[SensorNetArch] Problem definition

[Philip Levis]

On Jul 19, 2006, at 2:14 PM, Joe Polastre wrote:

>> I think the point you're getting at is right, that because space
>> overlaps and is not transitive, then fair queueing within one
>> airspace might not be fair within another...
>
> Let me think of an exact case where the hidden terminal can exploit
> this later this evening.
>
> Does this mean I get a tasty goodie baked by Phil? :)

Well, the SNA group seemed tepid on my cooking, so I also offered an  
alternative consolation prize of a cake from Massey's...

The intent of the problem had to do with the fact that a recipient  
can ignore the protocol quiet period. I.e., can nodes collude to use  
this to get more time. Since values are monotonically increasing, you  
can't just do a simple ping-pong. E>g., if A sends to B who sends to  
A, A still has to obey its first quiet period and may not have B's  
packet as the first to send.

Phil