[CUWiN-Dev] making the visualization less "jiggly"
David Young
dyoung at pobox.com
Sun Aug 21 18:38:17 CDT 2005
My friend Dave Healy and I came up with three recommendations for making
the route visualization less "jiggly" in a brainstorming session:
1 Set the natural length of springs according to each neighbor's
minimum ETX over a very long period of time. It is
reasonable to expect that the minimum ETX is stable, since
it is (very) roughly proportional to S/N, which is (very)
roughly proportional to distance in the rooftop environment.
We know this much for certain: rooftop nodes are not getting
any closer to each other [1], so the ETX on any link ought to
have a more-or-less stable floor.
2 Change the natural length of the springs gradually, to avoid
injecting energy into the system. (I believe Bill Comisky's
patch does that.)
3 Find a polynomial function that relates ETX to distance.
Use that function to set the natural length of springs based
on ETX. You can do this by fitting a curve to a graph of ETX
versus distance, where the distances are known (say that the
nodes latitude and longitude are known).
Dave
[1] Apart from Brooklyn, the universe is expanding. QED.
--
David Young OJC Technologies
dyoung at ojctech.com Urbana, IL * (217) 278-3933
More information about the CU-Wireless-Dev
mailing list