[daip] Re: Spectral smoothing

[Eric Greisen]

Karl Menten writes:
 > Dear Eric,
 > 
 > I need to apply spectral smoothing on a data cube to get rid
 > of ringing due to  inadequate velocity resolution (no continuum).
 > 
 > Hanning should take care of this. However, if I run XSMTH
 > with  OPCODE 'SMOT' and DPARM 1,4,0 and inspect the result with
 > ISPEC the ringing is still there. The intensity is reduced, but
 > the line doesn't look much broader.
 > 
 > When in ISPEC, I choose SMOOTH=1,0 the intentity is reduced by the
 > same amount as above, but no ringing remains. It seems as if in XSMTH
 > the weighting is done incorrectly. Could you please check this?
 > 
 > Another curiousity is that when XSMTH is used with OPCODE 'SMOT'
 > the NPOINT option does not work, i.e. the number of output channels
 > equals the number of input channels. For Hanning smoothing it would
 > make sense if the number of the output channels would be equal to the
 > number of the input channels minus 1.
 > 
 > I'm using 31DEC03 AIPS.
 >

The problem is that DPARM=1,4 is not Hanning but Gaussian.  DPARM=0,4
is Hanning.  Try that and let me know if the results are better.

Eric Greisen