AIPS HELP file for ROBUST in 31DEC09

As of Sun Nov 22 16:52:09 2009

HELP SECTION

ROBUST
Type: Adverb   (Real)
Use:  The robustness parameter, devised by Daniel Briggs, is
      used to temper the effects of uniform weighting.  Thus,
      the weight actually used for sample i is
                             F1 (Wi)
              WT  =   ----------------------
                        [F2 (Wj)]   +    S
      where F1 and F2 are 2 functions applied to the input
      weights (Fi (W) = W by default), the [ ] means the sum of
      the F2 of the Weights of all samples in the neighborhood
      of the i'th sample, and S is a function of ROBUST.  We use
                      (10**(ROBUST)) * <[F2 (Wj)]>
              S  =   --------------------------------
                                    5
      where < > denotes the average of the local sums over all
      visibilities.  Note that if ROBUST > 4, the weight sums
      have little effect and the weighting is close to natural
      weighting.  If ROBUST < -4, then S is very small and we
      have virtually pure uniform weighting with no tempering.
Null value:   NONE
Tasks:
  IMAGR.....Fourier transform and Clean UV data with a full
            range of calibration, weighting, Cleaning, and
            wide-field correction options.

EXPLAIN SECTION