Re: [Madwifi-users] RSSI vs. txpower

[Frank C. <fc...@co...>]

With that short of a distance, you are max'ing out the noise floor almost
instantly so the automatic gain control on the receiver is continuously
compensating and recalibrating to keep a clean enough signal for data.  So
the RSSI value you are reading is after it has been adjusted and not a true
reading.

If you want to play with it, look in if_ath.c in the driver source.
Unfortunately the way it is written, every time the radio compensates the
drivers reset the chips and recalibrates so technically it is bad for the
radios and also causes your throughout to go in the crapper (technical
term). Look in the ath_reset function... comment out the body of the
function and rebuild the drivers.  Then you'll get a true RSSI value with
out compensation.  Just dont forget to put it back before production use. :)

Anyhow hope it helps....

Frank 

> -----Original Message-----
> From: mad...@li... 
> [mailto:mad...@li...] On 
> Behalf Of Robert Owen
> Sent: Thursday, July 06, 2006 10:24 PM
> To: mad...@li...
> Subject: [Madwifi-users] RSSI vs. txpower
> 
> Hi all,
> 
> I did some RSSI measurements under different power levels. It 
> turns out that, when the txpower exceeds a certain threshold, 
> RSSI almost does not change.
> I wonder if this is normal or I did something wrong.
> 
> Here is how I did the experiments:
> I have two nodes, 6-7 feet apart in an indoor office 
> environment (there are other wireless networks), one 
> broadcasting and the other receiving.
> The sender broadcasts 1000 packets with 1 second intervals.
> I set the band to 802.11a, and controlled the txpower by 
> specifying it in file: /proc/sys/dev/ath0/txpowlimit
> I repeated the experiments for these different power levels, 
> one after another:
> 1,5,10,20,30,40,50,60mW
> 
> For each power level, it took 1000 seconds or about 17 
> minutes. So the background interference could have changed, 
> hopefully not much, during the whole session.
> I used click to retrieve the RSSI values from an ath0raw device.
> 
> Here are the results I got:
> 
> tx power (mW)          1         5         10        20       
>   30         40        50        60
> mean RSSI       23.314  23.331  25.686  32.149  35.778  
> 35.758  35.733  35.693
> 
> Initially, there's only slight RSSI increase with txpower 
> from 1mW to 5mW.
> Then it seems to be normal with txpower from 5mW to 30mW.
> But from 30mW on, RSSI almost does not change.
> 
> I wonder if it's possible that the txpower increases from 
> 30mW on did not actually take effect?
> 
> Thank you!
> 
>