[Antennas] Omnidirectional?

2004年4月25日 10:15:23 -0400

When people say "omnidirectional", they usually mean "no directivity in
azimuth". That is, stations on the horizon at all compass bearings will receive
the same power. As folks point out, you can get more gain towards the horizon
by sacrificing gain upward or downward. That's what colinear arrays do.
More-than-you-wanted-to-know department:
Antenna theorists know that a true omnidirectional (isotropic) linearly
polarized antenna is physically impossible. You can't build a practical antenna
that radiates into 4 pi steradians (all directions) equally, even in principle.
 That's a result of Maxwell's equations and some math. A continuous transverse
vector field on the surface of a sphere cannot have constant magnitude, which
can be verified (?) by looking at the hairs on a coconut. (Remove outer shell
first. Watch your cholesterol.)
This was known in 1909. For elliptical polarization, there is some wiggle room,
to coin a phrase. A neat paper (warning: math & physics) is at
http://puhep1.princeton.edu/~mcdonald/examples/isorad.pdf .
None of this stops you from quoting gain in dB over an isotropic antenna (dBi),
as if there were such a thing.
73, Martin