[USRP-users] Daughter board with a log response
Marcus D. Leech
mleech at ripnet.com
Fri Apr 10 16:40:07 EDT 2015
On 04/10/2015 04:17 PM, Venkatesh Sandilya wrote:
> We are feeding a pulse modulated signal at 1090 MHz to the input of
> the N200 and looking at the output using Gnuradio companion. The GRC
> flowgraph is really simple and it contains a USRP source, few throttle
> blocks, a complex to magnitude block (which is essentially doing
> sqrt(I*I + Q*Q)) and looking at the output voltage using a scope sink.
> We start with an input power of -80 dbm and increase it in steps of 5
> and observe the output voltage (in mv). We take these values of input
> power vs output voltage and plot it in Excel. I think the receiver was
> saturating around -30 dbm when the observed output voltage was around
Turn the gain down on the DBSRX2.
Also you say "when the observed output voltage was around 1340mV". Since
you don't get to see any absolute voltage values on the digital side,
just magnitude values that are *proportional* to the input voltage at
the antenna, I'm curious as to how you derived your 1340mV value?
The ADCs on the USRP motherboard are set with 1.2V references, so the
highest value that can be "seen" at the ADC input would be 1200mV, but this
is *AFTER* downconversion and gain and baseband filtering. You
typically have a *lot* of gain ahead of the ADCs.
So, I'd start with an input power level of around -110dBm, and work your
way up from there. It doesn't surprise me that -30dBml, with some
amount of gain on your DBSRX2 produces an output from the DBSRX2 that
saturates the ADCs.
> On Fri, Apr 10, 2015 at 1:12 PM, Marcus D. Leech via USRP-users
> <usrp-users at lists.ettus.com <mailto:usrp-users at lists.ettus.com>> wrote:
> On 04/10/2015 12:33 PM, John Keller via USRP-users wrote:
>> We are using a USRP N200 that has a DBSR2X daughter board. We
>> are decoding signals at 1090 MHz that have a very large dynamic
>> range Our old system used receivers that had log amps with a
>> dynamic range of over 80 db and are linear when you plot input
>> power vs output amplitude.
>> We plotted the output of the DBSR2X using a signal generator as
>> an input source. We then plotted the input power vs output
>> amplitude and it is very non linear. Do you have a daughter board
>> with a log characteristic or could you suggest another way.
>> Thank you.
>> John Keller
>> USRP-users mailing list
>> USRP-users at lists.ettus.com <mailto:USRP-users at lists.ettus.com>
> How are you plotting this?
> The power of a signal is proportional to the square of the
> voltage. The DBSRX2 receiver should be very linear, except at the
> very upper end, perhaps.
> Logarithmic amplifier chains are generally *ONLY* used in
> high-dynamic-range power detection, they aren't general-purpose RF
> The DBSRX2 on an N200 should have more than 80dB of dynamic range,
> depending on sample rates, and bandwidth settings of the DBSRX2 card.
> The instantaneous power of a signal can be calculated using:
> P = (I*I)+(Q*Q)
> Square-root and average to taste.
> So, what is it that you're actually measuring and plotting? The
> DBSRX2 should be quite linear over a fairly large dynamic range.
> USRP-users mailing list
> USRP-users at lists.ettus.com <mailto:USRP-users at lists.ettus.com>
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the USRP-users