I've assigned a homework that involves performance measurement of 802.11b
and also over mobileIP (HUT implementation).
I'm having students use a mix of the DELL/Cisco cards to connect to apple airports.
I also have some PCI cards (for the FA for mobileIP).
Standing right next to the base station, I get a signal strength of 100, but
a signal quality of "4" in linux. When I use windoz, I don't get this -- signal
quality (beacons, I'm assuming) is good. Same when using the Compaq WL100 in
windows - haven't figured out how to get signal quality from linux-wlan driver
Any idea why this might happen? They're all associated in ESS mode, using CAM
and on the proper channel and allegedly working at 11mbit/s. I unplugged all
microwaves, etc, but the fact that this isn't showing up in windoz makes me
think this is a driver problem.
Here's a sample Status from a PCI card in the same room as the apple AP,
about 20 feet away.
[itsydev-25] cat Status
Signal Strength: 100
Signal Quality: 3
SSID: Computer Science
Driver Version: airo.c 0.99zf 2000/03/25 16:27:17 (Benjamin Reed)
Manufacturer: Aironet Wireless Communications
Firmware Version: 3.57.01
Radio type: 2
Hardware Version: 20
Software Version: 0
Software Subversion: 0
Boot block version: 132
Dirk Grunwald, Assoc Professor,
Dept. of Computer Science, Boulder,
From: Jim Veneskey <jvene@bi...> - 2000-10-18 15:35:15
> Standing right next to the base station, I get a signal strength of 100, but
> a signal quality of "4" in linux. When I use windoz, I don't get this -- signal
> quality (beacons, I'm assuming) is good. Same when using the Compaq WL100 in
> windows - haven't figured out how to get signal quality from linux-wlan driver
> Driver Version: airo.c 0.99zf 2000/03/25 16:27:17 (Benjamin Reed)
> Device: PC4800B
> Manufacturer: Aironet Wireless Communications
> Firmware Version: 3.57.01
> Radio type: 2
Ok - the key here is you are using the "B" rev of the 4800 aka 340 series.
Signal quality is determined using a new algorithm for this series of
hardware and now, I believe, LOW numbers are good....
This was also confusing for the author of the WinDGS/ACU utility, so instead of
having the signal quality look terrible - the version of ACU/WinDGS you are using
detects it's a PC4800B/340 radio - and now judges it's quality of signal based on the
percentage of beacons received.
The forthcoming Linux version of ACU should display the signal strength/quality
in a comparable fashion to the Windows version.
No - I don't have a release date for it - but I hear it's due for BETA testing really
soon now... And no, I have no idea who the BETA testers are or how they have been
chosen - that's marketting's job I think...
| | Jim Veneskey
:|: :|: Software Test Engineer
:|||: :|||: 3875 Embassy Pkwy, Akron OH 44334
.:|||||||:..:|||||||:. Email: jvene@...