At 11:44 AM 12/7/2005, kb5my@starband.net wrote:
>What about using multiple copies of the RX-only version of the SDR-1000?
>Built & tested for $899 each, Analog Devices DDS LO (no PLL or VCO), I/Q
>output, $10 option to add capability for locking to an external 10 MHz or
>20 MHz reference, and the software written in C# (C Sharp) is open source.
> You can tweak it to your heart's content.
The SDR1000 is an interesting radio (I have one at home, and four at work),
but not really "ready for prime time" in this application. Here's why:
Most of the radio is implemented in the attached PC. You'd need to
dedicate a separate PC to each radio, and then, potentially, a fifth one to
do the array processing, for a few reasons:
- The existing software to drive the SDR1000 requires a parallel printer
port. In these days of "legacy interface free" computers, it's hard to
find a computer with one parallel port, much less 4.
- The existing software for the SDR1000 doesn't necessarily support
multiple instances running at the same time. It might work, but it's not
something that's been tested extensively, if at all, and the possibility
for bugs is huge.
- You'd need an amazing number of audio interfaces (2x4 in for the sdr, 1x4
out for the sdr, 1x4 for the array, 2 out for the array), and just handling
the audio channels could get to be a nightmare. It's not clear that the
audio card drivers that handle multiple channels will let you assign
different channels to different processes. The SDR does now support Virtual
Audio Cable (so you could route the output of the sdr to the input of
something like MixW, without using a physical audio channel), but there are
instabilities in that, and, again, there's the possibility of weird
interactions, when running 4 radios.
- While the software is opensource, it is undocumented, and not exactly
filled with comments. Flexradio has said that they are going to produce
some documentation "real soon now", as has Bob N4HY, but they are really
focusing on making the software work, rather than making it open (in the
usable documentation sense). The prospect of reverse engineering several
tens of thousand lines of code is not a pleasant one.
Perhaps in a year or two, when the planned rearchitecting of the SDR1000
code base happens (so that the user interface is split from the radio
processing and there's explicit support for multiple radios) this might be
a possibility. I look forward fondly to the idea of putting a small
diskless motherboard with a SDR1000 in a box, and it feeds an audio stream
out over the ethernet port.
There's also the basic problem that while you can feed the same reference
oscillator to multiple SDR1000s (and you'd be better off feeding the 200MHz
around, rather than using 10MHz), there's no way to synchronize them,
because the "load" line for the DDS isn't brought out to the
interface. One could build some special purpose interface hardware, but
that increases the complexity of the overall task.
What I was really looking for was 4 essentially plug and play boxes that
would feed out audio streams, and the SDR1000 is quite a ways away from there.
>73,
>Dan KB5MY/6 DM13nc
>
Jim, W6RMK
_______________________________________________
See: http://www.mscomputer.com for "Self Supporting Towers", "Wireless Weather
Stations", and lot's more. Call Toll Free, 1-800-333-9041 with any questions
and ask for Sherman, W2FLA.
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
|