[Pafgbt] BYU RT Correlator Data

[Rick Fisher]

Richard,

Thanks very much for the sample correlator data.  I may not get a chance 
to open the package until Monday, but I look forward to digging into the 
details.

Rick

On Wed, 1 May 2013, Richard Black wrote:

> Rick and Anish,
> Here is some sample data from the real-time correlator. I'm sorry it took so
> long to get to you.
> 
> I've also attached a python script that parses and plots the data. You'll
> need the numpy, pylab, h5py, time, sys, math, and poxy python libraries to
> run this code.
> 
> "plot_corr_h5_zk.py [options] <file_name>" will plot the data in a
> user-specified manner.
> 
> OPTIONS
> ------------------
> -a #
> Specify which antenna to plot against (all cross-PSDs or cross-correlation
> functions against #)
> 
> -a #_#
> Specify which two antennas to plot against each other (cross-PSD or
> cross-correlation function between the two #'s)
> 
> -a auto
> Plot PSDs or auto-correlation functions for all antennas
> 
> -p (xx, yy, yx, xy, all)
> Specifies which polarization should be plotted-- this refers only to how the
> data was treated in the correlator. For this sample data, we only use 'xx.'
> 
> -D
> Plots the cross/auto-correlation functions. If this option is omitted, the
> (cross-)PSDs are plotted instead.
> 
> For example, to plot the cross-PSDs against antenna 0, you would use the
> following command:
> "./plot_corr_h5.py -a 0 -p xx corr.2456087.47037.h5c"
> 
> If you wanted to plot the cross-correlation functions against antenna 0, you
> would use the following command:
> "./plot_corr_h5.py -a 0 -p xx -D corr.2456087.47037.h5c"
> 
> 
> If you wanted to plot the cross-PSD between antennas 0 and 1, you would use
> the following command:
> "./plot_corr_h5.py -a 0_1 -p xx corr.2456087.47037.h5c"
> 
> If you wanted to plot the auto-correlation PSDs for all the antennas, you
> would use the following command:
> "./plot_corr_h5.py -a auto -p xx corr.2456087.47037.h5c"
> 
> 
> 
> I hope this makes sense. Let me know if there is any confusion.
> 
> Thanks,
> 
> Richard Black
> 
>