On Tue, 16 Aug 2011 17:13:15 -0700, Jim Lux
wrote:
On 8/16/2011 3:16 PM, Dave Platt wrote:
In ,
Jeff wrote:
With the cardioid pattern, there's no ambiguity. With a deep null,
and a non-reflective environment, it's easy to use. However,
reflections off buildings and hills are a problem. The antenna is
roughly an omnidirectional antenna, and reflections will appear to
fill in the deep null. Not much can be done about that except to move
and try a different location. My favored method is by using a map.
Find a position, determine a bearing, and draw a line of position on
the map. Then, find a different location, and do it again. Repeat as
often as practical generating as many LOP's as possible.
I have seen recommendations that one do this by proceeding in a
straight line, taking bearings at positions of around 1/4 wavelength
or so. The effects of multipath will tend to cause the measured
bearing to wobble back and forth across an arc (as you go down the
line) and the true bearing will tend to be close to the center of that
arc.
I haven't tried this approach myself... but it might combine well with
your "take readings from lots of different locations" approach, by
helping cancel out some of the effects of multipath.
There's a photograph of this technique on a field with little flags at
different distances all oscillating around a single line illustrating it
in some ARRL publication (probably an antenna compendium, but maybe the
handbook or antenna book).
I think the photo was on interferometry for direction finding, where
they were trying to plot the wavefront pattern. I recall the photo,
but I couldn't find it in the ARRL Antenna Handbook (19th edition).
However, the handbook does have a full chapter on direction finding
(Ch 14) which includes the cardioid antenna.
The purpose of me using multiple bearings is simply to eliminate the
effects of reflections. It's not totally foolproof, but better than
chasing a single bearing.
There are better ways, but they're usually not very portable. A
rotating antenna direction finder can easily distinguish between the
incident signal from a reflection. Such direction finders display
something like the antenna pattern on a polar display. The incident
signal is very steady and does not move in azimuth. Reflections
jitter dramatically in both amplitude and azimuth. They also tend to
appear and disappear rapidly and erratically.
Rotating antenna DF antennas:
http://www.rockwellcollins.com/sitecore/content/Data/Products/EW_and_Intelligence/SIGINT/ANT-1040A_Airborne_Spinning_DF_Antenna.aspx
http://www.rockwellcollins.com/sitecore/content/Data/Products/EW_and_Intelligence/SIGINT/ANT-1040_Spinning_DF_Antenna.aspx
20 in diameter, 25 in high, 50 lbs.
Another method that I've used is to generate a narrow "beam" using two
identical yagi antennas that share a common reflector. The antennas
are oriented about 20 degrees apart with the reflector at the vertex.
A near perfect audible square wave drives two identical PIN diode
switches on the driven elements. While the antenna may have a -3dB
beamwidth of perhaps 10 degrees, the LOP of equal antenna signal
levels is extremely sharp, and often less than 0.5 degrees. A
synchronous demodulator, charge pump, differential amp, and zero
center meter complete the systems. You can also do it by ear by
listening to the null at the switching frequency. If the antennas and
PIN diode switches are perfectly symmetical, all boresight error
cancel. I originally contrived the system by reading about the Lorenz
beam bombing system used by the German's during WWII, but later
discovered that others had anticipated the idea.
--
Jeff Liebermann
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060
http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558