U.S. patent number 3,868,694 [Application Number 05/386,924] was granted by the patent office on 1975-02-25 for dielectric directional antenna.
This patent grant is currently assigned to The United States of America as represented by the Secretary of the Air. Invention is credited to Hans H. Meinke.
United States Patent |
3,868,694 |
Meinke |
February 25, 1975 |
DIELECTRIC DIRECTIONAL ANTENNA
Abstract
A dielectric directional antenna using a wedge shaped dielectric
with conducting exciters on each of the angular sides. One or both
of the exciters can be triangular with side indentations and one
exciter can be a ground conductor with a coaxial line feeding the
exciters. A three-terminal amplifier can be connected to the
conductors and the feed line.
Inventors: |
Meinke; Hans H. (Gauting
Bavaria, DT) |
Assignee: |
The United States of America as
represented by the Secretary of the Air (Washington,
DC)
|
Family
ID: |
23527649 |
Appl.
No.: |
05/386,924 |
Filed: |
August 9, 1973 |
Current U.S.
Class: |
343/753; 343/701;
343/785 |
Current CPC
Class: |
H01Q
23/00 (20130101); H01Q 9/40 (20130101); H01Q
9/0471 (20130101); H01Q 13/20 (20130101) |
Current International
Class: |
H01Q
9/40 (20060101); H01Q 23/00 (20060101); H01Q
13/20 (20060101); H01Q 9/04 (20060101); H01q
019/06 () |
Field of
Search: |
;343/753,754,755,783,785,701 |
References Cited
[Referenced By]
U.S. Patent Documents
Primary Examiner: Lieberman; Eli
Attorney, Agent or Firm: Herbert, Jr.; Harry A.
Claims
What is claimed is:
1. A directional antenna comprising:
a. a wedge-shaped dielectric member having a pair of surfaces
inclined to each other and joined at an intersection line;
b. a first conducting surface mounted on one of the pair of
dielectric surfaces and having the general shape of a triangle;
c. a second conducting surface mounted on and extending beyond the
other of the pair of dielectric surfaces with the first and second
conducting surfaces forming a line of constant characteristic
impedance; and
d. means for feeding the antenna, the feeding means being connected
to the apex of the triangle of the first conducting surface and on
the second conducting surface in the immediate vicinity of the
intersection line.
2. A directional antenna according to claim 1 wherein one of the
conducting surfaces is at least a quarter of a wavelength long.
3. A directional antenna according to claim 1 wherein the
characteristic impedance formed by the pair of conducting surfaces
is equal to the characteristic impedance of the feeding means.
4. A directional antenna according to claim 1 wherein the wedge
angle of the dielectric member is equal to arc cos c.sub.o /v.sub.p
.sqroot..epsilon..sub.r where c.sub.o is the speed of light,
v.sub.p is the phase velocity, and .epsilon..sub.r is the relative
dielectric constant.
5. A directional antenna according to claim 1 which further
comprises a three-terminal amplifier with one terminal connected to
the feed point of one conductor, one to the feed point of the other
conductor, and one to the inner conductor of the coaxial line.
6. A directional antenna according to claim 5 wherein the triangle
has indentations at the two sides that form the apex thereof.
7. A directional antenna according to claim 1 which further
comprises a pair of flat conductors mounted on the dielectric
member adjacent to the triangle but spaced therefrom and extending
to the intersection line and connected to the extended conducting
surface, the triangle and the flat conductor forming slots to
resonate in the operating frequency range.
8. A directional antenna according to claim 7 wherein the surface
of the wedge opposite the angle of the intersection line includes
angular indentations.
Description
BACKGROUND OF THE INVENTION
This invention relates to directional antennas, and more
particularly to dielectric antennas having a uniform dielectric
constant.
A dielectric directional antenna is an antenna whose directional
effect is produced by the incorporation of one or more dielectric
members in its near field. If several dielectric members are
present in the antenna, then the different members can have
different dielectric constants. An example of such a directional
effect produced by dielectric members are lenses and lens
combinations, but a true lens effect in the sense of geometric
optics occurs only when the dimensions of the lenses are a large
multiple of the wavelength. However, in the present invention, the
dimensions of the dielectric members are on the order of magnitude
of a wavelength (between a quarter-wavelength and a few multiples
of the wavelength) and consequently the rules of geometric optics
no longer apply.
