[Battlemesh] dBm values > colours

Micheldegeofroy micheldegeofroy at gmail.com
Wed Mar 7 23:01:00 CET 2012

 If you lable it as RSSI then the proposed seems fine... RSSI is RSSI 

What I would find interesting is how to grade a connection from user point of view.

It think it has to take into account the actual speed from the client all the way through to a specific point...a type of UTC of speed test on the web... Having seen a map of the world traffic ... that point is probably somewhere in the USA 


On Mar 7, 2012, at 22:18, Federico Capoano <federico at nemesisdesign.net> wrote:

> Il 07/03/2012 20.19, Jo-Philipp Wich ha scritto:
>> the Kernel uses RSSI values between -110 and -40 to calculate its
>> "signal quality", the formula is simple:
>>  quality = min(max(rssi, -110), -40) + 110
>> That gives 70 quality levels, divided by your three color choices gives
>> ca. 23 dB spans, so it would result in:
>> -110 .. -87  -> red
>> -86  .. -64  -> orange
>> -63  .. -41  -> green
> Thanks jow!
>> Whether this kind of quality indicator actually makes sense is
>> debatable, but thats one way to do it.
> Yea it's quite debatable. At the moment our values are quite not right http://map.ninux.org (click on dBm on the right column). We'd like to have values that are more useful in order to have a quick overview of the network.
> Maybe... could that range be restricted?
> I guess a link that is worse than -90 does not have much sense. I don't know if there is a very noticeable difference in performance between a -50 and a -40 link, anyway, links with those values happen rarely in our network, only if two nodes are very very near. I checked now (Info tab in upper right corner) and we have only about 10 links in that range of dBm values out of 132.
> The reason for which I ask you guys is because I'd like to hear opinions and experiences from other wireless communities ^_^
> Federico Capoano
> Web Designer & Web Developer
> Portfolio/Blog: nemesisdesign.net
> Twitter: @nemesisdesign
> PGP Key ID: 308BD46E
> _______________________________________________
> Battlemesh mailing list
> Battlemesh at ml.ninux.org
> http://ml.ninux.org/mailman/listinfo/battlemesh
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://ml.ninux.org/pipermail/battlemesh/attachments/20120307/b2272b0a/attachment-0001.html>

More information about the Battlemesh mailing list