[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: [AVT] RE: [dccp] Would a DCCP flow need to send 2x its preferred rate?



Sorry for the previous message. I was intending to make comments on the
following posting regarding "TFRC allowed rate".

--------------------------------------------------------------
...... When you're transmitting at
less than the TFRC allowed rate, your allowed rate could be twice what
you're transmitting.  If a loss event happens, your allowed rate is cut
back, but since it was twice what you were actually transmitting, no
effect.
This gives you the effective "protection" of transmitting at twice your
necessary rate, without actually doing it.

.......
----------------------------------------------------

Since TRFC allowed rate (R0) is estimated from observed loss rate (L) by
the DCCP flow, and the two are positively correlated, R0 may be
interpreted and used in different ways:

- If the DCCP flow can tolerate a loss rate > L, it can consider to
increase its sending rate. As long as its sending rate R < R0, it will
not be considered to be unfair to other TCP flows. It does not
necessarily means that when R < R0, it is always good to increase R.
Doing that could result higher loss rate when the end-to-end path is
lightly multiplexed with TCP flows.

- If the estimated R0 is reduced, even that it is still higher than R,
keeping the current sending rate for the DCCP flow may imply the flow
will experience higher loss rate.

The point: the adaptation of DCCP flow rate may need to combine the
considerations of both R0 and L, depending on the codec characteristics.
When the decoder can only perform error concealment under certain
pre-designed loss rate, then the L should dominate the adaptation,
assuming scalable media content is used, which may imply that R0 could
become irrelevant under some circumstances.  

Qiong Li



_______________________________________________
dccp IETF mailing list: dccp@ietf.org
list info:  https://www1.ietf.org/mailman/listinfo/dccp
wg charter: http://www.ietf.org/html.charters/dccp-charter.html