average in frequency before evaluating r.m.s.

[Fil8]

Right now RFInder estimates the RFI and scans through visibilities at the original spectral resolution of the observation.

The frequency binning is done after having found the RFI, considering the average of the % of RFI flagged per bin.

In principle one could average the visibility over frequency and then estimate the RFI in those large averaged bands?

• Does the RFI investigation results change?
  • It should not matter for the % of FLAGS, but it could be faster using casacore (see (issue #11 ))