[Battlemesh] Testbed - how to organize this year?

Juliusz Chroboczek jch at pps.jussieu.fr
Sat Mar 25 00:01:36 CET 2017

> my personal list of burning research questions ;)

I find your list intriguing, and would like to subscribe to your newsletter.

> - what's currently the optimal throughput metric to use?

Interesting question.  There's a tradeoff here between making the right
routing decisions and implementation complexity, so I'd reformulate it as
"at which point does adding complexity to metric computation yield
diminishing returns".  There's a serious lack of experimental data in this

> - does airtime fairness scheduling make sense in mesh?

Right.  I think we're all convinced that it does, but how to design an
experiment that proves it?

> - is it optimal to use 802.11s(norouting) instead of IBSS?

I plead incompetence.

> - what's the potential of per packet power control?

Very interesting, but difficult to measure.

Here are my additions to your list, obviously biased towards my research:

- is radio-diversity routing (bmx7, Babel-Z) worth it, and in what topologies?

- is RTT-based routing (Babel-RTT) useful in meshes?  (It's known to be
  useful in overlay networks, which is what it has been designed for.)

- is source-specific routing useful for the mesh community?

- can we use ToS for something?

- is there any sense in developing autoconfiguration protocols, or is
  DHCPv6-PD good enough?

-- Juliusz

More information about the Battlemesh mailing list