[Sdr] from Fred on Wed

Barry Buelow barry.buelow at gmail.com
Wed Apr 13 12:36:32 CDT 2011


repeating this to get it into the archive



http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=568-5093-ND

Other parts like Linear Tech are in the same price range. TI seems to be
double in cost as are Analog Devices.

Can we do this with a single ADC or do we need to do duals with I/Q (and 1/2
the sample rate)? I'm thinking a single faster converter and doing the I/Q
conversion either in the FPGA or in the Beagle board might work (a Hilbert
transform along the way somewhere?) ? But I am no expert on this. I dabble
in DSP enough to be dangerous. But this might be something to present to
someone around here who is a DSP/FPGA guru who wants to help.

Yes, it's SPI to control it and LVDS DDR to get the data out.. but that is
no issue if it's feeding a FPGA.. it's probably an advantage as less pins
need to be routed from the device to the FPGA.

When I looked into DAC's for TX, the price was about $12 for one that's
complementary.. obviously it's a lot easier to do a D/A...

But maybe $60-80 for the A/D and D/A's.. hopefully no more than $50 for a
FPGA. I think the DUC/DDC topology is very do-able if we don't have to do it
with I/Q to RF. Actually. it would likely be less expensive to do it this
way.. and 16-bits is "OK"..

80+50+100 for the rest of the board at cost means about a $250 board.. that
plus another $100 for RF parts (amplifier to 5W say, and a preamp, filters
and AGC) and $170 for a BB-xm (delivered with cables and Powersupply) is
about $550 dollars-ish.. so that's 1/3 of the RF Space receiver.. And that
will get us a Flex-1500 capability with out the Latency problems, and full
digital DDC/DUC... I think it can be done...



Fred
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://rf.org/pipermail/sdr/attachments/20110413/6c097d6e/attachment.html>


More information about the Sdr mailing list