CODEWITHSUNDEEP
pythonmatlabfft
Amanda Willis
Amanda Willis

Reputation: 1

different calculation output from Matlab and Python for Inverse Fourier Transform

I have Matlab code and I want to translate it into Python. here I have an array

python array

import numpy as np  
a =  [ 81.42663125 -5.6074337j,  -26.39236508 -2.20896985j, 0.+0.j,  0.+0.j, 0.+0.j, 81.42663125-5.6074337j, -26.39236508-2.20896985j, 0.+0.j, 0.+0.j, 0.+0.j, -41.21717133+13.27058111j, 0.+0.j, 0.+0.j, 0.+0.j, 0.+0.j,0.+0.j]
b = np.asarray(a)
inverse = np.fft.ifft(b)

Matlab array (same as python)

a =   [ 81.42663125 -5.6074337i;  -26.39236508; -2.20896985i; 0.+0.i;  0.+0.i; 0.+0.i; 81.42663125-5.6074337i; -26.39236508-2.20896985i; 0.+0.i; 0.+0.i; 0.+0.i; -41.21717133+13.27058111i; 0.+0.i; 0.+0.i; 0.+0.i; 0.+0.i; 0.+0.i];
inverse = ifft (x);

I tried to inverse it with numpy.fft.ifft it turns like this:

     [4.30321006-0.14763913j   5.66633932+3.89294213j  -0.79347125-5.89179416j
      6.84928702-3.05455565j   9.8032857 +2.39782633j  -2.65998456-3.6116546j
      10.67141555-4.33907798j  11.24112597+2.56947076j -2.57607321+0.82941132j
      11.85609673-4.26124876j  9.03685629+3.33776417j  -1.27864139+4.64168411j
      8.82623526-4.48145694j   5.49420633+2.5781028j   1.44185722+5.49124955j
      3.54488621-5.55845764j]

while the result from matlab looks like this

    [4.30321006250000   + 0.147639125000000i
     3.54488621443892   + 5.55845764203018i
     1.44185722459799   - 5.49124955185143i
     5.49420632611877   - 2.57810280349092i
     8.82623525708333   + 4.48145694500000i
     -1.27864138655093  - 4.64168410598597i
     9.03685628560142   - 3.33776416665201i
     11.8560967270980   + 4.26124875584204i
     -2.57607320833333  - 0.829411319166666i
     11.2411259692567   - 2.56947075987604i
     10.6714155516520   + 4.33907798310142i
     -2.65998456069909  + 3.61165460362471i
     9.80328570125000   - 2.39782632583333i
     6.84928701535530   + 3.05455564883182i
     -0.793471249351424 + 5.89179416040201i
     5.66633931998235   - 3.89294213097583i]

I tried with another matrix, it turns out different. I tried transpose the python array first but nothing good came. how can I obtain result same as matlab?

what I've changed:

  1. if you point out the output from the python has the same value as Matlab but disordered.
  2. Title "different calculation output from Matlab and Python for Inverse Fourier Transform"

Upvotes: 0

Views: 171

Answers (1)

eWizard
eWizard

Reputation: 11

I just checked the output for the ifft of your MATLAB array. I'm sure your first and second element are supposed to be [ 81.42663125-5.6074337i; -26.39236508-2.20896985i;. The spacing in between the real and imaginary part of these equations and the semicolon in your second element are causing you these issues. edit:

My python results:

My python results

This is what I have for my MATLAB code:

MATLAB Results

Unless you are talking about how MATLAB is outputting the array vs python, I believe you just have minor errors in the MATLAB. Python is going to be outputting is as a one-dimensional array

Upvotes: 1

Related Questions