[hpsdr] single board rigs *YES* please...

[Chris Albertson]

> I vote for a single-board Mercury. Since its PCB design has not
> started yet,
> it may be possible to add an USB interface to the unit and turn it
> into a
> "direct RF sampling receiver for masses". Not only such receiver will
> allow
> experimentation with software on a large scale, but it will also have
> many
> practical applications. It could be used as

One can design to any criteria.  You can choose to optimize
performance or cost.  The Softrock 40 series transceiver kits
are sold for $35 shipping included and give us one example
of a "design to cost" SDR radio.  Because of the low price
there are MANY of these in use and under construction.
A more sophisticated transceiver could be designed for a
price point of just under $100.

The first design decision should be the final selling price
of the kit.  

Another design decision is if you are willing to trade any
capability away for "modifiability" or "experimenter
friendliness".  I think experimenters prefer the radio to
be built using technologies that are at most only slightly
more advanced than what they are familiar with.  So stay
with what was available in the 1990s  Same on the software
side.  It is best to remain in the mainstream.

Once you have these kinds of politically derived requirements
written down you can start engineering.  There is still a lot
of room for inovation even when constrained with decade old
tech and a $100 price point.  One idea I had as the use of
multiple ADC sections.  The radio would have a (say) 9Mhz
first IF that is sent to a high speed 8-bit ADC (these ADCs
are now dirt cheap) but the first IF is also sent to a
second mixer that takes it to baseband and then to a
sound card.  The display is built using the 8-bit data
but you listen to weak signals with the 24-bit sound card
data.  The 8-bit section could also work for wide band
signals like ATV.  

I'd prefer using Ethernet and UDP to USB cable simply because
this totally eliminates the need for a driver on the computer.
This means you don't have to maintain a driver for every
verson of every OS.
It allows the device to be used over the Internet possibly
enabling some novel uses.  With Eithernet one could eliminate
a long feed line and put the transciever outdoors in a
water proof box.

That said, I think engineering decisions are secondary.  The
important points are social and economic.  The "right" device
is the one that brings developers and experimenters on board
even if that device is not state of the art.



Chris Albertson
  Home:   310-376-1029  chrisalbertson90278 at yahoo.com
  Office: 310-336-5189  Christopher.J.Albertson at aero.org
  KG6OMK


       
____________________________________________________________________________________
Need a vacation? Get great deals
to amazing places on Yahoo! Travel.
http://travel.yahoo.com/

 1191442991.0