Card Calibration

[pitcherj]

Hello,

I've upgraded from an early release of WirelessMon 2.0, and there have been many changes since the version I was running. I've noticed that you can now tell WirelessMon minimum and maximum dbm values that the wireless card will report. My problem is determining what in the world to enter for these values. My reason for investigating the choice of values here is that it appears the scale has changed and now I'm getting different readings than I did before. That is to say, when I do a signal strength map by clicking in the hexagons on the map it appears that I'm getting much different results for signal strength. I believe this to be related to the calibration setting which is now available. My wireless card is an Intel 3945ABG MiniPCI adapter. As of yet I've been unable to find minimum and maximum values that the card will report. Can anyone provide some guidance as to what I should do? I'm in the middle of a project at the moment and it's a tad bit annoying to have upgraded and find out that the scale is no longer the same. I know it's my fault for upgrading in the middle of the project, but the interpolation feature really attracted me.

Perhaps it would even be helpful if someone would provide me with values to enter in the minimum and maximum boxes so that the signal strength is interpreted as it was in previous versions of the program.

Jacob

[Tim (PassMark)]

If you don't have any vendor supplied software that reports the dBm strength you can run WirelessMon in debug mode and the strength for each sample should be in the log file on a line that looks like "GetAPList: ssid - -signal value".

The algorithm used to convert a dBm value to a percentage was tweaked slightly as well in that release and if you were using Vista the method used to get the signal strength was changed so a dBm value was retrieved instead of a strength value which could account for a large discrepancy between versions.

A specs document found through google (http://h18019.www1.hp.com/products/q...a/12510_na.PDF) had this line
"Receiver sensitivity 54 Mbps: -72 dBm, 11 Mbps: -89 dBm , 1 Mbps: -97 dBm", which would indicate you could put a min signal of -97 and a max of -72 , but it's likely the card could report a much higher value than -72 when next to an access point, this would just be the level at which a 54 Mbps link could be maintained at.

[quickstrike]

I'm curious about the card calibration feature.

It may be the answer of why my readings are a bit off...

My setup: I had two different machines running WirelessMon 2.0. One was Vista, the other XP. The Vista machine had an Intel NIC (in fact it is the Intel 3945ABG card in the initial post), and the XP machine had a Zonet NIC. They were in ad-hoc mode so I can pick up signal strength from each computer. At a certain distance apart, we took 10 readings and then averaged. The 10 readings were roughly taken at the same time. The readings were taken from the "RSSI" column in the main window of WirelessMon.

According to wirelessmon, the zonet card was transmitting at 100mW (20dBm). WirelessMon couldn't tell us what the Intel card was transmitting at, but that specs document in the previous post says the Intel card can transmit at 15dBm. One would expect that the zonet readings would be about 5 db weaker since the Intel card is transmitting with less power. Attached are my results.

What I see is the exact opposite! The zonet readings are higher.

My guess is that it has something to do with the card calibration feature??
I really didn't mess around with that feature and simply started the software up on the two machines and began taking measurements.


One other thing I am confused about is that I'd expect to see a consistent difference between the two machines, but in fact, the difference isn't uniform with distance. Looking at the two trendlines, at a shorter distance, the difference is about 5dB, but at the larger distances, the difference is about 10-12 dB.

I'm guessing this has to do with calibration as well and perhaps how WirelessMon maps the data?

Any help would be appreciated.

[Tim (PassMark)]

The calibration settings is a way of mapping the highest signal strength returned by a card to a % value as cards can return different values as a maximum/minimum strength, so different calibration values might affect the the readings you are getting. but differences between cards might also be related to other factors, like how directional the antenna is, power levels, frequency being used and local interference at a the time the test was done.