Re: [codec] WG Review: Internet Wideband Audio Codec (codec)

[Steve Underwood <[e-mail]>]

On 01/11/2010 10:10 PM, Herve Taddei wrote:
> It did not seem to be a major problem when for example IETF issued RFC 3592
> (iLBC). For this codec the PLC is basically a non normative link on the
> ITU-T G.711 PLC.
>
> Besides, I don't think you would have any trouble to propose at ITU-T some
> new appendices to G.711 and G.722 that could fit your goals. An appendix is
> non normative (a bit like the informative reference to G.711 PLC in iLBC).
> By the way, if I am not wrong, some basic ITU-T G.722 PLCs are RF.
>
> Not considering G.711 and G.722 because they do not have RF appendices does
> not sound to be optimal in term of interoperability, better to work on
> making RF some new appendices.
>    
PLC only affects the receiving end, so you can put anything you like in 
there, really. You can try to implement something non-patented, even if 
the one is the specs is patented. The VAD/DTX/CNG Appendix to G.711 is 
the one I was thinking of. It basically says "use the VAD/DTX/CNG from 
G.729 Annex B" and at least one of the patents you need to licence for 
G.729 relates to that Annex. VAD/DTX/CNG needs to be a properly defined 
part of the standard, because the two ends need to cooperate for that. 
The recent appendices to G.722 seem to have been specifically chosen to 
be encumbered.

Although RFC3591 (RFC3592 is about SDH) mentions the G.711 PLC, that 
isn't the PLC which is normally used with iLBC.

In the end G.711  and G.722 just aren't that interesting. G.711 was 
chosen to be the best they could do with 1957 technology. G.722 was 
designed with the goal of working in a standard 64kbps channel, and 
running on a $1.5 custom chip in 1986. These aren't particularly 
meaningful goals these days.

Steve