" RFI Eraser " noise reduction processing
has been added to the HL3D Java code.
It seems stable & effective, but I have a few more ideas for improvement.
Suggestion :
Process the data first with Filter Width = 0 to get an uncorrected baseline of your data
IF there are RFI spikes in the spectrum, try Filtering
Start with small filter width values, save the *.CSV file set & compare via Rinearn .
Over-correcting degrades the H data.
Alex Pettit
The graph contains 3 plot sets :
Uncorrected Data with RFI spikes
Small Filter Width correction results
Larger Filter Width correction results