Re: wild variation in signal level - flat noise level

Most often in my limited experience the noise level stays the same 
throughout. Are there any 2.4 GHz phones in the area? How long is the 
link in question? Do you have good line of sight? What are these 
antennas mounted on? Have you done a frenel zone calculation? How high 
off the ground are these? Is it related to weather/time of day/inversion 
layer effect? Perhaps there is an interfering AP on a similar channel. 
Connect the radios and antenna to your laptop and run netstumbler for a 
quick and dirty site survey. Pick the least used channel.

Sean

Brad Langhorst wrote:

>I've built a small community wireless network 
>One central AP with 2 radios
>One is for clients, the other is a backhaul link to a dedicated access
>point that connects to the firewall and the internet.
>
>I'm using smc radios (prism chipset - 200mW 1.7.4firmware, hostap) and
>panel antennae on both sides of the backhaul link.
>
>i see signal levels between -40dBm and -60dBm (20!db?)
>the noise level is almost always static at -92dBm
>
>The link quality metric is extremely variable  - going between 
>90 and ~0 dB
>
>here are the graphs
>http://charterstreet.dyndns.org/rrd/
>
>Sometimes i actually lose the connection 
>
>What could explain 
>1) the wild variation in "Link Quality"
>2) the less wild but still very big variation in signal level..
>
>obviously link quality depends on signal level but there's certainly
>something else in play. the noise level seems to be useless
>
>
>
>
>_______________________________________________
>BAWUG's general wireless chat mailing list
>[unsubscribe] http://lists.bawug.org/mailman/listinfo/wireless
>
>
>  
>
_______________________________________________
BAWUG's general wireless chat mailing list
[unsubscribe] http://lists.bawug.org/mailman/listinfo/wireless