Filter löschen
Filter löschen

How to compute diffraction integral using fast Fourier transform (fft2) ?

8 Ansichten (letzte 30 Tage)
Sohail Ahmed
Sohail Ahmed am 13 Jun. 2024
Kommentiert: Sohail Ahmed am 18 Jun. 2024
Hi there !
The following code computes Fresnel diffraction integral using fft2. It works for (obj.E = ones(Ny, Nx);) and performs the fft2 analysis, but when the input is (obj.E = zeroes(Ny, Nx);), it plots nothing. I checked the same code on python with zeros and it worked but in MATLAB it is not performing fft2 for zeros (gives only zero values). Kindly help me resolve this issue.
The Sheet.m (code):
classdef Sheet
properties
dx
dy
x
y
xx
yy
Nx
Ny
E
end
methods
function obj = Sheet(extent_x, extent_y, Nx, Ny)
obj.dx = extent_x / Nx;
obj.dy = extent_y / Ny;
obj.x = obj.dx * ((0:Nx-1) - floor(Nx/2));
obj.y = obj.dy * ((0:Ny-1) - floor(Ny/2));
[obj.xx, obj.yy] = meshgrid(obj.x, obj.y);
obj.Nx = Nx;
obj.Ny = Ny;
obj.E = zeros(Ny, Nx);
end
function rectangular_slit(obj, x0, y0, lx, ly)
% Creates a slit centered at the point (x0, y0) with width lx and height ly
mask = ((obj.xx > (x0 - lx/2)) & (obj.xx < (x0 + lx/2))) & ((obj.yy > (y0 - ly/2)) & (obj.yy < (y0 + ly/2)));
obj.E = obj.E + mask;
end
end
end
The main code :
% Simulation input
Lx = 1.4;
Ly = 0.4;
Nx = 2500;
Ny = 1500;
sheet = Sheet(2*Lx, 2*Ly, Nx, Ny);
% Slit separation
mm = 1e-3;
D = 128 * mm;
sheet.rectangular_slit(-D/2, 0, 22 * mm, 88 * mm);
sheet.rectangular_slit(D/2, 0, 22 * mm, 88 * mm);
% Distance from slit to the screen (mm)
z = 5000;
% Wavelength (mm)
lambda = 18.5*1e-7;
k = 2*pi / lambda;
% Initialize complex array for phase information
fft_c = fft2(sheet.E .* exp(1i * k/(2*z) * (sheet.xx.^2 + sheet.yy.^2)));
c = fftshift(fft_c);
% Plot with MATLAB
abs_c = abs(c);
% Screen size (mm)
dx_screen = z*lambda/(2*Lx);
dy_screen = z*lambda/(2*Ly);
x_screen = dx_screen * ((1:Nx) - Nx/2);
y_screen = dy_screen * ((1:Ny) - Ny/2);
figure;
subplot(2, 1, 1);
imagesc(x_screen([1 end]), y_screen([1 end]), abs_c);
colormap(gray);
axis equal;
xlabel('x (mm)');
ylabel('y (mm)');
title('Probability Density |\psi|^2');
xlim([-2, 2]);
ylim([-1, 1]);
subplot(2, 1, 2);
plot(x_screen, abs(c(Ny/2, :)).^2, 'LineWidth', 1);
xlabel('x (mm)');
ylabel('Probability Density |\psi|^2');
title('Probability Density |\psi|^2 along y=0');
xlim([-2, 2]);
  3 Kommentare
David Goodmanson
David Goodmanson am 17 Jun. 2024
Hi SK. if the input is all zeros, is there a reason that you expect that the output would not be all zeros? If anyting I might be questioning the python code that gives nonzero output for zero input.
Sohail Ahmed
Sohail Ahmed am 18 Jun. 2024
Yeah, there is a reason. The issue arises with the zero padding of the Fresnel inetgral while applying fft2.

Melden Sie sich an, um zu kommentieren.

Antworten (0)

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by