GRC+USRPx file sink problem

Hi,

GRC Setup (used with RFX and WBX cards)…

USRP1 —to—> FFT Graphical sink (shows signal spectrum as expected)
and —to—> Complex to IShort -----to----> File Sink (signals
are dumped
to disk as expected)

USRP2 —to—> FFT Graphical sink (shows signal spectrum as expected)
and —to—> Complex to IShort ----to----> File Sink (file
is all
zeros
)

Tried with 3.3.0 and git 3.3.1, and on three different
installations/USRPs.
Basically zero output with USRP2.
USRP2 works fine with python scripts eg, usrp2_fft.py with
–output-shorts.

Any ideas please?

Dave

On 08/18/2010 01:41 PM, David E. wrote:

zeros******)

Tried with 3.3.0 and git 3.3.1, and on three different installations/USRPs.
Basically zero output with USRP2.
USRP2 works fine with python scripts eg, usrp2_fft.py with --output-shorts.

Any ideas please?

Dave

My guess is that with USRP2, samples are scaled -1.0 to 1.0. Which
means that a straight conversion to
a short integer is going to yield zero most of the time. Try putting
a multiplier block in front of your
file sink–maybe scale those -1.0 to +1.0 to +/- 32767.


Marcus L.
Principal Investigator
Shirleys Bay Radio Astronomy Consortium
http://www.sbrac.org

Thanks, so it is.

Unfortunately though, I have discovered that the gain for an RFX card
cannot be
adjusted (min=0, max=0, step=0), requiring a massive multiply value in
gnuradio-companion, so its all very distorted. :frowning:

Cheers,
Dave

----- Original Message ----
From: Marcus D. Leech [email protected]
To: [email protected]
Sent: Wed, 18 August, 2010 20:26:38
Subject: Re: [Discuss-gnuradio] GRC+USRPx file sink problem

On 08/18/2010 01:41 PM, David E. wrote:

USRP2 —to—> FFT Graphical sink (shows signal spectrum as expected)
and —to—> Complex to IShort ----to----> File Sink (******file is all

My guess is that with USRP2, samples are scaled -1.0 to 1.0. Which
means that a straight conversion to
a short integer is going to yield zero most of the time. Try putting
a multiplier block in front of your
file sink–maybe scale those -1.0 to +1.0 to +/- 32767.


Marcus L.
Principal Investigator
Shirleys Bay Radio Astronomy Consortium
http://www.sbrac.org


Discuss-gnuradio mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio

Thanks, so it is.

Unfortunately though, I have discovered that the gain for an RFX card cannot be
adjusted (min=0, max=0, step=0), requiring a massive multiply value in
gnuradio-companion, so its all very distorted. :frowning:

Cheers,
Dave

It should not be distorted just because of a multiply in the signal
path, unless that multiplied value is
getting clipped when converting to an ISHORT–but that will only be
“distorted” downstream from that
conversion.

Most of the functions that operate on floats in Gnu Radio are perfectly
happy with pretty much the entire
machine-native range of floats. But certainly when you convert from a
floating-point value to an integer
value like a short, you have to be aware of conversion losses that can
screw you up, so scale appropriately.

Cheers


Marcus L.
Principal Investigator
Shirleys Bay Radio Astronomy Consortium
http://www.sbrac.org

This forum is not affiliated to the Ruby language, Ruby on Rails framework, nor any Ruby applications discussed here.

| Privacy Policy | Terms of Service | Remote Ruby Jobs