CVS-2-2 NMApplet empty bar explained

[Bill Moss <bmoss clemson edu>]

Wireless drivers report two types of quality information: results for an 
associated AP and a list of results for scanned AP's. These results are 
reported using

struct iw_quality
{
    uint8 qual;
    uint8 level;
    uint8 noise;
}

Unfortunately, as Dan pointed out in

http://www.redhat.com/archives/fedora-devel-list/2004-September/msg00767.html

drivers don't report the data in the same way for associated AP and 
scanned AP's. As an example, consider ipw2200-1.0.0 and NM CVS-2-2 on my 
T42, with these structures printed from NetworkManagerDevice.c. This is 
my home network with only one SMC AP in the scan list.

Feb  3 23:58:45 localhost NetworkManager: Activation (eth1) successful, 
device activated.

First, look at the results for the associated AP.

Feb  3 23:58:47 localhost NetworkManager: 
nm_device_update_signal_strength stats.qual.qual=76
Feb  3 23:58:47 localhost NetworkManager: 
nm_device_update_signal_strength stats.qual.level=203
Feb  3 23:58:47 localhost NetworkManager: 
nm_device_update_signal_strength stats.qual.noise=175

These same values show up in /proc/net/wireless

Inter-| sta-|   Quality        |   Discarded packets               | 
Missed | WE
face | tus | link level noise |  nwid  crypt   frag  retry   misc | 
beacon | 17
 eth1: 0000   76.  203.  175.       0      0      0      0      0        0

and in the output of iwconfig

Link Quality=79/100  Signal level=-53 dBm  Noise level=-81 dBm

The iwconfig level and noise values in dBm are obtained by substracting 
256 from the unsigned values [0 - 255] in the proc file and NM output. 
dBm is a measure of power.

The 79 is an RSSI value and 100 is RSSI_MAX. A quality percentage can be 
computed as

(100 * RSSI) / RSSI_MAX

Some drivers provide a lookup table to convert from dBm to RSSI along 
with RSSI_MAX. The ipw2200 driver does not do this. Instead, it 
originally used a linear model to convert from dBm to RSSI but the 
developers have accepted a more accurate quadratic model I submitted. I 
based by model on matching the percentages reported by the Windows XP 
driver for this card. That was the only thing I could think to do.

Now look at the results for the scanned AP's (list of one here). qual 
and noise are reported as 0. This is why the CVS-2-2 NMApplet bar is 
empty for the ipw2200 driver. NM uses the qual value of zero to create 
the bar length. We can convert the level of 198 to -58 dBm but NM has no 
idea how ipw2200 would convert this into a percentage. to make matters 
worse the dBm range varies from driver to driver. We are stuck. There is 
no way the NM nm_wireless_qual_to_percent function can deal with this. A 
new nm_wireless_level_to_percent function could be created to do this 
conversion but it would be guess work at best and impossible to 
normalize across all drivers.

Feb  3 23:58:50 localhost NetworkManager: 
nm_device_wireless_process_scan stats.qual.qual=0
Feb  3 23:58:50 localhost NetworkManager: 
nm_device_wireless_process_scan stats.qual.level=198
Feb  3 23:58:50 localhost NetworkManager: 
nm_device_wireless_process_scan stats.qual.noise=0

So what to do . . .

--
Bill Moss
Professor, Mathematical Sciences
Clemson University