Tutorials

Step-by-step examples that solve a concrete problem with SVETlANNa.

Every tutorial contains complete, runnable code. We recommend following along in a Jupyter notebook or a Python script.

Beam focusing 4f system Phase grating on an SLM Laguerre–Gaussian beam Phase retrieval

Difficulty

Tutorial	Level	Key concepts
Beam focusing	Beginner	`ThinLens`, `FreeSpace`, `Wavefront`
4f system	Intermediate	Fourier optics, spatial filtering
Phase grating on an SLM	Intermediate	`SpatialLightModulator`, diffraction orders, quantisation
Laguerre–Gaussian beam	Intermediate	Custom wavefronts, vortex beams, free-space propagation
Phase retrieval	Advanced	Iterative algorithms, Gerchberg–Saxton

Suggested order

Beam focusing

The basics: focusing a Gaussian beam with a lens. Introduces Wavefront, ThinLens and FreeSpace.

4f system

A classical 4f setup for optical filtering — Fourier optics put to work.

Phase grating on an SLM

A sinusoidal phase grating displayed on a SpatialLightModulator; the diffraction orders are compared with the Bessel-function prediction.

Laguerre–Gaussian beam

Building a vortex mode from scratch on the simulation grid and propagating it through free space, validated against the analytical solution.

Phase retrieval

The Gerchberg–Saxton algorithm: recovering phase from intensity measurements.

Prerequisites

Before you start we suggest reading:

Core concepts — the architecture of the library
SimulationParameters — setting up the grid
Optical elements — the available elements
Wavefronts — working with fields

Running the examples


# minimal template used throughout the tutorials
import torch
from svetlanna import SimulationParameters, Wavefront
from svetlanna.units import ureg
 
params = SimulationParameters.from_ranges(
    x_range=(-2*ureg.mm, 2*ureg.mm), x_points=512,
    y_range=(-2*ureg.mm, 2*ureg.mm), y_points=512,
    wavelength=632.8*ureg.nm
)
 
wf = Wavefront.gaussian_beam(params, waist_radius=0.5*ureg.mm)
print(f"Ready. Field shape: {wf.shape}")

To run on a GPU, install the CUDA build of PyTorch and call params.to('cuda') before creating any elements.