On Mon, Jun 27, 2011 at 11:02 AM, Tom Hendrick <sdtom...@yahoo.com> wrote:
> I have a really simple question that I can't seem to find the answer to > online (and pardon my limited signal processing background). I'm trying to > help a fellow colleague figure out a processing problem. > > An external code modulates a real-valued signal that is eventually fed into > GRC. For now the signal is generated with some arbitrary positive center > frequency and then a GRC script converts it to complex and then downshifts > and centers it at 0Hz. This isn't efficient because the modulation in the > external code takes a lot longer with higher sampling frequency. > > If the external code was modified to put the real-valued signal in the -5 > to 5kHz band (centered at 0Hz and assuming a real-valued signal can > represent negative frequencies), is the minimum sampling frequency to avoid > aliasing 20kHz or is it 10kHz? This is before the GRC script converts it > to complex. > > Thank you in advance.. Tom > > > > _______________________________________________ > Discuss-gnuradio mailing list > Discuss-gnuradio@gnu.org > https://lists.gnu.org/mailman/listinfo/discuss-gnuradio > > A few things: * A real valued signal has a frequency spectrum that is symmetric around 0Hz. So the negative frequency components are identical to the positive frequency components. * For your case, the minimum sampling frequency would be 10KHz. The signal being real or complex does not make a difference. If a signal occupies frequency -F to F at the baseband (centered), your Nyquist sampling rate is 2F. Hope this helps, Colby
_______________________________________________ Discuss-gnuradio mailing list Discuss-gnuradio@gnu.org https://lists.gnu.org/mailman/listinfo/discuss-gnuradio
