[e2e] Packet Corruption in a Wireless Medium!

[Spencer Dawkins]

Dear Arjuna,

Dave did a nice job of replying in a separate e-mail, so I'll
just send the "other" reply, asking "so, why do you want to
know?"

My own interest in ELN was for transport feedback - "since you
know it's not congestion loss, retransmit but don't slow down".
This worked nicely in isolation, and ran into a wall when you
were experiencing both congestion and transmission error loss
simultaneously - we couldn't come up with a decent mechanism
that says "take this loss seriously, not this next loss, but the
following loss is also serious", with RTT-scale feedback loops.

Phil Karn (among others) keeps reminding me that "transmission
errors" in a CDMA environment can ALSO reflect "congestion" - at
a cell level, when two senders are colliding with each other -
so not slowing either one down is not a good plan.

We wrote this up in http://www.ietf.org/rfc/rfc3155.txt,
following up on our wishful thinking in
ftp://ftp.rfc-editor.org/in-notes/rfc2757.txt.

Dave also referred to the possibility of protecting header
fields (so you can send ELNs to the right place based on source
addresses in the damaged packets) - UDPlite is trying to provide
some help here, but it's not clear to me how much help it can
provide.
(http://www.ietf.org/internet-drafts/draft-ietf-tsvwg-udp-lite-01.txt)

Hope this helps,

Spencer

--- arjuna at dcs.kcl.ac.uk wrote:
> Dear All,
>  I would like to know :
> 
> 1)Can the headers of ALL packets that are corrupted while
> traversing a
> wireless link be retrieved by the MAC layer at the receiver
> (as proposed
> by the Header Checksum Option), so that the ELN mechanism can
> detect
> packet corruption?
> 2)If no, what percentage of them can be retrieved
> successfully? Are there
> any papers that address this issue?
> Regards,
> Arjuna
> 
>