[AVT] Re: Proposed change to Enhanced CRTP

[Stephen Casner <casner@acm.org>]

The change to the Enhanced CRTP draft proposed in the message enclosed
below has received no objections on the AVT mailing list.  The change
was also discussed in the AVT session at the IETF meeting just
concluded in Vienna, where comments in favor were received.
Therefore, we consider this change request to be appropriately
reviewed by the working group and accepted.

                                                        -- Steve

On Fri, 11 Jul 2003, Tmima Koren wrote:

> We'd like to propose a change to the Enhanced CRTP draft,
> draft-ietf-avt-crtp-enhance-07.txt
> http://www.ietf.org/internet-drafts/draft-ietf-avt-crtp-enhance-07.txt
>
> One of the enhancements is to add a Headers Checksum (HDRCKSUM) when UDP
> checksum is not present (see section 2.2).
> The UDP checksum is used by the decompressor to periodically validate the
> reconstructed packets. When not present,
> a HDRCKSUM may be inserted by the compressor replacing the missing UDP
> checksum. The decompressor removes the HDRCKSUM after validating the
> reconstructed packet.
>
> Currently the HDRCKSUM covers the pseudo-IP header, the UDP header and the
> fixed part of the RTP header (first 12 bytes).
> Those are the only parts of the reconstructed packet that can be validated
> by the HDRCKSUM.
> The CSRC list is not included in the HDRCKSUM.
>
> We propose to include the CSRC list in the HDRCKSUM since the CSRC list is
> part of the header compression context.
>
> The CSRC list is stored at the decompressor context and inserted to the
> restored packets. If the HDRCKSUM does not cover the CSRC list and the CSRC
> list at the decompressor is errored, a bad CSRC list will be restored into
> multiple packets until the CSRC
> list changes and is resent by the compressor as an update.
> If the HDRCKSUM covers the CSRC list, an error in the CSRC list will be
> caught early and corrected.
>
> When there is a change in the CSRC list, the CSRC list is sent as an update
> to the context.
> The CSRC list should be treated in the same way as the other context
> updating changes and be included in the HDRCKSUM.
>
> On the other hand, adding the CSRC list to the HDRCKSUM requires more
> processing calculating the HDRCKSUM. Also if
> the accuracy of the CSRC list is not crucial and the CSRC list is included
> in the HDRCKSUM, then an error in the CSRC list
> will result in dropping a few packets while the contexts are getting
> synchronized.
>
>
> The proposed change to the text is as follows:
>
> Old:
>
>     The HDRCKSUM is calculated in the same way as a UDP checksum except
>     that it does not cover all of the UDP data.  That is, the HDRCKSUM is
>     the 16-bit one's complement of the one's complement sum of the
>     pseudo-IP header (as defined for UDP), the UDP header, and the first
>     12 bytes of the UDP data which are assumed to hold the fixed part of
>     an RTP header.  The extended part of the RTP header and the RTP data
>     will not be included in the HDRCKSUM.
>
> New:
>
>     The HDRCKSUM is calculated in the same way as a UDP checksum except
>     that it does not cover all of the UDP data.  That is, the HDRCKSUM is
>     the 16-bit one's complement of the one's complement sum of the
>     pseudo-IP header (as defined for UDP), the UDP header,     the first
>                                                            ^^^^
>     12 bytes of the UDP data which are assumed to hold the fixed part of
>     an RTP header, and the CSRC list.
>                  ^^^^^^^^^^^^^^^^^^^
>     The extended part of the RTP header beyond the CSRC list and the RTP data
>                                         ^^^^^^^^^^^^^^^^^^^^
>     will not be included in the HDRCKSUM.
>


_______________________________________________
Audio/Video Transport Working Group
avt@ietf.org
https://www1.ietf.org/mailman/listinfo/avt