carl9170 noise floor calculation

carl9170 noise floor calculation

[Manuel Mendez]

We are developing a product that uses a board similar to the DLink
DWA-160. We are sticking an amplifier before the usb radios. I am
pumping beacons at say -80dBm from a siggen and a program reading the
radiotap header prints them out.

I am seeing ~20dBm gain which looks right, when taking into account my
PA and the one that looks to be on the actual USB cards.

Is there a way to use the pre-gain values for noise floor calculations,
and reporting of signal strength?

Manny

Re: carl9170 noise floor calculation

[Christian Lamparter]

On Thursday 02 June 2011 22:46:53 Manuel Mendez wrote:
> We are developing a product that uses a board similar to the DLink
> DWA-160.
Uh, there are at least two DWA-160 AR9170 revision and mysteriously
users have reported strange problems with both?!

So, my question: why not base your product on ath9k_htc instead?
> We are sticking an amplifier before the usb radios. I am pumping
> beacons at say -80dBm from a siggen and a program reading the
> radiotap header prints them out.
> 
> I am seeing ~20dBm gain which looks right, when taking into account my
> PA and the one that looks to be on the actual USB cards.
> 
> Is there a way to use the pre-gain values for noise floor calculations,
> and reporting of signal strength?
probably not, in fact the whole noise floor and signal strength calculations
are not as straight forward as you might think?!

http://www.mail-archive.com/ath9k-devel@lists.ath9k.org/msg05777.html
http://www.spinics.net/lists/linux-wireless/msg50060.html
[there's more, just ask google about ath9k and noise]

Also, since you are adding an extra amplifier: Are you thinking about
long-range links? Because you'll need to tinker with the round trip time
MAC registers then.

Regards,
	Chr