[Olsr-users] Topology Set Fields

[Henning Rogge]

Am Thursday 22 January 2009 10:34:29 schrieb Bhat, Ishwara:
> Henning,
>
>
> 1. So LQ and NLQ are really for forward and reverse links between the two
> nodes?
OLSR use hello packages to measure the link quality to it's neighbors. By 
detecting lost hellos it calculates the LQ value, which indicates how good YOU 
can hear your neighbor. Your neighbor is doing the same (so it calculates how 
good YOUR NEIGHBOR can hear you). The result is transmitted with the hello 
packages to the other side, so both sides have both values.

>
> 2. What does the value in link quality signify ? Is there any relation
> between LQ or NLQ and values like RSSI or Received Channel Power Indicator
> ?
At the moment OLSR is unable to get any layer 2 data... I'm working on it at 
the moment because we want to experiment with layer 2 metrics.

> 3. If the Wi-fi card reads certain dBm value when it receives a packet from
> the neighbor, how is this translated to the LQ value ?
It's not.
We are using ETX at the moment, which is a pure layer 3 metric.

> 4. ETX: Expected Tx Count .. I understood how this is calculated. What is
> its significance? Does it tell us how many retires are required? Or it is a
> normalized value indicating the path cost ?
ETX is a measure for the cost of the link. A 802.11 unicast transmission is 
only successful if you can transmit your data and you get the ACK. So the cost 
should include both directions.

> 5. In topology set, do we indicate hop count anywhere ?
No, hopcount is not a good measure for "good links".

> 6. About "UP", please see the attached snapshot. I used 0.5.6-rc1 version
> and I get the screen like that. Also it lists "ILQ" not "NLQ". Is this
> meant to be NLQ itself?
Yes, it's just an error in the summary above the values... what you see is

Source-IP   Dest-IP   (LQ/NLQ)   ETX

>  Should I use different version of OLSRd ?
RC means "release candidate", so you might be interested in testing a newer 
0.5.6 version (0.5.6-r3 is the recent "stable" one, we have some patches in 
our queue which might lead to 0.5.6-r4 soon). Of course there is the 
development branchh too...

> 7. And, I browsed the code little bit. In tc_set.h I saw 2 structures. I
> was hoping to find mapping between the fields topology set shown on the
> console and the structure. But could not figure this out.
tc_entry is a node in the topology graph.
tc_edge_entry is an edge between two nodes.

Henning

*************************************************
Diplom Informatiker Henning Rogge
Forschungsgesellschaft für
Angewandte Naturwissenschaften e. V. (FGAN) 
Neuenahrer Str. 20, 53343 Wachtberg, Germany
Tel.: 0049 (0)228 9435-961
Fax: 0049 (0)228 9435-685
E-Mail: (spam-protected)
Web: www.fgan.de
************************************************
Sitz der Gesellschaft: Bonn
Registergericht: Amtsgericht Bonn VR 2530
Vorstand: Dr. rer. nat. Ralf Dornhaus (Vors.), Prof. Dr. Joachim Ender 
(Stellv.)

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 197 bytes
Desc: This is a digitally signed message part.
URL: <http://lists.olsr.org/pipermail/olsr-users/attachments/20090122/d626952f/attachment.sig>