Hi Tengfei,

Thanks for these questions.

1. We're not protecting against selective jamming on the minimal cell
indeed. This is a tricky problem, because beacons must include enough
information for nodes to find out when to tx/rx for the join procedure, and
a not-joined-yet node can do it then so can an attacker. Our threat model
is where the attacker is interested in degrading the performance for a
given [set of] node[s] in a running network.

2. This is a great point. We currently know of two approaches, both of
which involve a slight modification of MSF. (a) make MSF use the minimal
cell instead of autonomous for joining or (b) move autonomous cells to a
different slotframe than dedicated cells (e.g. to autonomous to slotframe
0, or dedicated to a new slotframe 2).

Best,
Simon

On Thu, Apr 4, 2019 at 6:47 PM Tengfei Chang <[email protected]>
wrote:

> Hello Authors,
>
> I just reviewed the draft. It reads pretty good for me! I only found a
> tiny typo error in Eq.1 where the 'c' is not defined actually, I believe
> you mean 'chOff'.
>
> Besides, I have two questions referring the usage of minimal cell.
>
> 1. What if the selective jamming applied on the minimal cell? Do you
> consider to resolve this case?
>
> 2. The joining traffic is going on  minimal cell in slotframe 0. Do you
> plan to use some strategy to regulate the traffic on minimal cell? I am
> asking this because this is different from what MSF is doing, which send
> joining packet over autonomous cell.
>
> Tengfei
>
> --
> Chang Tengfei,
> Pre-Postdoctoral Research Engineer, Inria
> _______________________________________________
> 6tisch mailing list
> [email protected]
> https://www.ietf.org/mailman/listinfo/6tisch
>
_______________________________________________
6tisch mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/6tisch

Reply via email to