AOMODEL

AOMODEL#

AOMODEL is a Python package for generating synthetic aero-optic phase screen data.

The main feature of this package is an implementation of the Re-whitened Vector AutoRegression (ReVAR) algorithm, which generates synthetic time series of images that match the statistics of measured data.

This package can be used to:

  • Fit the statistics of input data.

  • Generate synthetic data with the desired statistics.

  • Evaluate the statistics of an input data set.

For more information about the algorithm, see [4, 7].

See also the Boiling Flow package on GitHub: boiling_flow. This package implements a Fourier-based algorithm for generating synthetic aero-optic phase screens, including automatic parameter estimation from measured data. This model is designed for highly convective data with spatially stationary (i.e., homogeneous) statistics.

The benefits of Boiling Flow compared to AOModel are:

  • Ability to generate phase screens of any size.

  • The use of physically relevant parameters.

The drawbacks of Boiling Flow compared to AOModel are:

  • Synthetic data generated by Boiling Flow has less accurate statistics.

  • The model is less generalizable to different measured data sets.

Use AOModel if you need to generate synthetic data with statistics that closely match those of the measured data, or if the measured data is not accurately modeled by boiling flow. Use Boiling Flow if your measured data falls into the boiling flow regime and you need to extend the phase screens beyond the size of the measured data images. Alternatively, Boiling Flow can be used to estimate turbulence parameters from measured phase screen data.

Disclaimer: Approved for public release; distribution is unlimited. Public Affairs release approval # AFRL-2026-1309.