[fitsbits] ATNF/NRAO codes for polarization

[Tom Kuiper]

Thanks for the quick responses.  I'll try to answer the points raised.

Peter, I agree that definitions would be needed.  I don't have strong 
feelings about what they are, since you can usually get one from 
another, except possibly (N-1) which is usually close enough to N not to 
matter.

The reason for the higher moments (responding also to Thierry) is that 
I'm now writing code to record data from electrostatic discharges (on 
Mars, for now, others later?) which have non-Gaussian signal 
statistics.  So far, kurtosis only tells us that the signal is 
non-Gaussian.  I'm hoping that with higher order moments, we may be able 
to figure out what the signal statistics are, approximately.  That's 
just a hope so far and our firmware currently doesn't computer higher 
orders but that's the goal.

Brian, a separate axis seems like a good idea that I should have 
considered.  Currently we only compute RR* and LL* but cross-products 
should be considered too.  Unless I come back to this forum, I'll do 
that instead and leave the polarization codes alone.

Best regard,

Tom

On 12/16/2012 01:56 PM, Tom Kuiper wrote:
> I would like to extend these codes as follows to handle signal statistics:
>
>    Number Code      Meaning
> 11     mean (should be zero for Gaussian noise)
> 12     variance (proportional to power)
> 13     skewness (should be zero for Gaussian noise)
> 14     kurtosis (should be 3 for Gaussian noise)
> 1x     x = 5,7,9 (should be zero for Gaussian noise)
> 1y     y = 6,8 (should be 15, 105 for Gaussian noise)
>
> Does anyone anticipate a problem with that?  Of course, some analysis
> software might choke on this instead of gracefully ignoring unknown codes.
>
> Best regards
>
> Tom
>
> _______________________________________________
> fitsbits mailing list
> fitsbits at listmgr.cv.nrao.edu
> http://listmgr.cv.nrao.edu/mailman/listinfo/fitsbits
>