[Tinyos-devel] retransmissions in networking protocols

[Philip Levis]

On Oct 5, 2007, at 12:56 AM, Andreas Koepke wrote:

> But it uses the AMPacket.source -- but I'd like to use the  
> originator of
> the message (or network source). This way one can filter duplicates
> travelling over different paths.

I'll return to my age-old refrain: does this happen a lot? That is,  
is it a hypothetical problem or a real problem? Given that MLQI  
changes routes only on receiving beacons, and nodes send beacons  
every 30s, the routing topology tends to be very stable. I might  
believe that in the first few minutes this could happen a bit, but  
not once the protocol runs for a while. If you can show a  
quantification of the benefit source-based suppression gives over  
link-based suppression and it's reasonable, then that's great, and  
I'd say go for it. But there are so many cases in TinyOS's history  
where developers have tried to solve hypothetical problems in the  
code base and have instead merely introduced new ones (or made things  
worse) that churn for churn's sake is problematic, and one of the  
reasons a lot of people had trouble with TinyOS 1.x. Changes to well- 
tested and heavily-used protocols need a quantitative basis.

Om -- we have some detailed MLQI traces, right? Did we log duplicate  
suppression in them? Or, at the very least, we can definitely measure  
the duplicate delivery rate at the root, right? It would be good to  
know how bad the problem is.

Phil