0,0;5,10;10,20]; 
B(:,1) = 1:10; 
B(:,2) = interp1(A(:,1),A(:,2),B(:,1),'linear')
B =
          1.00          2.00
          2.00          4.00
          3.00          6.00
          4.00          8.00
          5.00         10.00
          6.00         12.00
          7.00         14.00
          8.00         16.00
          9.00         18.00
         10.00         20.00

[Justin Berquist]

That was just an example, I need to do it for a scenario where A is a 2xn matrix and the values within are not known. How would I go about that?

Thank you

[Chad Greene]

Which values are unknown?

Answer 2

[Star Strider]

You can certainly use the resample function. You simply have to create a timeseries object from your data.

See my Answer to your previous Question in: http://www.mathworks.com/matlabcentral/answers/282548-how-can-i-resample-a-set-of-data

[Justin Berquist]

A = xlsread('Canal5208RadiatorT_out.csv','','A:B'); C = length(A); A = resample(A,A(:,1):0.2:A(:,C));

What is wrong with this then? I just want my data to be linearly interpolated, adding 4 points inbetween every 2 original points.

[Justin Berquist]

When i simply do A = resample(A,5,1) my data is not linearly aligned

[Star Strider]

I would have to see your ‘Canal5208RadiatorT_out.csv’ file. If your data are regularly sampled (fixed sampling interval), linearly interpolating it is straightforward, using the linspace function, for instance, to generate the ‘query’ vector. If it’s not regularly-sampled, this becomes more difficult, because four points in every individual interval have to be created.