Finding Wavevector in 2D Domain

  • Context: Undergrad 
  • Thread starter Thread starter Crot
  • Start date Start date
  • Tags Tags
    2d
Click For Summary
SUMMARY

The discussion focuses on finding the wavevector in a 2D domain using the Fast Fourier Transform (FFT) method. The FFT provides the amplitudes for both positive and negative wave numbers (k_x, k_y), which are expressed as integer multiples of cycles per domain length. For an N-point FFT, the output order is typically from -N/2 to N/2, allowing users to identify the peak wavenumbers efficiently. This method is straightforward and effective for analyzing wave properties in a 2D space.

PREREQUISITES
  • Understanding of Fast Fourier Transform (FFT) algorithms
  • Familiarity with wave properties in a 2D domain
  • Knowledge of wavenumber representation and periodicity
  • Basic programming skills to implement FFT routines
NEXT STEPS
  • Research the implementation of FFT in Python using NumPy
  • Learn about wavenumber analysis in 2D Fourier transforms
  • Explore the differences between FFT and other Fourier transform methods
  • Investigate techniques for peak detection in FFT results
USEFUL FOR

Researchers, physicists, and engineers working with wave phenomena in two-dimensional spaces, particularly those utilizing FFT for wavevector analysis.

Crot
Messages
15
Reaction score
0
Dear forumers,

There is a wave computed in a 2D domain.
Could you please give me a hint how to find the wavevector of the wave?
I think I have to Fourier transform the wave in space but I cannot figure out how to
obtain that k_x and k_y from the transformation.


Crot
 
Physics news on Phys.org
There other ways to do this, but FFT is a straightforward (if brute force) approach.

You will want to look up the reference for your FFT routine (if you are using a prepackaged one). Typically, the FFT will give you amplitudes for +/- k_x, k_y with the index being wavenumber expressed as integer multiples of cycles per domain length (since the longest periodic wave in the domain completes 1 cycle). However, for the N-point FFT, you will likely get an order like -N/2, -N/2+1, ..., 0, N/2-1 cycles per domain for even N (or -N/2,...,0,...,N/2 for an odd N).

Anyway you are just looking for the peak wavenumbers in both k_x and k_y, right?
 
Thanks a lot for the answer. Now, it is clear!
 

Similar threads

  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 4 ·
Replies
4
Views
3K
  • · Replies 6 ·
Replies
6
Views
2K
  • · Replies 1 ·
Replies
1
Views
3K
  • · Replies 3 ·
Replies
3
Views
3K
  • · Replies 27 ·
Replies
27
Views
2K
  • · Replies 11 ·
Replies
11
Views
3K
  • · Replies 36 ·
2
Replies
36
Views
9K
  • · Replies 1 ·
Replies
1
Views
1K
  • · Replies 1 ·
Replies
1
Views
4K