Forum

I was wondering if anyone knew the maximum signal strength a device can receive before the reciever gets overloaded and throughput actually goes down?    I'm guessing there is a typical -dbm that is similar across radios etc.  I'll probably get a step attenuator and do test soon anyway but figured someone might know this offhand, which can help.
Thanks! 

The 802.11 Standard lists different levels for the different modes (e.g. b and g), but not in the way you might imagine.   It is specified with limits based on Packet Error Rate (PER) at a particular power level and usually for each rate.
 
For example, with 802.11b at 11 Mbps the PER limit is 8 % at -10 dBm and -4 dBm at 1 Mbps (also at 8 %).  For 802.11g at 54 Mbps the maximum signal level is -30 dBm for both 6 and 54 Mbps where both are measured at 10% PER.

If you have ever seen WLAN radio sensitivity graph, where the PER plot climbs on the right side of the graph, the numbers above reflect the left side of these same graphs.

This is not to say that the radio is deaf anywhere above those power levels, but in some cases that is true.    The left, maximum power, side of the graph looks similar to the right side, but usually rises at a much steeper rate. 

As technology progresses, both the maximum and minimum (ie the sensitivity values) are improving.   However the numbers in the stanadard haven't changed much, if at all.    For a comparison look at Cisco's latest AP sensitivity specs compared to say even 8 years ago.

Five years ago I would sometimes see radios that would choke on power levels of -20 dBm.   I can't remember how long it's been since I've seen a radio fails even above -10  dBm power.

Hope that all helps.