Forum: GNU Radio pdu and stream tagging in 3.6.3

I'm playing around with the new 
gr_pdu_to_tagged_stream/gr_tagged_stream_to_pdu blocks. Is there 
something that can translate between the new tag format and the standard 
tx_sob (start of burst)/tx_eob (end of burst)/tx_time tags?

I guess it's possible to write a packet scheduler if you fix a timestamp 
in the PMT metadata dictionary? I think this would involve writing a 
message source block that could attach PMT pairs containing tx_sob/PMT_T 
and tx_time/tuple(uint64_t secs, double frac_secs). A downstream block 
could read the pdu_length tag and put a tx_eob/PMT_T stream tag at the 
appropriate place. Does this sound like a reasonable approach?

Sean

On Fri, Feb 08, 2013 at 07:14:48PM +0000, Nowlan, Sean wrote:
> I'm playing around with the new gr_pdu_to_tagged_stream/gr_tagged_stream_to_pdu
> blocks. Is there something that can translate between the new tag format and
> the standard tx_sob (start of burst)/tx_eob (end of burst)/tx_time tags?

Hi Sean,

there's only one type of tags--I'm not sure what you're asking here.

> I guess it's possible to write a packet scheduler if you fix a timestamp in the
> PMT metadata dictionary? I think this would involve writing a message source
> block that could attach PMT pairs containing tx_sob/PMT_T and tx_time/tuple
> (uint64_t secs, double frac_secs). A downstream block could read the pdu_length
> tag and put a tx_eob/PMT_T stream tag at the appropriate place. Does this sound
> like a reasonable approach?

That sounds exactly like something I discussed with Johnathan recently.
Not everything you've described is there in GNU Radio, but it sounds
reasonable.

MB
--
Karlsruhe Institute of Technology (KIT)
Communications Engineering Lab (CEL)

Dipl.-Ing. Martin Braun
Research Associate

Kaiserstraße 12
Building 05.01
76131 Karlsruhe

Phone: +49 721 608-43790
Fax: +49 721 608-46071
www.cel.kit.edu

KIT -- University of the State of Baden-Württemberg and
National Laboratory of the Helmholtz Association

>On Fri, Feb 08, 2013 at 07:14:48PM +0000, Nowlan, Sean wrote:
>> I'm playing around with the new gr_pdu_to_tagged_stream/gr_tagged_stream_to_pdu
>> blocks. Is there something that can translate between the new tag format and
>> the standard tx_sob (start of burst)/tx_eob (end of burst)/tx_time tags?
>
>Hi Sean,
>
>there's only one type of tags--I'm not sure what you're asking here.

What I meant was that gr_pdu_to_tagged_stream doesn't tag with 
tx_sob/tx_time/tx_eob, and so the format has to be "translated" by a 
downstream block's work(...) function from PDU tags to burst tags.

>> I guess it's possible to write a packet scheduler if you fix a timestamp in the
>> PMT metadata dictionary? I think this would involve writing a message source
>> block that could attach PMT pairs containing tx_sob/PMT_T and tx_time/tuple
>> (uint64_t secs, double frac_secs). A downstream block could read the pdu_length
>> tag and put a tx_eob/PMT_T stream tag at the appropriate place. Does this sound
>> like a reasonable approach?
>
>That sounds exactly like something I discussed with Johnathan recently.
>Not everything you've described is there in GNU Radio, but it sounds
>reasonable.

I think all the elements are there in v3.6.3 to write such a block. But 
I'm confusing myself. In gnuradio-core/src/lib/io, 
gr_message_burst_source sort of implements this functionality (tx_sob & 
tx_eob, but no tx_time). It seems the PDU-to-tagged-stream concept is a 
parallel solution that adds a "length" tag and whatever is in the 
metadata dictionary, but doesn't use tx_sob/tx_eob tags explicitly. Am I 
reading this correctly?

-sean

On Mon, Feb 11, 2013 at 02:52:58PM +0000, Nowlan, Sean wrote:
> I think all the elements are there in v3.6.3 to write such a block. But I'm 
confusing myself. In gnuradio-core/src/lib/io, gr_message_burst_source sort of 
implements this functionality (tx_sob & tx_eob, but no tx_time). It seems the 
PDU-to-tagged-stream concept is a parallel solution that adds a "length" tag and 
whatever is in the metadata dictionary, but doesn't use tx_sob/tx_eob tags 
explicitly. Am I reading this correctly?

Yes, you are. The thing about sob/eob tags is, they're UHD-specific, so
the PDU-to-stream blocks probably aren't the right place to put them.

M

--
Karlsruhe Institute of Technology (KIT)
Communications Engineering Lab (CEL)

Dipl.-Ing. Martin Braun
Research Associate

Kaiserstraße 12
Building 05.01
76131 Karlsruhe

Phone: +49 721 608-43790
Fax: +49 721 608-46071
www.cel.kit.edu

KIT -- University of the State of Baden-Württemberg and
National Laboratory of the Helmholtz Association