Hi all, is there any way how to measure signal and noise levels on WRT54G(S) in dBm? "wl rssi" does not return dBm as far as I know.

iwconfig is able to return you the transmission power of your wrt54g, but I don't think it is possible to get the client side configured txpower using the rssi.

Oh, I'm sorry I didn't make myself clear. I need to measure signal quality of the AP (or the peer, simply the signal quality of another wireless device) from the client side. As I wrote before "wl rssi" and "wl noise" return something but the numbers are not in dBm thus are incompatible with numbers returned by different brand devices. iwconfig on Linksys does not return these numbers (but e.g. iwconfig on Sharp Zaurus PDA does) and iwspy causes instant reboot of the device (latest experimental release from April and custom built latest CVS too).

From the WL documentation (http://www.jawug.za.net/wl) it appears as if at least the noise is reported in dBm,
by the WL utility.

noise   Get noise (moving average) right after tx in dBm

In my experience, with wl rssi, and wl noise, they correlate very closely, so I'm going to go out on a limb and say that RSSI
is probably reported in dBm too.

Here is the output from one of my WRT's in client mode, for instance.

root@window:~# wl rssi
rssi is -59
root@window:~# wl noise
noise is -83
root@window:~# wl scan
root@window:~# wl scanresults
SSID: "linksys3"
Mode: Managed   RSSI: -59 dBm   noise: -84 dBm  Channel: 11
BSSID: 00:12:17:B7:8A:CB        Capability: ESS ShortSlot 
Supported Rates: [ 1(b) 2(b) 5.5(b) 11(b) 18 24 36 54 6 9 12 48 ]

It is clear that the output of "wl rssi" and "wl noise" is thus in fact, in dBm. Or whatever Broadcom's interpretation of it is.

Thank you TheRodent. I suspected rssi and noise to be in dBm but I was not sure. Thanks a lot.

Is there a way of measuring noise in (wds) ap mode. I get different readings on different ap's (different locations). for example on one ap I get noise is -24 on second -37 ? any ideas please, can the real level be atleas calculated from this value ? I mean is this -24 or -37 in some way real (accurate) value or is "totally taken from air" ?

Could someone explain how to actually interpret the RSSI values returned?  Say would -90dBm be better over -40dBm?

AFAIK when it comes to these values, higher means stronger. This is a bit counter-intuitive, since they are represented with negative numbers -- but in this case, a RSSI of -40 dBm would be higher (stronger) than one of -90 dBm.
Similarly, with noise you are looking to minimize it, therefore a noise of -90 dBm is lower (and thus better) than one of -40 dBm.
Keep in mind that decibels (dB) are measured on a logarithmic scale (as opposed to a linear one, like distance for example), hence -40 dBm is not "twice" as good as -80 dBm, it is actually 10000 times more powerful.

Thanks man, this makes much more sense to me now.  Perhaps something about this would do good in the wiki somewhere, since I couldn't really find it elsewhere.

afaict the values for 'wl noise' in wds mode are completely bogus.

as an example do 'wl noise' and it will say -37 or so, constantly.

then do a couple site scans, it will suddenly jump to -97 and stay there.

not sure wtf is up with that.