next up previous contents
Next: Other AIPS++ Dependencies Up: SDCalc Design Previous: IO Issues

Operations

The following summary of operations that SDCalc will provide was generate by a quick re-reading of the UniPOPS manual plus a re-reading of the wish list for the GBT that Rick Fisher has written. These are the immediate operations (for beta release) that SDCalc will offer. Additional operations will be added. A section of this document should eventually appear describing those additional operations.

Typical Calibration Schemes
There are several levels here. The GBT needs these schemes in order to do the very basic operation of combining the several sub-integration phases (on, off, on+cal, off+cal, etc.) into a single integration. More generally, there are probably a small set of standard operating modes used universally that we should be able to reduce automatically (position switched, frequency switched, beam switched, etc). The issue here, can these be described in a telescope independent manner. The sub-integration to integration averaging will in general be telescope specific. It would be nice if we could plug in telescope specific methods to some uniform place in SDCalc. The generic calibration operations need to be catalogued and described.
Shift and Align
Given a starting SDRecord, shift any other SDRecord with a compatible x-axis so that the x-axes align. There should be a number of user controllable shift methods from integer only, to simple spline interpolation to a full FFT based shift.
Average SDRecords together
Allow for different averaging schemes.
Fitting
The result should go in a number of different places as indicated by the user (attach as a parameter vector or fitted values to the SDRecord, to a Glish record, use it immediately by removing it from the data, store the residuals in place of the data). For baseline removal, it should be possible to use the baseline from another SDRecord (or even use an SDRecord as if it were a fitted baseline, which it might in fact be).
Smoothing
Boxcar, Hanning, Gaussian, arbitrary function
Interpolation
To a different x-axis.
General FFT tools
Moments
Galaxy profile analysis
Stolen from UniPOPS which in turn was stolen from Analyze. This is included here for completeness. The operations summarized here represent all of the one-dimensional functionality of UniPOPS not already found in Glish. This capability will be developed last after a few rounds of feedback from the initial users of SDCalc.


next up previous contents
Next: Other AIPS++ Dependencies Up: SDCalc Design Previous: IO Issues

Bob Garwood
Fri Jul 11 17:07:42 EDT 1997