Re: [Discuss-gnuradio] AWGN using E310

[John B. Wood]

On 01/05/2017 02:22 PM, Marcus Müller wrote:
> Hi John,
>
> sure! Just use the noise source with a noise variance (amplitude) low
> enough to allow for the resulting amplitude to very rarely exceed 1.0
> (because that would lead to artifacts/non-gaussian distribution) (or,
> better yet, use the fast noise source, which preallocates a definable
> amount of noise that it then plays over and over again), and connect it
> to the USRP sink.
>
> Notice that the E310 isn't sold calibrated – you'll need to calibrate
> the noise power generated using e.g. a calibrated spectrum analyzer
> yourself.
>
> Best regards,
>
> Marcus
>
>
> On 01/05/2017 08:18 PM, John B. Wood wrote:
> > Hello, all.  This may be a "RTFM" question but with the stock tools in
> > GRC (or a non-GRC generated .py file) can a gaussian white noise
> > generator be readily implemented on an Ettus E310 to provide a
> > broadband RF noise output (some level in dBW/Hz) over a specified
> > bandwidth (within the capabilities of the E310) that can be used to
> > evaluate noise performance of radio receivers?  If not feasible then I
> > will probably have to purchase a noise generator. Thanks for your time
> > and comment.  Sincerely,
> >
> >
> >                                              J. Wood
> >                                              U.S. Naval Research
> > Laboratory
> >
> > _______________________________________________
> > Discuss-gnuradio mailing list
> > address@hidden
> > https://lists.gnu.org/mailman/listinfo/discuss-gnuradio
Hello, Marcus and all.  I finally got around to trying to implement the 
above suggestion on an Ettus E310 but am a bit puzzled.  The suggestion 
is to connect the GRC noise gen block to a USRP sink. What I want to 
accomplish is to use the E310 (USRP source?) to output white noise over 
a prescribed frequency band (e.g. 1-2 GHz) such that an E310 transmit 
antenna port could be then  connected to the device under test (DUT) for 
noise figure determination.  Seems like a simple two-block GRC flow 
graph.  Thanks for any clarification.  Sincerely,


                          J. Wood
                          U.S. Naval Research Lab