Re: [52N SWE] 3D on SOS

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

Re: [52N SWE] 3D on SOS

Dillabough, Craig
Hi Alessandro,
I am very new to SWE and not an expert by any means (I am still
struggling with the data model), but reading your message I was
wondering if the following work-around might help.

Create a new Phenomenon which is called DEPTH, and then with each
observation record the sensor depth along with the property you are
observing. This would duplicate the depth measurements (since you would
be re-recording it with each observation), but it would certainly make
queries fairly straight forward I would think.

Regards,
Craig



-----Original Message-----
From: [hidden email] [mailto:[hidden email]] On Behalf
Of Alessandro Oggioni
Sent: Monday, August 09, 2010 10:13 AM
To: Sensor Web developer communication
Subject: Re: [52N SWE] 3D on SOS


Dear Christoph,
thanks for your reply and sorry for the delay in answering you.
Unfortunately for my data model that aims to multiply the FOI for each
depth is not suitable to my dataset, in fact only on Lake Maggiore
have the presence of 27 stations (27 FOI) in each data are collected
at depths all times different, the FOI would multiply exponentially.
The model that you have suggests the depth as a feature of FOI, but
for me it would be more appropriate to have the depth as a feature of
the observation. Even with the two other models that suggest the depth
appears to be a description of the Phenomenon or procedures may not be
so easy to make a query where you want all the observations collected
in the data associated with a given depth.
Is there any paper or OGC document you could see to be able to compare
these aspects?
Thanks a lot
Alessandro

2010/8/2 Christoph Stasch <[hidden email]>
>
> Dear Alessandro,
>
>
>
> if the procedure is always the same, I'd model it as observations at
different FOIs. For example, let z be the height at a point (x,y), then
the FOIs would be:

>
>
>
> foi1(x,y,z1)
>
> foi2(x,y,z2)
>
> foi3(x,y,z3)
>
>
>
> Unfortunately, the 52N SOS currently only supports 2-dimensional
geometries for features of interest. If you want to use the
implementation, there would be two workarounds:
>
>
>
> - model a phenomenon for each height and use the metadata section in
the description of the outputs in the SensorML description to describe
the height of each component
>
>
>
> -model a procedure for each height and use the sml:position of the
sensor description to provide the 3-dim position information.

>
>
>
> Cheers,
>
>
>
> Christoph
>
>
>
> Von: [hidden email] [mailto:[hidden email]] Im
Auftrag von Alessandro Oggioni
> Gesendet: Freitag, 30. Juli 2010 12:19
> An: Sensor Web developer communication
> Betreff: [52N SWE] 3D on SOS
>
>
>
> Dear all,
> I would ask to list the issue of the third dimension (depth for
example) in the SOS service. In fact working in the aquatic environment
we have sensors that collect continuous measurements at a station that
has a FOI always identical, with a sensor that detects the same measures
at different depths. So I am to have taken observations with procedures,
FOI and fenomenon identical but with different timestamp, collected at
different depths.
> Recently, we have made many tests on the possibility of using the
function DataArray for insert data, but all the examples found referring
to array of data that were to compose a composite_fenomenon composed of
two different procedures.
> We also searched the OGC we did not find anything useful, maybe we
have not searched well enough? Anyone have any ideas? To provide some
bibliographic reference?
> Many thanks
> Alessandro
>
> _______________________________________________
> SWE mailing list
> [hidden email]
> http://www2.52north.org/mailman/listinfo/swe
>
_______________________________________________
SWE mailing list
[hidden email]
http://www2.52north.org/mailman/listinfo/swe
_______________________________________________
SWE mailing list
[hidden email]
http://www2.52north.org/mailman/listinfo/swe
Reply | Threaded
Open this post in threaded view
|

Re: [52N SWE] 3D on SOS

Alessandro Oggioni
Thank Craig, I try to use this data model. Alessandro

2010/8/13 Dillabough, Craig <[hidden email]>:

