From: Ken Clements (Ken@Innovation-On-Demand.com)
Date: Mon Jul 09 2001 - 15:50:48 MDT
Brian D Williams wrote:
> >From: Eugene Leitl <Eugene.Leitl@lrz.uni-muenchen.de>
>
> >I understand you concerns about interoperability, but the
> >situation is unfixable. You have to have to use a different
> >transport layer. You can always tunnel IP over it, too. The
> >easiest entry points into geodetic routing is 802.11b grassroot
> >networks and LEO sat clouds, because they don't rely on existing
> >infrastructure.
>
> Okay, so you plan to use encapsulation to accomodate IP.
>
> I still don't see the routing part, you're not using a distance-
> vector (Bellman-Ford) nor a link-state (OSPF or Dijkstra) protocol
> or even a hybrid (EIGRP, BGP), essentially it seems to be nothing
> more than a simple broadcast which would eat up your bandwidth, not
> to mention propagation delays, and what about loops?
>
> Brian
>
Yes, this is a big problem. At the moment I am sitting in the
IEEE802.11 meeting in Portland OR, and we are stuck with some of these.
The trouble is that wireless within 802 has to use the 802 layers above
the MAC. The result is that we cannot properly address wireless routing
here (transport layer).
The infrastructure companies want to use 11b as the last link ONLY.
This means that routing happens on the wires, not over the air. If you
can do routing that way, you cannot beat it. If you need to route in
the air, it impacts things across layers and the general algorithms used
for wires (fibres) do not work well without modifications for the nature
of wireless propagation, connection density and error rates.
-Ken
This archive was generated by hypermail 2.1.5 : Sat Nov 02 2002 - 08:08:36 MST