 News |
FAQ
|
| Search
|
Home
|
| Getting Started | Documentation | Glish | Learn More | Programming | Contact Us |
|
| Version 1.9 Build 1367 |
|
The ``atomic measurement'' produced by a telescope is known as a Yeg. Exactly what is the most atomic piece of data will vary between the different types of telescope. For example, for an interferometric array, a correlation or visibility (the output of a single correlation channel, for a given baseline, at a given time) is the most atomic piece of data. For a single dish, it may be a single total power measurement for a single integration time. Although the Yeg is the atomic unit as seen at the level of an application program, this is not to say that this will be the basic unit of storage. The result of a real observation includes a set of Yegs and other information (which can be placed in TelescopeModel as described below).
A Yeg is basically unintelligent. Apart from the data values, it knows little more than the parameters needed to tie it to the telescope setup used to observe the Yeg. For example, a Yeg for an interferometer will contain a real and an imaginary part, a weight (or perhaps noise estimate), time stamp, some way of determining its correlator element, and perhaps some direct way of determining its receptors. An Interferometric Yeg may also contain u,v and w coordinates although these are examples of attributes which might be calculated ``on the fly''. The split between what belongs in a Yeg, and what should be determined using the TelescopeModel is discussed later, but may not always be obvious.
All Yegs contain a reference to their parent Telescope object. A Telescope object embodies a framework which relates the Yeg to the sky. This framework is embodied in references to two sorts of objects, a TelescopeModel and an ImagingModel. The intelligence for interpreting a Yeg (e.g., how to calibrate a Yeg or how to form an image from a set of Yegs) resides in these models. The Telescope object does not attempt to describe a real telescope - it merely references models of how we believe the telescope behaves. The actual models used will vary according to the actual telescope, the observing modes, and the astronomical requirements. The models may be very simple or as sophisticated as required. For example, a simple TelescopeModel may simply include the antenna-base gains for each antenna or Receptor; a complex VLBI model would include many more effects such as polar axis motion. Similarly the ImagingModel might be a conventional two-dimensional Fourier transform method, or a three-dimsnsional Fourier transform for wide-field imaging.
A complete observation database will be described by a set of Yegs known as a YegSet (the measured data) and the associated Telescope, embodying models of how these are related to the sky.In this sense, a GBT observation would have a single Telescope object, as would a VLBI observation or a 12 hour run by the VLA, etc.