manet@ietf.org
[Top] [All Lists]

Re: [manet] Metric encoding for OLSRv2

Subject: Re: [manet] Metric encoding for OLSRv2
From: Henning Rogge
Date: Thu, 1 Jul 2010 14:35:21 +0200
On Wed June 30 2010 00:33:45 Teco Boot wrote:
> Your alternative and the "- 15" option do not differ that much.
> So it is acceptable for me.
The "infinite" metric is not that important (it's a nice to have I think), as 
long as we can encode a cost of zero.
 
> Having a coding scheme with a maximum near to 2^24 - 1 would be nice, but a
> 1 to 10^6 ratio would be good enough for almost all scenarios.
value = (a + 2^m)*2^b - 2^m
0 <= value <= 1015792

What would be a good default link value for this encoding ?
I would suggest (a=0,b=15) = 524272 or (a=0,b=14) = 262128.

This would give us lot's of range for metrics better than hopcount and some 
space for links worse than it.

Henning Rogge

-- 
Diplom-Informatiker Henning Rogge , Fraunhofer-Institut für
Kommunikation, Informationsverarbeitung und Ergonomie FKIE
Kommunikationssysteme (KOM)
Neuenahrer Straße 20, 53343 Wachtberg, Germany
Telefon +49 228 9435-961,   Fax +49 228 9435 685
mailto:henning.rogge@xxxxxxxxxxxxxxxxxx http://www.fkie.fraunhofer.de
GPG: E1C6 0914 490B 3909 D944 F80D 4487 C67C 55EC CFE0
_______________________________________________
manet mailing list
manet@xxxxxxxx
https://www.ietf.org/mailman/listinfo/manet
<Prev in Thread] Current Thread [Next in Thread>