[hpsdr] HPSDR Project Proposal - Spread Spectrum

[Cliff Chism]

A Project Proposal For Calypso - The Spread Spectrum Transceiver

I have recently been investigating FPGA and CPLD programming. As is usual,
it is difficult to really get into such an endeavour, without first having a
goal. Of course, understanding the HPSDR system, as it currently stands, is
one very good reason. But, that is more a matter of comprehending something
that has already been done. I wanted a project that would advance the radio
art. The one I came up with will be a major endeavour not only with respect
to programming, but also in RF theory. As the idea currently stands, it is
well beyond anything I've ever tried before (that's part of the challenge)
and in consultation with Phil Harman VK6APH, decided that the idea should be
presented to the members of the HPSDR list, to see if there might be group
interest in such an undertaking. I am almost positive that it can be done.
Whether or not, I am personally capable of such a task, remains to be seen.
But, with the help of the more than capable members of this group, I believe
it can be achieved.

Spread Spectrum has the distinction of being one of the few areas of amateur
communication that has not been embraced by the community, at large. A few
have made notable forays into the field (AMRAD performed a large number of
experiments in the 80's and 90's, resulting in the publication of the ARRL
Spread Spectrum Sourcebook, no longer being published, but still an
excellent, if dated reference). But, it has largely languished in obscurity,
I think, due mainly to its inherent complexity. If the means might be
perfected, or at the least made usable, the benefits to the amateur bands
could be staggering, especially considering the congestion that exists
today, sure to only increase in the coming years.

This idea is in its infancy. As I currently understand the situation, the
main areas of difficulty lie in acquisition and synchronization of the SS
signals. In order to break the problem down into individual goals, I have
decided to first concentrate on generating the pseudorandom noise sequence
via an FPGA algorithm. Step 2 would be to get a hardwired transmitter and
receiver to sync on the generated signal, and develop a method of forward
error correction. Step 3, Acquisition, appears to be the most difficult part
of the problem. I propose that an interim solution might be to use an
automated and conventional narrowband direct link between the transceivers,
synchronized via GPS timecode (ala Gibraltar, which I am told, is
progressing) and able to pass correction data between them, to establish the
parameters of the link and the PN sequence key, and then to quickly switch
to spread spectrum. This might be construed as cheating, but provides us
with a usable SS system, where experimentation might continue, in order to
develop a method where acquisition is possible "on the fly".

In very fruitful discussions with Phil Harman VK6APH, he suggested
using a short PN sequence for initial synchronization and programming
the FPGA to try all possible keys at once, in parallel. This could lead to
extremely fast synchronization of the signals, and cut
down on the "waiting period". Then, a longer key could be exchanged via the
established link, leading to an even more efficient use of bandwidth.

So, assuming that the idea is intriguing to a number of members, I am
proposing a transceiver, using Direct Sequence or Frequency Hopping Spread
Spectrum technology. Just about every other facet is subject to negotiation,
pending further review of technological and legal issues. The FCC has seen
fit to encourage experimentation on SS, but at the same time, has put a
myriad of obstacles in the way of experimentation. It remains to be seen how
much can be accomplished within those guidelines, and how amenable they
might be to relaxation of the rules, in the face of a workable and efficient
system. They have permitted variations, in the form of STA's to a few
experimenters, in the past.

For the time being, I am planning on using Ozy as a development platform for
the FPGA work, to establish the basic functions, and "doability" of the
idea. I will also have access to a Digilent Nexsys2 Spartan 3e FPGA
development
board (Thanks for the head's up, Phil C.) and Lyle's KK7P DSP module, for
further experimentation.

Please think of this proposal as a sounding board, to the group, to get a
sense of whether the interest might be there, for a project of this type.
Any constructive comments would be most welcome. I believe that the project
is doable. I believe that it will be a great challenge (especially for me,
personally). And, I believe that, if successful, it would lead to a great
advancement of the radio art.

73,

Cliff Chism - KE5FJV