Bob - On Sat, Oct 01, 2005 at 02:56:26PM +0000, Robert McGwier wrote: > > The paper you included and the mathematical and electrical phenomenon > you are talking about applies to the analog to digital converter, the > receiver only. [chop] > > You cannot do the reverse on transmit. It is a "one to many" problem. > The only way you can get a higher frequency is for some (BAD) > nonlinearity to be producing the higher frequency content.
You _can_ do the reverse on transmit. Most real-life D-A chips put
out fast edges every time the sample updates. This gives a 1/f
amplitude dependence for the intermodulation products Jorge
mentioned. You need a band pass filter to select the one you
want. At higher frequencies, the power degrades, and other
properties get worse. The first few lines come out pretty clean,
though.
You don't have to believe me and Jorge. Get out a spectrum
analyzer and see for yourself! My experiments had a DAC902
from TI/Burr Brown clocked at 80 MS/s, with a nominal output
at 30 MHz. Lines were nice and crisp at 50, 110, and 130 MHz,
with the predicted 1/f amplitude envelope. The 50 MHz line
happened to be the one I kept with a band-pass filter. Works
great, with excellent linearity.
YMMV, especially depending on your choice of DAC.
- Larry
signature.asc
Description: Digital signature
_______________________________________________ Discuss-gnuradio mailing list Discuss-gnuradio@gnu.org http://lists.gnu.org/mailman/listinfo/discuss-gnuradio