When a directional effect is to be produced by means of relatively
small dielectric members, a wave field is generated in the
dielectric by means of an exciter or primary radiator and the wave
field is radiated from the dielectric into the surrounding air
space. The exciter can be located outside the dielectric in the
surrounding air, inside the dielectric, or on the surface of the
dielectric. Ordinary lenses are a well known case in which the
exciter is outside the dielectric. A disadvantage of such
excitation is that the waves emanating from the primary radiator
are partially reflected at the dielectric boundary surfaces.
Furthermore, with small dielectric directing members there is the
danger of pass-by radiation and therefore the useful rays contain
only a portion of the available power. The remaining power strays
around in the space and generates undesirable secondary radiation.
An exciter inside the dielectric is described by P. Mallach in
Fernmeldetechnische Zeitschrift, Vol. 2 (1949), pp. 33-39. The
mounting of an exciter and the replacement of defective exciters in
the interior of the dielectric without an air gap between the
exciter and dielectric results in engineering difficulties if the
exciter has a complicated shape. From an engineering viewpoint, the
simplest exciters are those which have a flat construction and are
pasted or pressed onto the surface of the dielectric. Such flat
exciters which are pasted onto the surface of a dielectric
hemisphere are described in G. Niedermair, Dielectric Spherical
Antenna, Dissertation, Technische Universitat Munchem, May 1971.
However, due to internal resonances of the dielectric member, this
spherical antenna has a very frequency-dependent directional effect
and large frequency ranges with pronounced secondary radiation,
which makes the spherical antenna unusable for many
applications.
The present invention overcomes the above-mentioned
disadvantages.
SUMMARY OF THE INVENTION
In the present invention the dielectric member and the flat exciter
are attached to the surface of the dielectric member in such a
manner that the frequency dependence of the directional effect of
the antenna becomes extremely small. For a given total volume of
the dielectric member, the focusing of the radiated energy around
the principal radiation direction is greatly reduced and
undesirable side lobes of the radiation pattern are kept very small
in the largest possible frequency range. A very useful criterion
for the fulfillment of these requirements is that the antenna has
no appreciable backward radiation in a very large frequency range,
i.e., no appreciable radiation in a direction which is opposed to
the principal radiation direction.
It is therefore an object of this invention to provide a novel
directional antenna.
It is another object to provide a dielectric antenna that is
independent of frequency.
It is still another object to provide a dielectric antenna that has
limited secondary radiation.
These and other objects, features and advantages of the invention
will become more apparent from the following description taken in
connection with the illustrative embodiments in the accompanying
drawings.
DESCRIPTION OF DRAWINGS
FIG. 1 is a diagram of a wedge-shaped dielectric with a pair of
exciters;
FIG. 2 is a diagram of a nonsymmetrical antenna having a
wedge-shaped dielectric with different exciters;
FIG. 3 is a diagram of the wave condition along the exciter line of
that shown in FIG. 2;
FIG. 4 is a diagram showing how the antenna of FIG. 2 is fed with a
coaxial line;
FIG. 5 is a diagram showing the antenna feed system of FIG. 4
integrated with an amplifying device;
FIGS. 6 and 7 are diagrams showing variations in the shapes of the
conductor or exciter;
FIGS. 8 and 9 are diagrams showing the use of additional conductors
on the surface of the dielectric to form parallel capacitances;
and
FIGS. 10-11 are diagrams showing a technique for a low reflection
front as a termination for the dielectric antennas of FIG. 3 by
adding additional wedges.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
Referring to FIG. 1, dielectric member 3 has a wedge shape in the
vicinity of the feed points a and b of the two flat exciters 1 and
2. Two approximately or exactly plane boundary surfaces 4 and 5 are
inclined at an angle with respect to each other. The two flat
exciters 1 and 2 are mounted to these two surfaces. The surfaces 4
and 5 intersect approximately or exactly in a straight line 6. At
least one of the two exciters has a triangular apex a or b which
lies in the immediate vicinity of the intersection line 6 and is a
connecting terminal of the antenna. In a symmetrically fed antenna,
the antenna has a symmetrical construction in which the two
exciters have the same size, shape, and symmetrical positions on
the dielectric wedge. In nonsymmetrical antennas, the exciters 1
and 2 differ. For example, with coaxial feeding as shown in FIG. 2,
the exciter 2 can be a large conducting surface and can project
over the boundaries of the dielectric wedge, such as in the
conducting surface of an aircraft. The shape of the boundary
surfaces 4 and 5 of the dielectric can very greatly. However, in
the present invention it is preferred that the two exciters do not
both project over their corresponding boundary surfaces of the
dielectric.
