DSP Radios

PaulKB8N at aol.com PaulKB8N at aol.com
Sun Aug 20 18:33:53 EDT 1995

Doug's comments on "ham radio on a card" bring up a couple of very
interesting and germane issues.  The first is that of the concept of the
"virtual transceiver" and what level of acceptance it will have in the
community.  The USAF developed a prototype "virtual receiver" for its HF
sites a couple years ago.  This was not a receiver on a card as Doug
describes, but rather total computer control of the WJ and Racal military
receivers.  The computer screen displayed the receiver and the operator was
able to access the controls through the use of touchscreen and mouse inputs.

The operators who tested this concept did not like it!  At the very minimum,
they wanted the tactile feel of a tuning knob and the responsiveness of at
least minimal physical controls.  As the "hamcard" concept evolves, I believe
that there will still be a necessity for some type of physical controls,
probably something that sits next to the keyboard and is tied to the
computer.  It seems obvious to me that commercial DSP available on
inexpensive sound cards will be the audio processor of choice in this scheme.

The thought of generating RF (even at low levels) and bits and bytes within
the same box (though arguably that is happening already under controlled
circumstances) makes me nervous.  The opportunity for undesireable byproducts
would be dependent on the design of the motherboard and peripherals of the
computer, in conjunction with the "hamcard" design.  I don't believe that
there would be universal compatibility and only the best engineered systems
and bus designs would work.

I believe that the future instead lies in the "hamcube" concept.  This would
be a tower mounted, heat-sinked box that would contain the transceiver and PA
connected to the shack by fiber optics.  Certainly some of the noise and
signal losses that exist in our receivers today are the result of long coax
runs, coax switches and relays, and proximate rotor and other control cables.
 A single fiber could control "hamcubes" at several locations, along with all
the rotor and antenna switching controls.   The antenna field could be miles
from the shack, with no perceptable degradation of performance. Whatever
noise there is with this arrangement could be easily handled within the DSP
soundcard of the computer.  The soundcard stereo output with the ability to
mix inputs from several sources would provide great flexibility for the
single op.

My take on the future of DSP radios is that fiber optics holds the key to
reduced S/N ratios, that DSP soundcards will become the noise reduction
system of choice, that the "hamcard" concept (though interesting) still has
some limitations that fiber can overcome with a remoted "hamcube" system, and
that some minimum number of "real" tactile controls in this otherwise virtual
domain will be needed by the operator to make this system work.
     73, Paul KB8N

More information about the CQ-Contest mailing list