Re: [AVTCORE] Framemarking in video packets

[[e-mail]]

"shihang (C)" <[e-mail]> writes:
> Are you suggesting that we modify the RFC 8285 to reserve a global ID
> for framemarking extension? By using the fixed global ID, you remove
> the signaling process when using the framemarking extension, right? 

I don't have a truly solidified proposal.  I suppose we could propose
that, *if the sender was to use framemarking according to the proposal*
it would have to use a particular extension ID.  But that doesn't
suffice for my purpose, which is for a router to be able to (1)
determine which packets are according to the proposal, and then (2) be
able to isolate the framemarking extension easily.  The reason is that
even if we declare a fixed global ID for the extension, that is at most a
16-bit fixed value to be searched for in the packet, which would have a
lot of false positives for the identification step.

The novel part of my proposal is that it supports the router being able
to identify many more layers of the video stream than would be possible
with a single AF-set of DSCP codepoints, and it does so by providing a
way to insert a global, fixed 32-bit value that shows where the
framemarking extension is.

Of course, there may be better methods.

Dale