RFX2400 Transmit power

Hi, all!

Now I’m using USRP1 and RFX2400 d’board. In our project, we have to be
able
to reduce transmit power down to -40dBm. Please correct me if my
approach is
wrong.

If we set tx_pga=0dB, the transmit power only depends on the level of
signal. So, +/-32767 can give us maximum transmit power 17dBm because
the
maximum transmit power of RFX2400 is 17dBm.

Since the power is a square of signal level, we can get -40dBm if we
choose
signal level +/-45 roughly.

Am I right?


View this message in context:
http://www.nabble.com/RFX2400-Transmit-power-tp21575529p21575529.html
Sent from the GnuRadio mailing list archive at Nabble.com.

Hi Yong,

I think with maximum signal levels of +/- 45 on the digital side your
signal would be severely distorted due to quantization noise. Perhaps
the easiest thing to do would be to put a mini-circuits attenuator
inline with the SMA output of the RFX2400 board? Not sure if this will
work for your application though.

-Brian

Yep,

Brian’s right. Doing what you propose would be equivalent to reducing
the DAC to ~7bit and thus reducing signal-to-quantisation-noise-ratio by
54dB (!). You don’t want that.

Try subdev.set_gain(). You can see the available gain range with
subdev.gain_range() (subdev is the object to access the subdevice, try
usrp.selected_subdev() ). Perhaps that will help.

MB

Thanks all!

As I know so far, RFX2400 doesn’t have amplifier we can control. So
set_gain() doesn’t affect any transmit power. Right?
Now I’m trying to control PGA gain on FPGA side, but I confused. AD9862
has
pga which can be controled in range [0 ~ -20dB].

self.u.set_pga(0,-20)
pga_gain = self.u.pga(0)
print 'PGA gain ',pga_gain

After running this code, I can see pga() returns -20. But transmit power
is
not changed. What’s wrong?


View this message in context:
http://www.nabble.com/RFX2400-Transmit-power-tp21575529p21586587.html
Sent from the GnuRadio mailing list archive at Nabble.com.

Yong J. Chang wrote:

You can use the PGA to reduce the gain on the RFX-series boards, but the
problem is that the analog bias points for the quadrature modulator will
change. You may be able to reduce 5-10 dB, but it may not operate well
beyond that.

The other way to change the output power is to reduce your digital level
sent over the USB. Yes, this will reduce your dynamic range, so you
won’t want to do -40dBm this way, but 10 or 20 dB in reduction is fine.

Matt