> Hi Alessandro,
> I am very new to SWE and not an expert by any means (I am still
> struggling with the data model), but reading your message I was
> wondering if the following work-around might help.
>
> Create a new Phenomenon which is called DEPTH, and then with each
> observation record the sensor depth along with the property you are
> observing. This would duplicate the depth measurements (since you would
> be re-recording it with each observation), but it would certainly make
> queries fairly straight forward I would think.
>
> Regards,
> Craig
>
>
>
> -----Original Message-----
> From: [hidden email] [mailto:[hidden email]] On Behalf
> Of Alessandro Oggioni
> Sent: Monday, August 09, 2010 10:13 AM
> To: Sensor Web developer communication
> Subject: Re: [52N SWE] 3D on SOS
>
>
> Dear Christoph,
> thanks for your reply and sorry for the delay in answering you.
> Unfortunately for my data model that aims to multiply the FOI for each
> depth is not suitable to my dataset, in fact only on Lake Maggiore
> have the presence of 27 stations (27 FOI) in each data are collected
> at depths all times different, the FOI would multiply exponentially.
> The model that you have suggests the depth as a feature of FOI, but
> for me it would be more appropriate to have the depth as a feature of
> the observation. Even with the two other models that suggest the depth
> appears to be a description of the Phenomenon or procedures may not be
> so easy to make a query where you want all the observations collected
> in the data associated with a given depth.
> Is there any paper or OGC document you could see to be able to compare
> these aspects?
> Thanks a lot
> Alessandro
>
> 2010/8/2 Christoph Stasch <[hidden email]>
>>
>> Dear Alessandro,
>>
>>
>>
>> if the procedure is always the same, I'd model it as observations at
> different FOIs. For example, let z be the height at a point (x,y), then
> the FOIs would be:
>>
>>
>>
>> foi1(x,y,z1)
>>
>> foi2(x,y,z2)
>>
>> foi3(x,y,z3)
>>
>>
>>
>> Unfortunately, the 52N SOS currently only supports 2-dimensional
> geometries for features of interest. If you want to use the
> implementation, there would be two workarounds:
>>
>>
>>
>> - model a phenomenon for each height and use the metadata section in
> the description of the outputs in the SensorML description to describe
> the height of each component
>>
>>
>>
>> -model a procedure for each height and use the sml:position of the
> sensor description to provide the 3-dim position information.
>>
>>
>>
>> Cheers,
>>
>>
>>
>> Christoph
>>
>>
>>
>> Von: [hidden email] [mailto:[hidden email]] Im
> Auftrag von Alessandro Oggioni
>> Gesendet: Freitag, 30. Juli 2010 12:19
>> An: Sensor Web developer communication
>> Betreff: [52N SWE] 3D on SOS
>>
>>
>>
>> Dear all,
>> I would ask to list the issue of the third dimension (depth for
> example) in the SOS service. In fact working in the aquatic environment
> we have sensors that collect continuous measurements at a station that
> has a FOI always identical, with a sensor that detects the same measures
> at different depths. So I am to have taken observations with procedures,
> FOI and fenomenon identical but with different timestamp, collected at
> different depths.
>> Recently, we have made many tests on the possibility of using the
> function DataArray for insert data, but all the examples found referring
> to array of data that were to compose a composite_fenomenon composed of
> two different procedures.
>> We also searched the OGC we did not find anything useful, maybe we
> have not searched well enough? Anyone have any ideas? To provide some
> bibliographic reference?
>> Many thanks
>> Alessandro
>>
>> _______________________________________________
>> SWE mailing list
>> [hidden email]
>> http://www2.52north.org/mailman/listinfo/swe
>>
> _______________________________________________
> SWE mailing list
> [hidden email]
> http://www2.52north.org/mailman/listinfo/swe
> _______________________________________________
> SWE mailing list
> [hidden email]
> http://www2.52north.org/mailman/listinfo/swe
>
_______________________________________________
SWE mailing list
[hidden email]
http://www2.52north.org/mailman/listinfo/swe