Thread: [Madwifi-devel] noise and signal level
Status: Beta
Brought to you by:
otaku
From: Dennis B. <den...@go...> - 2007-03-26 14:17:43
|
Hello list! This is output on my machine: herakles@cheetah:~$ iwconfig lo no wireless extensions. wifi0 no wireless extensions. eth0 no wireless extensions. sit0 no wireless extensions. ath0 IEEE 802.11g ESSID:"turbo" Mode:Managed Frequency:2.437 GHz Access Point: 00:16:B6:ED:E4:E9 Bit Rate:54 Mb/s Tx-Power:16 dBm Sensitivity=0/3 Retry:off RTS thr:off Fragment thr:off Power Management:off Link Quality=62/94 Signal level=-240 dBm Noise level=-46 dBm Rx invalid nwid:4491 Rx invalid crypt:0 Rx invalid frag:0 Tx excessive retries:0 Invalid misc:0 Missed beacon:0 Have a look at the "Signal level". Could THIS value be correct? If I disconnect the two antennas of a Linksys accesspoint, this value raises to roundabout 3dBm. The accesspoint is 50cm away from my laptop and it is positioned on the same desk. Another point: how is the "Link Quality" value calculated? Thanks in advance, Dennis |
From: Benoit P. <ben...@fr...> - 2007-03-26 18:15:34
|
Dennis Borgmann wrote: > Hello list! > > This is output on my machine: [cut] > ath0 IEEE 802.11g ESSID:"turbo" > Mode:Managed Frequency:2.437 GHz Access Point: > 00:16:B6:ED:E4:E9 > Bit Rate:54 Mb/s Tx-Power:16 dBm Sensitivity=0/3 > Retry:off RTS thr:off Fragment thr:off > Power Management:off > Link Quality=62/94 Signal level=-240 dBm Noise level=-46 dBm > Rx invalid nwid:4491 Rx invalid crypt:0 Rx invalid frag:0 > Tx excessive retries:0 Invalid misc:0 Missed beacon:0 > > Have a look at the "Signal level". Could THIS value be correct? If I > disconnect the two antennas of a Linksys accesspoint, this value raises > to roundabout 3dBm. The accesspoint is 50cm away from my laptop and it > is positioned on the same desk. > > Another point: how is the "Link Quality" value calculated? I have been investigated this type of issue recently since i've seen incorrect value in the rssi reported through iwpsy (it was reported packets from another bss). Doing that, i've seen that the hardware reports two things: - rssi as a signed 8 bits in the hardware descriptor - the level of noise. I am not all sure that rssi is SNR and that signal = noise + rssi However, the iwspy API (and probably the iwconfig as well) use UNSIGNED 8 bits. Thus, a rssi value of -10 leads to 246 in unsigned. Link Quality is simply RSSI / RSSI Range. In your case, RSSI = 62 and RSSI Range = 94 Since the formula, Signal = Noise + RSSI is used, you get: Signal = (unsigned)-46 + (unsigned)62 Signal = 210 + 62 Signal = 16 Now, look in iwlib.c (wireless-tools source code), it substract 0x100 before printing, leading to -240 dBm ! I really like to know what is the correct signness/behaviour troughout this code path! btw, i think your results is a proof that rssi is independant from noise measurement. Ie, signal = rssi + some constant and noise = whatever HAL reports. In my case, i've seen rssi values ranging from 95 to -10, which nearly no values between 20 and 40. Best regards, Benoit |