Hi all, I need to do some wideband spectrum measurements with more than 25MHz bandwidth and therefore have to use the wire format complex int8 in the UHD-USRP source. But this gives me a wrong spectrum with additional signals at wrong frequencies in the FFT spectrum. When I go back to wire format complex int16 all is fine, but there the maximum sample rate is 25Msps which is not enough, I need the 50Msps. Please see the attached file that shows the problem. Even when I use wire format complex int8 at a sample rate of 25Msps the delivered spectrum is wrong. Is this a problem of the UHD driver or the firmware of the N210, or do I made something wrong in my GRC flowgraph? Best regards Manfred

On 06/24/2014 04:50 PM, Manfred Weiss via USRP-users wrote: > Hi all, > I need to do some wideband spectrum measurements with more than 25MHz > bandwidth and therefore have to use the wire format complex int8 in > the UHD-USRP source. But this gives me a wrong spectrum with > additional signals at wrong frequencies in the FFT spectrum. When I go > back to wire format complex int16 all is fine, but there the maximum > sample rate is 25Msps which is not enough, I need the 50Msps. Please > see the attached file that shows the problem. > Even when I use wire format complex int8 at a sample rate of 25Msps > the delivered spectrum is wrong. Is this a problem of the UHD driver > or the firmware of the N210, or do I made something wrong in my GRC > flowgraph? > > Best regards > Manfred > > > _______________________________________________ > USRP-users mailing list > USRP-users@lists.ettus.com > http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com Could you try changing your peak=0.01 to peak=0.05 and see if that changes your results? -- Marcus Leech Principal Investigator Shirleys Bay Radio Astronomy Consortium http://www.sbrac.org

One other quick thought…the N210 daughter boards typically have an approximately 40MHz analog passband, and this filter rolls off slowly…it's possible you might see some spectrum alias and fold in at the edges of your 50MHz. -Ian On Jun 24, 2014, at 2:12 PM, Marcus D. Leech via USRP-users <usrp-users@lists.ettus.com> wrote: > On 06/24/2014 04:50 PM, Manfred Weiss via USRP-users wrote: >> >> Hi all, >> I need to do some wideband spectrum measurements with more than 25MHz bandwidth and therefore have to use the wire format complex int8 in the UHD-USRP source. But this gives me a wrong spectrum with additional signals at wrong frequencies in the FFT spectrum. When I go back to wire format complex int16 all is fine, but there the maximum sample rate is 25Msps which is not enough, I need the 50Msps. Please see the attached file that shows the problem. >> Even when I use wire format complex int8 at a sample rate of 25Msps the delivered spectrum is wrong. Is this a problem of the UHD driver or the firmware of the N210, or do I made something wrong in my GRC flowgraph? >> >> Best regards >> Manfred >> >> _______________________________________________ >> USRP-users mailing list >> USRP-users@lists.ettus.com >> http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com > Could you try changing your peak=0.01 to peak=0.05 and see if that changes your results? > > > > -- > Marcus Leech > Principal Investigator > Shirleys Bay Radio Astronomy Consortium > http://www.sbrac.org > _______________________________________________ > USRP-users mailing list > USRP-users@lists.ettus.com > http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com

Ach! No, never mind, I was thinking of X300 not N210…N210 sample rate is fixed @ 100MHz….this shouldn't be an issue for you. On Jun 24, 2014, at 2:18 PM, Ian Buckley <ianb@ionconcepts.com> wrote: > One other quick thought…the N210 daughter boards typically have an approximately 40MHz analog passband, and this filter rolls off slowly…it's possible you might see some spectrum alias and fold in at the edges of your 50MHz. > -Ian > > On Jun 24, 2014, at 2:12 PM, Marcus D. Leech via USRP-users <usrp-users@lists.ettus.com> wrote: > >> On 06/24/2014 04:50 PM, Manfred Weiss via USRP-users wrote: >>> >>> Hi all, >>> I need to do some wideband spectrum measurements with more than 25MHz bandwidth and therefore have to use the wire format complex int8 in the UHD-USRP source. But this gives me a wrong spectrum with additional signals at wrong frequencies in the FFT spectrum. When I go back to wire format complex int16 all is fine, but there the maximum sample rate is 25Msps which is not enough, I need the 50Msps. Please see the attached file that shows the problem. >>> Even when I use wire format complex int8 at a sample rate of 25Msps the delivered spectrum is wrong. Is this a problem of the UHD driver or the firmware of the N210, or do I made something wrong in my GRC flowgraph? >>> >>> Best regards >>> Manfred >>> >>> _______________________________________________ >>> USRP-users mailing list >>> USRP-users@lists.ettus.com >>> http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com >> Could you try changing your peak=0.01 to peak=0.05 and see if that changes your results? >> >> >> >> -- >> Marcus Leech >> Principal Investigator >> Shirleys Bay Radio Astronomy Consortium >> http://www.sbrac.org >> _______________________________________________ >> USRP-users mailing list >> USRP-users@lists.ettus.com >> http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com >

On 24/06/2014 22:50, Manfred Weiss via USRP-users wrote: > Hi all, > I need to do some wideband spectrum measurements with more than 25MHz > bandwidth and therefore have to use the wire format complex int8 in the > UHD-USRP source. But this gives me a wrong spectrum with additional > signals at wrong frequencies in the FFT spectrum. When I go back to wire > format complex int16 all is fine, but there the maximum sample rate is > 25Msps which is not enough, I need the 50Msps. Please see the attached > file that shows the problem. > Even when I use wire format complex int8 at a sample rate of 25Msps the > delivered spectrum is wrong. Is this a problem of the UHD driver or the > firmware of the N210, or do I made something wrong in my GRC flowgraph? Hello Manfred, your "wrong" spectra look strangely symmetric (especially the ones on slides 7 and 8). From the top of my head I cannot say what is the problem, but it may be in the FFT computation more than in the data stream. Have you tried to visualize the data in the time domain? If the spurious peaks you see in the spectra are really there the signal distortion should be easy to identify. Cheers, Daniele