From: Holger Schurig <hs4233@mail.mn-solutions.de>
To: Dan Williams <dcbw@redhat.com>
Subject: Re: [WIP, RFC] libertas: allow scanning via "iw"
Date: Wed, 7 Oct 2009 11:38:49 +0200
Cc: linux-wireless <linux-wireless@vger.kernel.org>
References: <200910061633.08237.hs4233@mail.mn-solutions.de> <1254869377.8034.8.camel@localhost.localdomain>
In-Reply-To: <1254869377.8034.8.camel@localhost.localdomain>
MIME-Version: 1.0
Content-Type: text/plain;
  charset="iso-8859-15"
Message-Id: <200910071138.49179.hs4233@mail.mn-solutions.de>
Sender: linux-wireless-owner@vger.kernel.org

> > * signal isn't reported in dBm or mBm. The scan response from
> >   the firmware tells me some "RSSI" value.
> 
> #define MRVDRV_NF_DEFAULT_SCAN_VALUE          (-96)
> #define SCAN_RSSI(RSSI)                       (0x100 - ((u8)
(RSSI)))
> 
>       iwe.u.qual.level = SCAN_RSSI(bss->rssi);
> 
>       rssi = iwe.u.qual.level - MRVDRV_NF_DEFAULT_SCAN_VALUE;
> 
> so it's more or less dBm.

I'm not that sure about that. I made a test and moved with my 
laptop away from an access-point. In the table below you find 
the signal from ipw2200/iwconfig. And in the second column the 
raw "u8 rssi" from libertas' scan response:

ipw2200 libertas
signal  rssi
-28     38
-36     42
-42     44
-32     46
-38     47
-41     42
-49     47
-49     48
-49     49
-49     48
-47     59
-49     65
-52     54
-55     57
-43     73

So, it's everything but an RSSI, because an RSSI would be high in 
the vincinity of the AP and decrease when farther away. The name 
is bogus, don't put too much value in it :-)

SCAN_RSSI(rssi) = 0x100 - rssi = 218. That doesn't look like a 
dBm either.

-- 
http://www.holgerschurig.de