H.248 Annex M IPR

[Taylor, Tom-PT [NORSE:B901:EXCH]]

You are right. This looks like a mistake in Annex F/H.323. It should say "#
of frames" in the diagram and in the text below. The current text conflicts
with the rest of H.323 and H.225.0 and H.245.

Paul, another one for the implementor's guide?

-----Original Message-----
From: stefan Jung [mailto:stefan.jung at dsp.se]
Sent: Monday, January 08, 2001 1:16 AM
To: H.323 Implementors
Subject: OpenLogicalChannel parameter


Hi All,
I need help with the intepretation of a parameter in the Open Logical
Channel structure.
The problem I have is when i read the AnnexF specification.  In the AnnexF
chapter 10.1.1.1 there is a layout of the logical channel structure that
shall be used by the SET device for the codecs G.711 and G.729. In the
structure there is a parameter denoted "# samples".
The definition is :  For codecs G.711 A law 64 kbit/s, G.711 56 kbit/s u law
64 kbit/s, ulaw 56 kbit/s and G.729 Annex A this component contains the
number of samples -1 per audio packet as defined in H.245.

The problem with this definition is that it is not specified in that way in
H.245. In H.245 it is stated that all saple based codecs are considered as
frame based with a frame length of 1 ms.

So how shall the "#samples " be interpreted ?

If it is actually samples then the maximum value seems a bit small.
The parameter is 8 bits which gives a maximum of 256 samples. This
corresponds to 32 ms in G.711.

A more reasonable interpretation would be to consider the parameter as
"frames" instead. This would then give a more flexible range of the packet
size and it would fit better with the definitions of H.245.

Any input to clarify would be appreciated.

Best regards
/Stefan Jung


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.packetizer.com/pipermail/sg16-avd/attachments/20010108/44fbd053/attachment-0001.htm>