LF: LF band plan]
Michael F. Skelly
Sat, 30 Jan 1999 14:33:52 -0500
The digitization and retransmission of the received content of LF
signals is the subject of one of my recurring daydreams. It's not
yet well-conceived, but my thinking is this:
1. The accuracy and precision of a received LF signal probably
require 8 to 16 bits representation for reasonable retrieval and
2. The wavelength of the signals prevents much directionality of
antennas on residential properties.
3. The limits to sensitive reception are primarily local noise and
nearby broadcast signals.
4. It may be possibly to cancel some local noise at individual
receiving sites and to achieve directionality by using well-
separated receiving sites in an antenna array. The dimensions
of the separation would be on the scale of a half wavelength or
better. (Some of the FCC and military HF phased arrays have
diameters on the order of 100-2000 meters, with up to 48 antenna
sites. One claimed to be able to locate a transmitter by
triangulation within 1 km on the opposite side of the earth. One
never knows whether to believe military braggadocio.)
5. By combing the digital representations of the analog LF signals,
and applying some relatively straightforward deconvolution algebra,
it ought to be possible to abstract directionality from the phase
lag components, and to cancel local noise from individual sites.
Additionally, variations in receptionsensitivity at different sites
could be treated as terms in the deconvolution.
6. Personally, I've never cared much for Fourier operations in
deconvolution, but this seems to be a good place for them. It's
fortunate that those Motoroola boards already have that available.
A convenient way to share the digitizations may be as their
Fourier transforms. From there, the deconvolution (including
filtering as well as signal abstraction) could be done with a
variety of available tools.
7. At this point, I wake from my daydream.