Hi Marcus. I forgot to state - I'm using the X310 with UBX 40MHz bandwidth dauhterboard. I've been using one of UHD 4.0 releases, and I just recently changed to the latest UHD 4.1 release where I see same behavior.
I realize now that I'm looking at manual for 3.15 version of UHD - no surprise there was a mismatch and that "skip_duc=1" does not do anything. I'm using the "HG" image, and it lists connection between DUC and Radio as 'static', similar to here: https://kb.ettus.com/Getting_Started_with_RFNoC_in_UHD_4.0 So unless there is some config register or similar for the DUC, then I have to rebuild the image to make the connection dynamic, or remove the DUC altogether? The x310 seems to accept several master clock settings in the range 180-200MHz. Regards, Andreas. -----Original Message----- From: Marcus Müller <marcus.muel...@ettus.com> Sent: onsdag 24. november 2021 12:35 To: usrp-users@lists.ettus.com Subject: [USRP-users] Re: Creating sinc-spectrum This message is from an EXTERNAL SENDER - be CAUTIOUS, particularly with links and attachments Hi Andreas, which USRP is this about? My guess is the X310, and you're using the BasicTX daughterboard. Is that correct? To talk about features of the DSP architecture, it's also (very) relevant to know which version of UHD we're talking about! > I have tried to add skip_duc=1 which is mentioned in some of the > documentation, but the output spectrum does not change. That's true, I only find a single seemingly orphaned code mention of skip_duc... but it was only mention (if I remember correctly) in the documentation of the N3xx, so that would conflict with my X3xx guess... > Is this command outdated or not supported? I wonder since the HG FPGA > image has listed that the routing between DUC and radio are “static”. This depends on the UHD version; generally, RFNoC will allow you to feed full-rate signal into the radio frontend. The "newer" the UHD, the more flexible it gets, essentially. > have noticed that I get something closer to what I want if I choose an > odd interpolation order, by setting master_clock_rate to 190 MHz instead of 200MHz. Uh! That now completely clashes with my guess, because the X310 doesn't support 190 MHz MCR at all, I thought. But yes, if you use an odd interpolation on any third series USRP, you get "less good" spectrum, because none of the relatively good half-band filters can be used, and the adjustable CIC filter needs to do it all, leading to sidelobes, which in turn lead to more squarishness. Best regards, Marcus On 24.11.21 11:40, Andreas.Bertheussen--- via USRP-users wrote: > > Hi, > > I’d like to create (imitate) the sinc(f/fs) spectrum that would be > generated from an unfiltered zero-order hold DAC output, clocked at 10 > Mhz. Main characteristics are a peak around center frequency, rolling off to > nulls at +- 10 MHz around the center frequency. > > The IF signal is a random BPSK sequence. > > When I stream the BPSK sequence at 10 MHz , I get a flat spectrum > between +- 5MHz since the DUC is doing its job in upsampling to 200MHz DAC > rate. > > I have tried to add skip_duc=1 which is mentioned in some of the > documentation, but the output spectrum does not change. Is this > command outdated or not supported? I wonder since the HG FPGA image has > listed that the routing between DUC and radio are “static”. > > I have noticed that I get something closer to what I want if I choose > an odd interpolation order, by setting master_clock_rate to 190 MHz > instead of 200MHz. The upsampling ratio becomes 190MHz/10MHz = 19, > which is odd, and I get some warnings about half-band filters and CIC roll-of > in the console. > > In this mode, what does the DUC do to the signal? Is it equivalent to > bypassing the DUC? > > How do I best avoid any effects introduced by DUC? > > Regards, > > Andreas. > > > ---------------------------------------------------------------------- > -------------------- > > CONFIDENTIALITY > This e-mail and any attachment contain KONGSBERG information which may > be proprietary, confidential or subject to export regulations, and is > only meant for the intended recipient(s). Any disclosure, copying, > distribution or use is prohibited, if not otherwise explicitly agreed > with KONGSBERG. If received in error, please delete it immediately from your > system and notify the sender properly. > > _______________________________________________ > USRP-users mailing list -- usrp-users@lists.ettus.com To unsubscribe > send an email to usrp-users-le...@lists.ettus.com _______________________________________________ USRP-users mailing list -- usrp-users@lists.ettus.com To unsubscribe send an email to usrp-users-le...@lists.ettus.com ________________________________ CONFIDENTIALITY This e-mail and any attachment contain KONGSBERG information which may be proprietary, confidential or subject to export regulations, and is only meant for the intended recipient(s). Any disclosure, copying, distribution or use is prohibited, if not otherwise explicitly agreed with KONGSBERG. If received in error, please delete it immediately from your system and notify the sender properly. _______________________________________________ USRP-users mailing list -- usrp-users@lists.ettus.com To unsubscribe send an email to usrp-users-le...@lists.ettus.com
