[Olsr-dev] Maximal LQ less than 1 on Ethernet

[Henning Rogge]

I just did a few experiments to discover the reason why the LQ of perfect 
links (switched ethernet) does not go up to 1.

The reason seems to be a combination of three points:
- we use a exponential moving average in the default LQ calculation
- we save the LQ values directly in the 1 byte network format
- we use a fixed point math system in the default LQ calculation

This two points creates a rounding problem that prevent the LQ value to rise 
above certain values.

The exponential depends on the following formula, with default alpha as 0.05:
 
    new_LQ = old_LQ * (1 - alpha) + measured_lq * alpha

the second part of this formula seems to get small enough that we cannot store 
it in a 1 byte value.

So we have to stop using the exponential average or change the number formats 
for LQ calculation, but I'm not completely sure about the effects of both 
sollutions.

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.)