Getting Started

Documentation

Glish

Learn More

Programming

Contact Us

Version 1.9 Build 1556

News	FAQ
Search	Home

Subsections

• Self-calibration and Imaging
• Image analysis and U,V model fitting

Image Formation and Analysis

The ability to process CE data implies that the AIPS++ infrastructure development covers several aspects of VLBI processing. Notably phase (only) and amplitude self-calibration routines, coordinate handling, model-fitting, imaging algorithms will become available because of these demands.

Self-calibration and Imaging

For VLBI processing it is important that (self-)calibration, data inspectionediting, transformation and deconvolution are integrated closely, as is available in DIFMAP. When implementing such features special care should be taken to warn (novice) users against the dangers of making the ``data fit the model''. It should be possible to easily ``mix-and-match'' self-calibration, transformation, and deconvolution tools, for example, using CLEAN to deconvolve in the early stages, and NNLS or MEM later on when CLEAN would begin to break-down. This also demonstrates the need to make self-calibration use a generic model, which may be CLEAN-components, an image, or a Gaussian model, or a combination.

More specific VLBI considerations are listed below:

• it should be possible to edit the list of clean components for self-cal

• the possibility of using CLEAN polygons should be included

• difference mapping should be supported
• complex deconvolution is required to handle ``asymmetric'' arrays (including telescopes with single polarization feeds)
• fringe rate mapping (AIPS equiv: FRMAP)

Image analysis and U,V model fitting

Although this strictly not an ``image analysis'' tasks, we note that model fitting should be possible in both the image as well as u,v-planes using point source, Gaussian or spherical models. It should be possible to use the resultant model in the same way as a CLEAN component model in calibration and self-calibration. Specific items are listed below:

• joint fitting for all four polarization parameters (as is done in the Brandeis package)
• for Space VLBI it is required that model-fitting can cope with sparse u,v coverage
• it should be possible to display model data (i.e. from CLEAN components, fitting etc.) with observed andor processed data simultaneously as well as data with model subtracted or divided

• there should be tools available to do spectral line profile fitting, in order to derive velocities ands widths

• in contour plotsgrey scale plots, choose coordinate system when creating the plot file

• possibly tools for gravitational lens transformation models
• fitting of Zeeman profiles should be possible

• it will be possible to develop tools to do proper motion analysis

In general we think that AIPS++ should improve the capabilities to produce publication quality figures from the data reduction package. For the VLBI community it important to be able to plot maps of sources at different epochs or frequencies in a single figure.