I had noticed that the first ten samples of my application were always coming up as zero regardless of whether it was 8-bits on host or 16-bits on host. I went down the path of trying to reproduce a minimal example to share with this mailing list, however I then realized that even the example "rx_timed_samples.cpp" was also producing the same results. This occurred with two different b200minis and a B210 with UHD 4.7.0.0. I had to de-boostify the source code to run on my system and specify a center frequency and receive gain, but am able to reproduce this issue every time. I've confirmed that a signal is present by using a signal generator for one setup and also with an OTA setup tuned to 2421 MHz with AGC disabled and receive gain set to 70 dB. Has anyone observed this issue or can anyone else reproduce it using the timed receive example as well? Sorry if I am missing critical details that would help diagnose the issue, let me know if any additional information would be helpful. - Chris

On 07/11/2024 14:53, Chris Wozny wrote: > I had noticed that the first ten samples of my application were always > coming up as zero regardless of whether it was 8-bits on host or > 16-bits on host. I went down the path of trying to reproduce a minimal > example to share with this mailing list, however I then realized that > even the example "rx_timed_samples.cpp" was also producing the same > results. This occurred with two different b200minis and a B210 with > UHD 4.7.0.0. I had to de-boostify the source code to run on my system > and specify a center frequency and receive gain, but am able to > reproduce this issue every time. I've confirmed that a signal is > present by using a signal generator for one setup and also with an OTA > setup tuned to 2421 MHz with AGC disabled and receive gain set to 70 dB. > > Has anyone observed this issue or can anyone else reproduce it using > the timed receive example as well? Sorry if I am missing critical > details that would help diagnose the issue, let me know if any > additional information would be helpful. > > - Chris > The signal must necessarily pass through some digital filtering on the way between the antenna and your application.   Those digital filters have a certain length, and thus group delay.   That filter must necessarily have *some* value already   in it prior to your samples being presented to it.

Thank you! The sample delay induced by the filter taps makes sense. Is there a best practice that folks go with to deal with this or is it so insignificant that people don't care? I was thinking of either requesting 10 more samples and skipping the first 10 when writing to the filesystem / reading them into the software that consumes them. On Thu, Nov 7, 2024 at 3:03 PM Marcus D. Leech <patchvonbraun@gmail.com> wrote: > On 07/11/2024 14:53, Chris Wozny wrote: > > I had noticed that the first ten samples of my application were always > > coming up as zero regardless of whether it was 8-bits on host or > > 16-bits on host. I went down the path of trying to reproduce a minimal > > example to share with this mailing list, however I then realized that > > even the example "rx_timed_samples.cpp" was also producing the same > > results. This occurred with two different b200minis and a B210 with > > UHD 4.7.0.0. I had to de-boostify the source code to run on my system > > and specify a center frequency and receive gain, but am able to > > reproduce this issue every time. I've confirmed that a signal is > > present by using a signal generator for one setup and also with an OTA > > setup tuned to 2421 MHz with AGC disabled and receive gain set to 70 dB. > > > > Has anyone observed this issue or can anyone else reproduce it using > > the timed receive example as well? Sorry if I am missing critical > > details that would help diagnose the issue, let me know if any > > additional information would be helpful. > > > > - Chris > > > The signal must necessarily pass through some digital filtering on the > way between the antenna and your application. > Those digital filters have a certain length, and thus group delay. > That filter must necessarily have *some* value already > in it prior to your samples being presented to it. > > _______________________________________________ > USRP-users mailing list -- usrp-users@lists.ettus.com > To unsubscribe send an email to usrp-users-leave@lists.ettus.com >

 On 07/11/2024 14:53, Chris Wozny wrote:

 always

 minimal

 that

 system

 an OTA

 to 70 dB.

 using

 The signal must necessarily pass through some digital filtering on
 the
 way between the antenna and your application.
    Those digital filters have a certain length, and thus group delay.
 That filter must necessarily have *some* value already
    in it prior to your samples being presented to it.

 _______________________________________________
 USRP-users mailing list -- usrp-users@lists.ettus.com
 To unsubscribe send an email to usrp-users-leave@lists.ettus.com

On 07/11/2024 15:46, Chris Wozny wrote: > Thank you! The sample delay induced by the filter taps makes sense. Is > there a best practice that folks go with to deal with this or is it so > insignificant that people don't care? I was thinking of either > requesting 10 more samples and skipping the first 10 when writing to > the filesystem / reading them into the software that consumes them. I think it depends very much on the application. There are other things that may be "surprising" if you aren't used to systems that sample the physical world.  If you   start sampling right after a re-tune event, for example, there will inevitably be transients in the data.  This can be   kind of unsettling for people who have spent most of their time in the strictly-digital world of computers and   at the higher-layers of comms systems, where everything is already nicely quantized for you and delivered in   nice neat bits... > > On Thu, Nov 7, 2024 at 3:03 PM Marcus D. Leech > <patchvonbraun@gmail.com> wrote: > > On 07/11/2024 14:53, Chris Wozny wrote: > > I had noticed that the first ten samples of my application were > always > > coming up as zero regardless of whether it was 8-bits on host or > > 16-bits on host. I went down the path of trying to reproduce a > minimal > > example to share with this mailing list, however I then realized > that > > even the example "rx_timed_samples.cpp" was also producing the same > > results. This occurred with two different b200minis and a B210 with > > UHD 4.7.0.0. I had to de-boostify the source code to run on my > system > > and specify a center frequency and receive gain, but am able to > > reproduce this issue every time. I've confirmed that a signal is > > present by using a signal generator for one setup and also with > an OTA > > setup tuned to 2421 MHz with AGC disabled and receive gain set > to 70 dB. > > > > Has anyone observed this issue or can anyone else reproduce it > using > > the timed receive example as well? Sorry if I am missing critical > > details that would help diagnose the issue, let me know if any > > additional information would be helpful. > > > > - Chris > > > The signal must necessarily pass through some digital filtering on > the > way between the antenna and your application. >    Those digital filters have a certain length, and thus group delay. > That filter must necessarily have *some* value already >    in it prior to your samples being presented to it. > > _______________________________________________ > USRP-users mailing list -- usrp-users@lists.ettus.com > To unsubscribe send an email to usrp-users-leave@lists.ettus.com > > > _______________________________________________ > USRP-users mailing list --usrp-users@lists.ettus.com > To unsubscribe send an email tousrp-users-leave@lists.ettus.com