Flagging CABB data

Is MIRIAD being a pain? Let us know your experience.

Moderator: Mark.Wieringa

Post Reply
ste616
Site Admin
Posts: 220
Joined: Thu Feb 04, 2010 3:27 pm
Location: Paul Wild Observatory Narrabri NSW

Flagging CABB data

Post by ste616 »

Flagging data from CABB can be quite difficult as there are a large number of channels and the data volume is great. MIRIAD has undergone some improvements recently to assist users while flagging CABB data.

The routine blflag has been improved, and a new option made available: nofqaver. Using this option means that blflag will not average all channels together before displaying them, so now you can use the same point-and-click flagging as before with a large bandwidth where only narrowband interference is present. You can also now specify channel as one of the blflag axes.

The new routine pgflag is a replacement for the obsolete tvflag, which had the significant problem of not being usable on screens with more than an 8-bit colour depth. pgflag uses standard PGPLOT displays, and allows you to flag via a familiar waterfall plot of channel vs time. All of the actions of tvflag are replicated, along with some new actions, although they are now all controlled through keypresses instead of an on-screen menu, so be sure to read the MIRIAD man page for pgflag before beginning.

There is also an automatic flagger in development by Emil Lenc based on pieflag. Contact Emil if you'd like to know more (and hopefully he will reply to this post with a description of his flagger!)
cheers
Jamie Stevens
ATCA Senior System Scientist
ste616
Site Admin
Posts: 220
Joined: Thu Feb 04, 2010 3:27 pm
Location: Paul Wild Observatory Narrabri NSW

Re: Flagging CABB data

Post by ste616 »

This thread was becoming quite informative about mirflag, but may have become unruly for later reading about general CABB flagging.

So I've moved the other posts to a new thread, which is all about mirflag! It can be found at http://atcaforum.atnf.csiro.au/flagging ... g-t13.html
cheers
Jamie Stevens
ATCA Senior System Scientist
cat001
Posts: 14
Joined: Tue Mar 09, 2010 3:58 pm

Re: Flagging CABB data

Post by cat001 »

Hi Jamie,

For atlod, we have the option of using rfiflag. Which flags channels at frequencies that are know to be bad. I looked at the file where the bad channels are defined: /nfs/atapplic/miriad/cat/rfiflag.txt, and the channels 513, 1025, and 1537 which are know bad channels are not in the file.
Are the channels suppose to be in the file? To add bad channels (or create my own file of bad channels), is the first and last number included in the flagging? so if I were to add 513 514 which channel is it flagging? 513,514 or both?

thanks,
Catarina
ste616
Site Admin
Posts: 220
Joined: Thu Feb 04, 2010 3:27 pm
Location: Paul Wild Observatory Narrabri NSW

Re: Flagging CABB data

Post by ste616 »

Hi Catarina,

Channels 513, 1025 and 1537 are flagged by the correlator in the RPFITS file it writes. Data from channels flagged by the correlator (either automatically or by using the fflag function) cannot be recovered, since the correlator only writes 0 for these channels, and then flags them. So to directly answer your first question: no, there is no need for these channels to be in the file since they will always be flagged in all files.

Now, it is true you can make your own file that atlod will then use as its flagging source. It is advisable that you copy /nfs/atapplic/miriad/cat/rfiflag.txt to your working directory, and add rules to that file, since atlod will only read the first flagging file it encounters. However, you cannot specify bad channels with this file, only bad frequencies. In that way, RFI can be flagged independent of what the central frequency was for the observation. If you want to flag bad channels, you will need to do that manually with uvflag.

I hope this helps!
cheers
Jamie Stevens
ATCA Senior System Scientist
madsarah
Posts: 3
Joined: Wed Mar 17, 2010 5:19 pm

Re: Flagging CABB data

Post by madsarah »

Hey Jamie,

Can you tell us (or is there a file somewhere that tells us) which channels the "birdie" option in atlod flags due to the CABB 640 MHz clock harmonics self-interference? At 7mm is it something like 257,641,1409,1793 ...? Given that birdie also flags 100 edge channels each side, we wondered whether you could flag the bad birdie channels by hand so that you don't have to lose 100 channels as well.

Thanks,
Sarah.
Mark.Wieringa
ATCA Expert
Posts: 297
Joined: Mon Feb 08, 2010 1:37 pm

Re: Flagging CABB data

Post by Mark.Wieringa »

Hi Sarah,

the channels flagged are 640,256,768,1408,1280,1920,1792,1176,156,128,1152 plus 1 (for each), this is approximately the order of strength.

The 100 edge channels really are not adding much, they are on the edge of the 2 GHz filter. I based the number on what people were using in their CABB reduction scripts.

Cheers,

Mark
Post Reply