#Terminology

• source - a unique source name.
• obsid - a unique observation identifier: <SOURCE>_<MJD{.5f}>
• obsidt - an obsid plus tuning: <OBSID>_<A/B/C/D>
• obsrec - a unique observation/hardware identifier: <OBSIDT>_<CNODE>
• obsrec file - the filename holding the obsrec information (generally <OBSREC>.npz, see note below)
• tracks - an on_tracks.py class that contain all info to generate a track along with metadata etc.
• obsinfo - a file containing information on obsids for a given experiment: obsinfo_<MJD>.json. Among other fields, it has the key "Sources" that contain a list of tracks
• ods - operational data sharing
• tba - telescope boresight avoidance

note: uvh5 files are put into directories that map to an obsrec:
/mnt/primary/ata/projects/pID#/YYYY-MM-DD-HH:MM:SS/uvh5...../Lo[A/B/C/D].C[####]/uvh5......uvh5

#Observing Recipe The focus right now is to observe the Starlink Direct-to-Cell (DTC) satellites.

##Find satellites## Need to find and set up the satellites. This is currently clumsily interactive. The main problem is that THERE ARE SO MANY!

Be sure to update the tle files with on_updatetle.py. The default is that there is a subdirectory called tle where it puts and reads the information.

Make sure that you have an appropriate parameters.yaml file.

Use the planning tool from within ipython and get 'tracks'

from obsnerd import onp_plan
plan = onp_plan.Plan()
plan.get_tracks(satname='some-search-string', start='now+20m', duration=number_of_minutes, auto=True)

If auto=True you can't chose more and it will automatically go through all of the steps below.

For example plan.get_tracks(satname='*', start='now+20m', duration=60, auto=True)

Also, if auto=False and if this looks like a good set, choose the ones you want (there is an auto-choose mode by default):

plan.choose_tracks(auto=True)

If you want to continue with that set,

plan.proc_tracks()

This will write obsinfo_<MJD>.json/.npz

##Observe##

1. on the VNC open a terminal
2. put the file obsinfo_<MJD>.npz onto obs-node1 as ~/rfsoc_obs_scripts/p054/obsinfo_rados.npz
3. type on_obs_prep.py --add-to-calendar
4. type aoctkuser.py --enable-rados
5. hit the Observe button and if you are confident select yes twice.
6. sit back and watch the action

##Process the data## The data now sits in that deeply nested directory structure in the note above. You now want to dump the autocorrelations for the antennas you want to much smalller 'npz' files, which means you need to find them, generate a bash script, and run the bash script.

Do this while logged into obs-node1 and in the ~/rfsoc_obs_scripts/p054 directory

1. Find them by typing on_gen_dump.py ? at the terminal
2. Generate the scripts by typing on_gen_dump.py <date-from-above> (use on_gen_dump.py -h for options). This generates two scripts: dump_autos.sh and download_files.sh.
3. Run the generated script bash dump_autos.sh (this will take a very long time so do it under a screen). This generates npz files containing the specified autocorrelations per antenna/polarization/cnode. The name of the file is the obsrec defined above
4. Move the download_files.sh to your local machine and run to download the files if you wish to process them locally.

##Generate the dashboard## In the directory with the obsinfo file and the npz files in the appropriate directory specified therein (if not the default data directory), you can view the data and generate the dashboard.

To view a dashboard of a source from the command line: on_look.py <SOURCE>

For options, type on_look.py -h

E.g. on_look.py STARLINK11139DTC2_A_C0352 --lo A --dash

If you want to generate and save all of the data in the obsinfo file, type on_gen_dash.py <MJD>, which will write a bash script file (default dash.sh).

Running the 'dash.sh' script won't display any data, but rather save them to png files. You can concatenate them and write them to a pdf file by using a image tool like magick: magick *.png out.pdf