Klaus von der Heide wrote:

Hello LF-Friends,

the negative result of the recent transatlantic experiment only says
that the usual ham methods for LF contacts are not adequate for a
transatlantic distance. A weak signal never excludes information
transmission, it only reduces the information bit rate. Why not try,
as a first step, to get one single bit over the ocean?
The communication theory says that BPSK is optimum, and a
bandwidth considerably larger than the information bit rate is better
than a small one.

I therefore, propose the following experiment:

1. Both, transmitter and receiver, must be synchronized to an
    atomic clock, i.e. all oscillators that determine the carrier
    or the symbol rate. Especially the sampling frequency of
    a DSP (not the processor clock) must be synchronized.
    A soundcard normally cannot.

2. The symbol rate is exactly  1 bit/s.  Starting every minute,
    a constant random pattern of 60 bits is sent out.
    Appropriate matched filters must be used at both ends.

3. At the receiving end, a DSP adds the 60 new  values to
    60 accumulators.

4. The contents of the accumulators is correlated with the
    known random bit pattern. After many hours or days
    (or years?) the correlation must become significant.

5. It is important to suppress the non-Gaussian noise as
    good as possible before the data are accumulated.

As I mentioned some weeks ago, phased array antennas
at both ends would increase the signal by many dBs.
There isn't any dought that hams can cross information
over the atlantic. The minimum information of a ham QSO
is 50 bits in either direction. The open questions only is:
Is the possible information bit rate  1 bit per day  or is it
1 bit per 5 minutes  or in other words: can a QSO run
within 10 hours or can it definitely not.

73 de Klaus, DJ5HG

Good idea Klaus. let me know when you want to start. But would it not be
more simple to send some slow dashes on a specific frequency . Each
amateur participating would have a time slot.
Amateur 1 on the hour commences dashes of an agreed on/off period for 5
minutes. Then a silent period of 5 minutes to avoid confusion with the
next participant.
Amateur 2 starts at H+10 for 5 minutes then 5 minutes silence
Amateur 3     "         H+20 for 5 minutes then 5 minutes silence etc

Time needs to be fairly accurate but no atomic standard is necessary and
all would conform to the same dash duration and off period.
Callsign at normal speed for ID purposes sent immediately after the 5 min
acty period.
Keep it simple.
de G3KEV