A pipeline for the analysis and simulation of drift scan radio data.
draco is a set of building blocks designed for the analysis of the transit radio data with the m-mode formalism (see the papers arXiv:1302.0327 and arXiv:1401.2095 for details). It is being used as part of the analysis and simulation pipeline for CHIME though is (and will remain) telescope agnostic. It can:
- Simulate time stream data from maps of the sky (using the m-mode formalism)
- Add gain fluctuations and correctly correlated instrumental noise (i.e. Wishart distributed)
- Perform various cuts on the data
- Make maps of the sky from data using the m-mode formalism
It does not do some of the key steps in radio data analysis, notably RFI flagging and calibration. The implementations we had were too specific to CHIME so they have been left out, until a more generic version is produced.
To do this it depends on various related packages:
- driftscan: for modelling the telescope and generating the computationally intensive products required for simulation and analysis.
- cora: for modelling and simulating the radio sky
- caput: provides infrastructure for building these packages
It also depends on the usual suspects: numpy
, scipy
, healpy
, h5py
and skyfield
.
draco can be installed with pip
in the usual way:
$ pip install git+https://github.com/radiocosmology/draco.git
or by downloading the package and running the setup.py
script:
$ cd draco
$ python setup.py install
The full documentation of draco
is at https://radiocosmology.github.io/draco/.