[Ground-station] UI for SDR

[Phil Karn]

On 6/30/18 16:57, Mike Parker wrote:

> This might be achieved in a variety of ways, depending on the details of the digital signal.  A pilot tone, an unmodulated CW carrier, or even an unmodulated digital sequence that can be detected with conventional spectral analysis comes to mind.

I was going to suggest this as soon as I saw you mention your problem.
Like a lot of people I'm planning to junk TV service on cable so I need
to put up some broadcast antennas. I don't expect much of a problem for
the channels whose towers I can visually see towers a few km away, but I
may have a problem with some of the other mountains that I can't see.

I've been interested in ATSC pilots as poor man's frequency references
so I'd already made a survey. Turns out they're NOT all locked to GPS,
which raises the question of which to trust. With one pilot carrier, I
knew what frequency I had. With two, I was no longer sure...

Anyway, I decided to collect some statistics in the hope that clustering
would reveal the accurate ones. This being southern California, I
figured I could hear a lot more with a narrowband receiver than I could
see with a TV. This turned out to be true; there was an audible pilot
carrier on just about every channel still used for TV broadcasting.
("Repacking" is further decreasing the number.) Because there's no ID on
a pilot, this might create some confusion as to which station you're
seeing unless you can get the azimuth at least somewhat close to start
with. And then you have the multipath and sidelobe problems you mentioned.

When I had DirecTV a long time ago I remember it being hard to align the
dish. The receiver had a pointing meter on the screen, which of course I
couldn't see from the roof. It did generate a tone I could hear through
the window, but it had a fairly long lag time and a nonlinear response
that still made it tedious. As we all know, any lag in the feedback path
of a servo loop tends to make it unstable...

Phil