[e2e] characterizing video conferencing traffic..

[Spencer Dawkins]

(Spencer_talking_out_his_hat)

It looks like this paper is putting one H.261 frame in each UDP
packet, because I think I can derive the pie chart on packet
size distribution from the repeating pattern of full-sized
I-frames, separated by shorter P-frames and B-frames, modulo
RTCP overhead.

I haven't looked at H.261 in several years, but I'm remembering
that the number, and order, of P-frames and B-frames between
I-frames was implementation-specific (maybe they've picked a
right answer by now - I know MPEG used to be
implementation-specific)

So, I don't think this is a traffic "mix" per se, it's a traffic
flow between any two endpoints using a specific implementation.

(/Spencer_talking_out_his_hat, or at least, I'd like to think
so)

Someone who actually KNOWS will correct me, of course...

--- Prasad Calyam <pcalyam at oar.net> wrote:
> hi,
> I was wondering if anyone knows of any research related to the
> packet
> size breakdown of video conferencing traffic..
> 
> I saw a Cisco whitepaper-
> http://www.cisco.com/warp/public/105/video-qos.html
> which has a pie chart showing the packet size breakdowns but
> does
> not explain how this conclusion was drawn..
> 
> I appreciate if anyone could give me more information..
> 
> Thanks!
> -Prasad
> 
>