Version 1.9 Build 1367

News	FAQ
Search	Home

Next: Data Display and Editing Up: No Title Previous: Introduction

Subsections

General Nature of the Data

This section includes requirements which will affect the data system explicitly, or implicitly because of the nature of the observed data.

Instrumental Requirements

A number of requirements are generally applicable to single dish and interferometer data:

A full measurement of the electromagnetic field requires four complex (polarisation) parameters, which can be converted to Stokes parameters; support for this should be fundamental.
Polarisation measurements may be time switched if all four polarisation parameters are not correlated simultaneously.
Multiple frequency bands may be simultaneously observed (e.g. for observing multiple lines simultaneously or multi-frequency synthesis), with variable numbers of channels within each band.
The frequency axis may be non-linear (e.g. as produced by acousto-optical spectrometers) and time variable.
Data for combination from different observations may have different (but overlapping) numbers of spectral channels and channel widths. These may need to be accomodated within the one dataset.
Focal plane arrays with arbitrary geometry (e.g. field rotation through the observation) characteristics must be supported.
Mosaiced observations may have many ( $\sim$ 1000) pointing centres. This must be supported at both the uv dataset and image dataset level.
The measured polarisations, frequencies and pointing centres may be rapidly time switched (e.g. these may change every integration).
Time-series data of profiles and visibilities (e.g. pulsar data with bin number as a data axis) needs to be supported.
Error measures or estimates (e.g. weights) should be regarded as standard in an observation.
Correlation data may be 16-bit integers or 32-bit floating point.

Requirements specific to single dish observations

Total power time series.
Total power measurements, irregulary spaced in space and/or time for imaging.
Series of spectra, including the case where the size of the spectrum is variable.
Bit-field data for raw pulsar data.
Interferometer autocorrelation data.

Requirements specific to interferometry

Interferometer arrays should be regarded as being generally inhomogenous (although this should not preclude taking advantage of homogeneity where possible) in a number of ways which should be handled transparently whenever possible:

Polarisation characteristics (e.g. linear or circular feeds and equatorial or alt-azimuth mounts) differ among antennas or arrays of antennas.
Antenna sizes and system temperatures may differ widely.
The procedures and input data for a priori antenna calibration may vary from antenna to antenna.
Space VLBI will require support for variable antenna positions.
Integrations times may vary from baseline to baseline.

VLBI

VLBI support is implicit in many requirements, but in addition, support must be provided for the many correlator formats which will prevail, including MkII, S-2 and K-4, as well as the MkIII and VLBA modes.

The data system should support the merging of correlation data and calibration data from different correlators and allow the user to deal with duplicate correlations.

Monitor/history data and site/instrument specific data

There should be support for history associated with data. It should be possible to import and make use of history information from ``previous incarnations'', such as on-line observing logs.

Monitor and meterological data need to go directly into dataset.
Easy access to observe-time information which cannot be put into the dataset at the time of observation.
Flexible addition to the dataset of observatory dependent monitor data.

Other observed data types

It is important that the data system be as extensible as possible to support new data data types in future:

Triple correlation data, including the case where one of the visibilities has a different frequency.
Optical interferometer data.
Images; at some level, the data system should support the requirements of image handling, rather than having one data system for images and another for all other data. In general, ``vector images'' should be supported, allowing scope for complex images, with associated errors etc., as well as double precision images.

General (Non-Instrumental) Data-System Requirements

The data system file format(s) should be accessible from all supported machines without conversion.
There should be no distinction between ``multi-source'' and ``single-source'' files.
Applications should function on data and data-cubes in arbitrary sort order; all sorting should be hidden to the user.
Coordinate handling must be very general:
- Support for data with non-regular increments in coordinate axes (such as an optical velocity axis).
- Coordinate systems and ephemeris information required for astrometry, and near field imaging will be supported.
This will also assist in the merging of data sets.
Support for data errors should be fundamental to the data system, providing a basis for support of data error handling in a variety of applications:
- Easy generation of error models;
- Error propagation through a series of tasks;
- Error images associated with astronomical images;
- Plotting error bars on spectra and profiles;
- Automatic warning if contour levels are below noise;
- Error-based blanking in display of results;
- Properly formatted errors when data are extracted for tabulation.
Simultaneous processing of ``associated'' datasets, such as different (e.g. by calibration, integration time, fringe fitting etc.) versions of the same observation, so that the best can be selected later.
A processing history should be maintained for each data set that can be reviewed by the astronomer.
The user should have access to both data values and the `header' values that govern the interpretation of the data.

Next: Data Display and Editing Up: No Title Previous: Introduction