[Battlemesh] Dynamic TX power

fboehm fboehm at aon.at
Thu Dec 31 17:08:20 CET 2015


Am 31.12.2015 um 13:34 schrieb Bastian Bittorf:
> * Ferry Huberts <mailings at hupie.com> [31.12.2015 13:23]:
>> To be truely effective this needs to be solved in the hardware, or at
>> least very very close to it.
>
> this is wrong. It can be fully done in software,
> e.g. ath5k/ath9k can set the power per *packet*
> and you can solve this in an opensource-driver (rate-control!)
> better than if the vendor does it in hardware.
>
> bye, bastian
> _______________________________________________
> Battlemesh mailing list
> Battlemesh at ml.ninux.org
> http://ml.ninux.org/mailman/listinfo/battlemesh
>
Just a short addition to full-power beacons:

Not always the best idea because beacons (with low modulation rate) will 
be heard by distant nodes that might not be able to achieve meaningful 
throughput later on. The result could be waste of valuable airtime 
because of a weak link with low modulated data packets and many 
retransmissions.

But you can also leave it to the routing protocol to detect such sketchy 
links.

Franz



More information about the Battlemesh mailing list