Re: [Roll] ETX metric

[Philip Levis <pal at cs.stanford.edu>]

On Oct 12, 2009, at 3:09 PM, Maxence Dalmais wrote:

> I am saying that when using wireless links, a node should be aware  
> that his communication could interfere with other communication.
> Less powered is the transmission signal, less noise is seen by  
> neighbours.

It also means that it's more greatly affected by interference from  
neighbors: there is no free lunch.

> A low powered node will also need less energy to communicate in a  
> less noisy environnement.

In low-power wireless chipsets, CMOS dominates power costs. E.g., a  
chip that draws 60mW transmits at 1mW. Reducing transmit power  
typically does not conserve much energy.

> So, it appears to me logical that when others metrics are egals to  
> choose the less power needed links.

I don't know of any strong experimental evidence supporting this  
conclusion. Can you point me at some?

Phil