From Peter Martinez G3PLX
Congratulations to Clint KA7OEI on the hardware implementations of
PSK31 for low-frequency work. Personally I think the hardware
implementations of PSK31 have been rather neglected. Perhaps this is
a side-effect of the software version being cost-free! Some
interesting applications of PSK31 have probably not yet seen the
light of day as a result. Clint's work is valuable in this respect.
Here's a few of my own ideas for exploring the PSK31 scene with
hardware for VLF and other areas:-
1. Transmitting "squarewave" PSK31, although it would be considered
very antisocial on HF radio, is a valid way of doing it when we are
not concerned with adjacent-channel QRM. I am thinking of underground
induction field communication at VLF, and other non-radio
applications where we are fighting the noise level and not the QRM. A
Class C power amplifier, designed more like a switchmode PSU than a
transmitter, fed with PSK31 generated with an XOR gate, will be ideal
for this task. Be careful that if this kind of transmitter is
connected to a very high Q VLF antenna, there can be disastrous
effects when the transmitter reverses phase and the high level of
stored energy in the antenna circuit is dumped back into the
transmitter! Check the reverse ENERGY capability of the transmitter
rather than the reverse POWER rating. This probably doesn't apply to
USA lowfer activity, but could apply to the European 136kHz
allocation where you might have a 1 kW transmitter to radiate the
permitted 1 watt e.r.p.
2. Note that if the transmitter is hard-limited in this way, the
receive filter in the current software PSK31 receive-side is not
quite optimum, since this is matched to the cosine envelope of the
software PSK31 transmit-side. Slightly better results (by which I
mean performance in Gaussian noise and not adjacent-channel
rejection) can be achieved with a filter matched to the square-wave
transmit envelope. The classic "Integrate and dump" filter (or the
equivalent using analogue delay lines) is right for this task, and is
quite easy to build in hardware. More good reasons to experiment with
simple hardware PSK31 receivers.
3. On HF the PSK31 receive signal is demodulated incoherently, by
comparing the phase of one symbol with the previous one. The
transmission is encoded differentially to match this, but we can gain
an extra 3dB of weak-signal performance if the path is stable enough
to let us demodulate coherently. We don't even need to reconfigure
the transmission coding to do this, since we can get the same answer
as differential incoherent demodulation by using coherent
demodulation followed by differential decoding. VLF applications,
whether radio or induction-field, are certainly stable enough. The
only penalty will be that it will take a little longer to lock in at
the start of a transmission, but we could easily switch between
incoherent and coherent if that's a problem. This is another area for
exploration with simple hardware demodulators.
4. Although QPSK is noticably better that BPSK on HF where there is
QSB and burst noise, when the limiting factor is Gaussian noise, BPSK
is better. This could mean that simple BPSK may be the best choice
for some non-HF applications of PSK31. The Lowfer/Medfer area might
suffer from atmospheric static which could mean QPSK had the
advantage there, but for underground reception of VLF, or laser
working, or perhaps even underwater acoustics, hard-limited BPSK
transmit and coherent integrate-and-dump would be my suggestion for
the best way to go.
73
Peter
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
to unsuscribe from the list send to [email protected] a message
with a text line as follow: unsubscribe psk31 or unsubscribe psk31-digest
More instructions on PSK31 Webpage: http://aintel.bi.ehu.es/psk31.html
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
