FAST (Fourier domain Adaptive optics Simulation Tool)

FAST is a simulation tool that utilises a Fourier domain AO model to enable rapid Monte Carlo characterisation of free space optical links between the Earth and satellites. For more details, see the paper https://doi.org/10.1364/OE.458659.

Documentation is hosted at ojdf.github.io/fast.

WARNING: FAST is under active development and the simulation API and configuration is not fixed. If you find a bug, or something is not working as you would expect, please email me or make an issue on GitHub, it's likely I've broken something!

Requirements

See requirements.txt

It is recommended to install pyFFTW to speed up FFTs.

Installation

PyPI

pip install fast-aosim

From Source

For the latest developments, clone this repository run pip install -r requirements.txt to get the required dependencies. Then, ensure that the FAST directory is on your PYTHONPATH, or navigate to the directory and run

pip install .

To confirm correct installation, try running

cd test && python test_script.py

in the FAST directory, which should run a short simulation. You can also run the unit tests with

pytest -v test/tests_pytest.py

Configuration

Config is handled currently by python scripts, an example of which is shown in test_params.py. The config file should define a dictionary p which contains all of the configuration information required. If any values are missed, they will be replaced by defaults.

To start a simulation either in the python shell or in a script, you can either pass the filename of the script defining the config dictionary, or you can provide the dictionary itself, which can be useful if you are scanning through parameters. So

import fast
sim = fast.Fast("your_config_file.py")

or

import fast
p = {your config}
sim = fast.Fast(p)

Running the simulation

The simulation is run by calling the run() function on the sim object. This will compute the phase screens and log-amplitude values, and then compute the detected power or phase/amplitude for coherent detection. The results are stored in sim.result and are also returned from run(), i.e. res = sim.run(). Results in either dB or power units can be obtained from e.g. sim.result.dB_rel (for dB relative to the diffraction limit) or sim.result.power. See the documentation for more details.

To save the simulation results as a fits file with header information, use sim.save(filename).