[daip] help with UVLSF, please

[Gustaaf van Moorsel]

On 11 Dec 2007, at 22:12, Eric Greisen wrote:

>> UVLSF with high order will only work if there is really a significant
>> continuum signal and you can use e.g at least half the channels in the
>> fit.  In particular, the lower numbered channels need to be line
>> free.

It seems to me that Elias' data suffered from a lack of continuum,
not so much a lack of channels at each end.

If that is the case, and the aliasing is not very time variable,
as my tests suggest, couldn't we do the UVLSF fit on the bandpass
calibrator (or maybe even better, the phase calibrator) only?
Surely in those cases the continuum is strong enough for UVLSF to
work reliably.  That will give us the correct functional form of
the aliasing for each baseline.  We can then use this functional
form for the source data, which then only need to be scaled,
thereby reducing the number of free parameters from 4 to 1.

I don't know how feasible it is to implement such a thing; for now
I'd like to hear if you think this is an approach that might work.
For instance, is the shape of the aliasing on a calibrator trans-
ferrable to the source?

Gustaaf