Rafael, > Could you please provide the data points in your example so that we can > test different methods.
See attachment. To view an example of the discontinuities, zoom in on the interval between 0.4 and 0.5 T (Y-scale). I normally have to zoom twice to get a good spread. > > Note that in the moving median filter solution presented there is no > propagation of errors because the original dataset is always used and > only one filtering pass is made using a very short 3-point filter. Okay. I see that (now). However, looking at my data, I don't think that a single pass with a window of 3 will produce the results I need. That means either widening the window, or iterating the filter (AFAIK?). > > Regards, > Rafael > Thanks for looking at this. Buk. ____________________________________________________________ Can't remember your password? Do you need a strong and secure password? Use Password manager! It stores your passwords & protects your account. Check it out at http://mysecurelogon.com/password-manager
HB1M_Core25_No_Field_No_Epoxy_800Am.initial.rdat
Description: Binary data
_______________________________________________ users mailing list users@lists.scilab.org http://lists.scilab.org/mailman/listinfo/users