[Nfd-dev] Techniques for prioritizing audio in NdnCon

[Junxiao Shi]

Hi Jeff

I have read the paper "Self-prioritization of audio and video traffic".
The paper aims at prioritizing audio/video traffic over download traffic on
the user access link. It detects audio/video streams by looking at their
traffic characteristics (constant send rate), and sends those packets
before others.
Mainly, it is a congestion control problem.

Prioritizing audio over video, or prioritizing base layers over enhancement
layers, is a different problem from above.
It's hard for network to determine whether a packet is audio / base layer,
without being told so. I think it's not too bad to peek the Name in an
application specific strategy (#2000 would give us composable strategy
which could be customized for every application namespace).
The strategy for these applications can then prioritize audio / base layer
packets when there's a congestion.

Yours, Junxiao

On Mar 28, 2015 10:06 PM, "Burke, Jeff" <jburke at remap.ucla.edu> wrote:
>
> Hi Junxiao,
>
> I will do some digging, but this is interesting:
>
> Bonald, Thomas, Luca Muscariello, and Norberto Ostallo.
"Self-prioritization of audio and video traffic." Communications (ICC),
2011 IEEE International Conference on. IEEE, 2011.
> http://perso.telecom-paristech.fr/~bonald/Publications_files/BMO2011.pdf
> (Instead of identifying packets as part of a "flow", one would use
namespace or priority flag hints.)
>
> There are follow up papers to this one, too. Interesting to note that
Muscariello (and a later co-author, Carofiglio) are now doing ICN research.
>
> (There are RTCP receiver reports in RTP streaming, too. )
>
> Jeff
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.lists.cs.ucla.edu/pipermail/nfd-dev/attachments/20150331/1f37b450/attachment.html>