[Radioig] voltage UCD?

Tue Apr 4 11:13:11 CEST 2023

Hi again,

Meanwhile, it has turned out that stat.fourier is not a match, because ...

@Mark:
... it's channelized, dual-polarisation, beamformed voltages. So, the 
data product is primarily a time series of complex voltages (amplitude 
and phase). Mentioning PSRFITS was somewhat distracting. Upon further 
investigation it turns out that the format was an earlier design 
assumption. As is the format does not support such voltages.

@All:
Clearly, the design of the voltage recording data product needs to 
settle, before we can make the case for a new UCD.

Apparently, I've stumbled across a new use case. So, thanks all for 
helping to clarify.

Best
Markus

On 31/3/2023 19:40, Mark Kettenis wrote:
>> From: Markus Dolensky <markus.dolensky at uwa.edu.au>
>> Date: Fri, 31 Mar 2023 08:16:47 +0000
> Hi Markus Do & others,
>
>> Hi Markus De., Francois, Mireille,
>>
>> @Markus De:
>> We are talking about raw (calib_level = 0), complex voltage data (o_ucd =
>> stat.fourier?) from a future radio interferometer that records them in near
>> realtime (dataproduct_type = timeseries).
> Correct me if I'm wrong, but I think what you're really talking about
> is "I/Q data" that describes the in-phase (I) and quadrature (Q)
> components of a (downmixed) signal.  The I and Q components are often
> combined in a single complex number but that really is just a
> mathematical convenience.
>
> Note that reconstruction of the original signal (which is presumably
> what is needed/wanted in the subsequent analysis) you need to know the
> frequency of the mixer signal (LO) that was used to construct the I
> and Q components.
>
>> So yes, stat.fourier is a possible candidate. I will try to source an
>> opinion from one of the Pulsar astronomers designing the data product.
> The existence of stat.fourier.amplitude suggests that stat.fourier
> would indeed be complex number; but the I/Q representation isn't
> really a fourier transform of the original signal...
>
> Is a new UCD "atom" that indicates I/Q data appropriate?  With the
> understanding that this would be a complex number?  Not sure if there
> is a need to have a UCD that represents the indivicual I and Q
> components.
>
>> I have taken note of the UCD maintenance process.
>>
>> @Francois
>> re ObsCore extension and ADU:
>> Likewise, I'm following up with the pulsar team on the nature of the voltage
>> data. Timezone difference suggests that won't happen before next week.
>>
>> @Mireille
>> Thanks a lot for jump-starting the RFM.
>>
>> @All
>> Depending on said internal clarifications, we may or may not propose
>> something new. Given that the respective data product does not yet exist,
>> there is no particular urgency and no shortage of uncertainty.
>>
>> For now, thanks for the great considerate responses!
>> Markus
>>
>> Markus Dolensky
>> Technical Leader
>> International Centre for Radio Astronomy Research
>> The University of Western Australia | M468, 35 Stirling Highway | Perth WA
>> 6009
>> T +61 8 6488 5072
>>
>> On 30/3/2023 22:33, Mireille LOUYS wrote:
>>
>>   Hi Markus , hi radio astronomers,
>>
>>   I think we can discuss the definition of a new term, if the requirement
>>   is to label a measurement corresponding to some voltage.
>>
>>   The current procedure for UCD addition is to prepare a short document to
>>   explain the context for adding a new term.
>>   This can be done in a short file called VEP-UCD.
>>   I have prepared such a file , as included here below.
>>
>>   The first step is to discuss the detailed information of the VEP, and
>>   the group can finalise the requirement for this UCD :
>>
>>   here is my proposal , to be completed by your inputs as
>>   radioastronomers.
>>
>>   ----  starts of: VEP-UCD-012_voltage------
>>   Vocabulary extension proposal for UCDs /
>>   Vocabulary: UCDlist 1.6
>>   Author: mireille louys for radio interest group
>>   Date: 2023-03-29
>>
>>   New Term: phys.voltage
>>   Action: Addition
>>   Label: Voltage (measured by an instrument)
>>   Prefix: Q
>>   Description:  Voltage quantity
>>    
>>   Relationships: can be combined with instr.param for observing
>>   configuration
>>   Rationale:
>>   Radio data in raw format can measure a voltage quantity as signal,
>>   before the data is calibrated.
>>   'phys.voltage' can be used as well to identify a parameter of the
>>   observation configuration used with an instrument.
>>
>>   Discussion:
>>   To be completed by radio and semantics WG remarks
>>   Pros and cons ...
>>   Other existing ucd terms that can be used instead...
>>
>>   Response: (To be completed by UCD board after discussion)
>>
>>   ----  end of VEP-UCD-012_voltage------
>>
>>   When we agree, this can be added on the RequestForModificationPage here:
>>   https://aus01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwiki.ivoa.net%2Ftwiki%2Fbin%2Fview%2FIVOA%2FUCDList_1-6_RFM&data=05%7C01%7Cmarkus.dolensky%40uwa.edu.au%7Cd2c757102bb543bc8b5f08db31dcc714%7C05894af0cb2846d8871674cdb46e2226%7C0%7C0%7C638158597453614121%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=HYgQCepCGS%2F2RDAjLR%2BDhRgLuKyW4MN3pn9dxLhUGu0%3D&reserved=0
>>
>>   many thanks for your inputs to finalise this first step.
>>
>>   cheers , Mireille
>>
>>   Le 30/03/2023 à 13:21, Markus Demleitner a écrit :
>>
>>   Hi Markus,
>>
>> On Tue, Mar 28, 2023 at 06:30:55AM +0000, Markus Dolensky wrote:
>>
>>   Can somebody suggest a UCD for complex voltage data, please?
>>
>> Context is the Obscore o_ucd for PSRFITS data.
>>
>> Hm... it would seem nobody has an immediate solution, and reviewing
>> the current UCD list make me suspect there's nothing remotely
>> matching -- although I have to admit that I've no clue of PSRFITS...
>> Does what you'd like to annotate have a complex part because of phase
>> information in a fourier transform?  If so, perhaps stat.fourier
>> could remotely work?
>>
>> With that caveat, I suppose the Right Thing™ to do would be to
>> propose a new UCD.  This is less of a hassle than it might sound like
>> -- see
>> https://aus01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fivoa.net%2Fdocuments%2FUCDlistMaintenance%2F20191007%2FREC-UCDlistMaintenance-2.0-20191007.pdf&data=05%7C01%7Cmarkus.dolensky%40uwa.edu.au%7Cd2c757102bb543bc8b5f08db31dcc714%7C05894af0cb2846d8871674cdb46e2226%7C0%7C0%7C638158597453614121%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=NBEV%2Bb3KdFXuXTAEsX4%2F0tvOzPDCrPz7JgbVh07kW1A%3D&reserved=0.
>>
>> However, I suppose it would be a good idea to discuss the RFM here
>> first before handing it over to Semantics.
>>
>>           -- Markus
>>
>> -- 
>> --
>> Mireille Louys,  MCF (Associate Professor)
>> Centre de données CDS		IPSEO, Images, Laboratoire Icube
>> Observatoire de Strasbourg	Telecom Physique Strasbourg
>> 11 rue de l'Université		300, Bd Sebastien Brandt CS 10413
>> F- 67000-STRASBOURG		F-67412 ILLKIRCH Cedex
>> Tel: +33 3 68 85 24 34
>