|Subject:||Re: LF: phase shift compensation question - and the moon|
|From:||Markus Vester <[email protected]>|
|Date:||Fri, 3 Mar 2017 13:17:33 -0500|
|Dkim-signature:||v=1; a=rsa-sha256; c=relaxed/relaxed; d=mx.aol.com; s=20150623; t=1488565054; bh=6QMLGaMMSqE2yAxoIhXbpkXCZtTChSr67jBkcN+wCVk=; h=From:To:Subject:Message-Id:Date:MIME-Version:Content-Type; b=JWrNRAcRQkUmlI0bWNbK6htsof+u1aYmGJ9I/2muPIm1eyOYzMIlzKKl4x8HVYwOL R5k6nmHpnopLxHf50X3yTcSBK7kwGRkKiDICWa7fA99mH1tSMLmO7NtZu5l7yR1/gV PE8vNRQbp9I+uQZVDxcfm20zAFlcfqs50uH81+Kw=|
this is an interesting question.
At VLF frequencies and over long distances, diurnal phase patterns seem to be pretty regular and repeatable so they could be predicted by calculation. At higher frequencies and intermediate ranges, the situation is complicated by multipath / multimode propagation, producing more random phase and fading patterns. As you say, the phase effects could still be compensated using a pilot signal, provided that the frequency and the geometrical path are similar. But how close would they need to be "similar", or in other words, what is the temporal and spatial coherence width of a wavefront?
I have been observing Loran-C stations for many years, and I believe that these data could provide sonme insight. At night, the impulse response consists of a number of differently delayed skywave components. Most of the energy is contained within about a millisecond, indicating coherent fading patterns over frequencies deviations up to about a kHz. The question of spatial correlation is a bit harder as we have only sparse data from a limited number of locations. Until 2010, an interesting example were three stations in the Newfoundland chain (Fox Harbour, Comfort Cove and Cape Race, GRI 7930), which are a few hundred km apart and used to have very similar looking impulse responses: http://df6nm.bplaced.net/LoranView/LoranView_1002.htm . On the other hand, the signal from Caribou ~1000 km further away already showed a quite different pattern.
BTW unfortunately DCF39 would not be a good pilot, as the frequency drifts quite a bit, and the carrier is spread by phase jumps from the FSK telegrams. HGA22 seems to be intrinsically more stable (a few ppb), but it is also contaminated by the same modulation.
Another interesting application could be unspreading VHF EME signals. Weak signal reception over the moon could be much improved if long coherent integration times could be used. But this is limited to about a second by libration spreading, caused by the superposition of many scattering centers on the lunar surface. The complex sum is highly dependent on the view angle, which varies over time because TX and RX both move with earth's rotation, and also to a lesser degree due to the excentric and tilted lunar orbit.
However unlike ionospheric fading, the associated speckle pattern is rather deterministic, literally "carved in stone". So in principle if we had collected a database of a priori measurements of phase factors for all accessible aspect angles, we could predict and undo the phase evolution and then use arbitrarily long integration and small noise bandwidth. However this database would need to be quite large (~ 10^10 viewangles at 144 MHz).
On the other hand, noticing that earth is moving beneath a fixed speckle pattern, the same fading pattern would be repeated to an observer further west some minutes or hours later. Thus we could possibly use a CW signal from a big gun as a pilot, and then apply a delayed version of the conjugate phase to receive a weak station located on the same latitude but further west.
would it be possible to use the phase shift from a strong VLF signal to
Von: Jacek Lipkowski <[email protected]>
An: rsgb_lf_group <[email protected]>
Verschickt: Fr, 3. Mrz 2017 15:12
Betreff: LF: phase shift compensation question
compensate the phase shift for an amateur VLF transmitter? for longer
transmissions (when the ionosphere height changes etc) over big distances
could also work for LF, like receiving DCF39 (which is probably very
stable, even if it's not intended for timing) on 139kHz in Australia to
measure the phase shift of the path from Germany, so that we can
compensate the changing phase shift of an amateur transmission on 137kHz,
so that we could use a long transmission time in ebnaut.
would that work? and would it be worth it?
Jacek / SQ5BPF
|<Prev in Thread]||Current Thread||[Next in Thread>|