I fixed vtmatch.
I passed last night's signal through vtmatch and got
a fairly decent set of spikes. I noticed a spike
at 19:30 which turned out to be an un-announced
VO1NA transmission which I decoded
19:30 3.8 dB
I won't release the new vtmatch yet because there's
still a little problem with it.
What are we testing tonight? The phase appears not
steady enough for much longer duration messages.
The signal is too strong for longer symbols.
We could try shorter symbols, say 0.5 seconds
and more characters, perhaps with a stronger code.
It would be nice to try
coding: 16K21A
symbols: 0.5 seconds
characters: 30
duration: 1728 seconds, 00:28:48.00
which is about 6dB weaker than we've been decoding.
Unfortunately that's too big to decode on 32 bit Windows.
It needs about 3.5 GByte of RAM and a 64 bit system.
Alternatively, for about the same signal strength
coding: 8K19A
symbols: 1.0 seconds
characters: 30
duration: 1712 seconds, 00:28:32.00
which should decode on 32 bit Windows but the code
is weaker.
Joe wrote:
> Wonder if there is a chance of a VK7 reception
> if QRO and longer symbols.
Also needs to maintain phase for longer. Might need
some work on the oscillator, or are we limited
by path stability?
--
Paul Nicholson
--
|
