Skip to content

EoRImaging/FHD

Repository files navigation

FHD

Fast Holographic Deconvolution

Introduction

FHD is an open-source imaging algorithm for radio interferometers, specifically tested on MWA Phase I, MWA Phase II, PAPER, and HERA. There are three main use-cases for FHD: efficient image deconvolution for general radio astronomy, fast-mode Epoch of Reionization analysis, and simulation. A license for IDL 8.2 or above is required.

For further details, please see Sullivan et al 2012 or Barry et al 2019a.

Quick Start Guide

  1. Acquire an IDL 8.2 or above license and install.
  2. Install FHD.
  3. Acquire a data file in the uvfits format. For an explanation of the format and instructions for acquiring public MWA data is available in the Inputs resource.
  4. Create a wrapper script with the appropriate keywords. For examples of FHD run wrappers see the Examples resource. For an explanation of the keywords see the Dictionary resource.
  5. Read and abide by the Community Guidelines.

Useful Documentation Resources

  • Dictionary: FHD is run by specifying keywords in a wrapper that is passed to the main code. These keywords change the functionality of FHD for fine-tuning and flexibility. Please visit the dictionary for keywords and their definitions.
  • Assumptions: Some mathematical assumptions are built into the framework of FHD. Please visit the assumptions page to explore the parameter space.
  • Examples: Please visit the examples page for brief tutorials on specific cases.
  • Outputs: For documentation of the various outputs, see the outputs page.
  • Inputs: FHD inputs beam models, calibration files, and sky model catalogs. Currently available inputs are described on this page.
  • Publications: FHD has been used in a variety of publications, documented on this page.

Installation

Download the latest versions of these libraries/repos.
a. This repository
b. The fhdps_utils repository
c. The NASA astro IDL library
d. The coyote library
e. The HEALPix library
f. Optional: the eppsilon power spectrum generation code

Add these libraries to your IDL path. Be sure to use the correct separator (; or : ) and include the ‘+’ sign at the start of +path/to/library/. Optionally, you can create an IDL startup file to set the IDL path.
a. Windows: !PATH = !PATH + ';' + Expand_Path('+path\to\library\')
b. Unix: !PATH = Expand_Path('+path/to/library/') + ':' + !PATH

Install HEALPix

Install Imagemagick if not already present.

Suggested test commands:
a. print,cgHasImageMagick()
prints error if coyote library not installed, returns 0 if coyote library installed but Imagemagick not installed correctly, and returns 1 if both are installed correctly.
b. astrolib
prints error if astro IDL library not installed correctly, prints message “ASTROLIB: Astronomy Library system variables have been added” if installed correctly.
c. init_healpix
prints error if HEALPix not installed correctly
d. imagefast,randomN(5,256,256),file_path='some/output/path/testimage.png'
prints error if FHD not installed correctly, generates a 256x256 pixel image of poisson noise at path 'some/output/path/testimage.png' if installed correctly and read/write permissions are set correctly.

Community Guidelines

We are an open-source community that interacts and discusses issues via GitHub. We encourage collaborative development. New users are encouraged to submit issues and pull requests and to create branches for new development and exploration. Comments and suggestions are welcome.

Please cite Sullivan et al 2012 and Barry et al 2019a when publishing data reduction from FHD.

Maintainers

FHD was built by Ian Sullivan and the University of Washington radio astronomy team. Maintainance is a group effort split across University of Washington and Brown University, with contributions from University of Melbourne and Arizona State University.