interpolation of a structure array

Hello, I have a structure array which consists of 39 variables (LFHD, RFHD, etc.) Each of those variables contains 3 double arrays of size 1 x ?. I would like to interpolate those double arrays from 1 x ? to 1 x 101 (0-100%) so that they are all the same length and I can average across them.

I am a bit of a MATLAB novice. Any assistance would be greatly appreciated.

Answer 1

[Matt J]

fcn=@(f)interp1(f, linspace(1,numel(f),101)  );

for i=1:numel(TrajXLstance)
  
    TrajXLstance(i) = structfun(fcn, TrajXLstance(i),'uni',0);

end

[Josh Tome]

This is great! Thank you. Can I ask what the "f" represents?

[Josh Tome]

Sorry for the 2nd question before you answer the 1st, but I have another structure array similar to the OP, but instead of 1 x ? double arrays it has 3 x ? double arrays. When I try the same code, instead of interpolating the ? number, it interpolates the 3, so I end up with double arrays of size 101 x ?. Any thoughts?

[Matt J]

Can I ask what the "f" represents?

It represents the value of a struct field.

When I try the same code, instead of interpolating the ? number, it interpolates the 3

Transpose:

fcn=@(f)interp1(f.', linspace(1,numel(f),101)  ).';

[Josh Tome]

when I use the transpose on the 3 x ? double arrays at first glance it appeared to work by creating arrays of 3 x 101. However, looking closely at these arrays it compressed the actual data by half, and then filled the rest of the 101 points with NAN's. For instance a 3 x 66 array of data was converted into a 3 x 33 array of data with the last 3 x 68 array of NANs. Any idea why that would happen?

[Matt J]

Try this insead,

fcn=@(f)interp1(f.', linspace(1, size(f,2) ,101)  ).';

[Josh Tome]

Yes, that seems to work. Thanks again for your help!