A symmetrical antenna can be produced by symmetrically combining
two equal nonsymmetrical antennas with a conducting base surface 2
as shown in FIG. 2 and then neglecting the intermediate conducting
base surface. Therefore, only the nonsymmetrical form of FIG. 2 is
described in the following, it being assumed that the conductor 2
is a very large plane surface and that the principal radiation
direction is directed along this plane.
If the antenna is to have less frequency-dependent characteristics
in a larger frequency range, then there must be less
frequency-dependent variation of the currents on the exciter
surface. Such a current variation is obtained if the two exciter
surfaces together form a line of nearly constant characteristic
impedance, called an exciter line. With a wedge shaped dielectric
as in FIG. 2, exciter 1 should have approximately a triangular
shape with the antenna fed between the conducting base surface and
the triangular apex a which lies in the immediate vicinity of the
intersection 6. The above-mentioned exciter line is a dissipative
line, since along its entire length it gives off energy to a wave
field which is built up in the dielectric and from there is
radiated into the surrounding space. If the exciter line is given a
minimum length of a quarter-wavelength and the described line
losses are large enough, a nearly progressive wave whose shape is
only slightly frequency dependent travels on this line in a
direction away from the feed point a. The input impedance of the
exciter at the feed point is then nearly equal to the
characteristic impedance of the exciter line, independent of
frequency. For example, if the characteristic impedance of the
exciter line is made equal to the characteristic impedance of
coaxial feed line 7 of the antenna shown in FIG. 4, the feed line
will have a nearly matched termination in a large frequency range
by the connected exciter system.
The special effect of an exciter line with a progressive wave which
is applied directly to the dielectric is explained in the
following. It is sufficient to consider the effect of a
transmitting antenna, since the wave processes of a receiving
antenna obey the reciprocity law and travel in the same manner but
in the opposite direction. A generator connected between the feed
point a and the surface 2 in FIG. 2 generates between exciters 1
and 2 a wave which travels along the exciter line. The wave
condition along line 1 is illustrated schematically in FIG. 3. The
wave along line 1 travels in air above the line and in the
dielectric below the line. In the immediate vicinity of the line
there is formed a near-field wave with a phase velocity v.sub.p
which, in accordance with the mixed dielectric, lies between the
phase velocity of a pure air-wave (velocity of light c.sub.o) and
the phase velocity c.sub.o /.sqroot..epsilon..sub.r of a wave
completely in the dielectric (.epsilon..sub.r is the relative
dielectric constant). Then at a certain distance from the line 1,
the wave in air and the wave in the dielectric is very different
and no longer has the form of the known TEM conduction waves. In
the air space, this wave is delayed relative to its natural
velocity, which would be the velocity of light. Such delayed waves
have the nature of surface waves in which all field strengths
decrease approximately exponentially with increasing distance from
the conductor. FIG. 3 shows the characteristic shape of the
electrical field lines E of such surface waves in air. See, for
example, H. Meinke, Introduction to Electrical Engineering at
Higher Frequencies, Vol. 2, FIG. 27, Springer-Verlag, Berlin, 1966.
The power radiated directly from the antenna conductor into the
surrounding air space is very small. The delaying effect of the
conduction wave is the reason for the low lateral and backward
radiation of the directional antenna, according to the invention,
which occurs almost independently of frequency if faults due to
internal reflections in the dielectric, which will be described
later, are avoided.
Furthermore, because v.sub.p > c.sub.o /.sqroot..epsilon..sub.r
the near field wave of the conductor 1 is faster inside the
dielectric than a plane wave completely in the dielectric. The
result is that in the dielectric there is formed a wave in which
the preferred direction of energy movement is directed obliquely
into the dielectric. This oblique wave direction results for a
conductor segment shown in FIG. 3. The phase fronts of the wave in
the dielectric always travel with the velocity c.sub.o
/.sqroot..epsilon..sub.r. The phase velocity in the direction of
conductor 1 is v.sub.p and, together with the velocity c.sub.o
/.sqroot..epsilon..sub.r forms the illustrated right triangle with
the angle .alpha., where
c.sub.o /.sqroot. .sub.r = v.sub.p . cos .alpha. (1)
or
cos .alpha. = c.sub.o /v.sub.p . .sqroot..sub.r (2)
Thus, such an exciter pressed onto the surface of a dielectric
member is capable of radiating the energy of its conduction wave
obliquely into the dielectric and therefore is specially suited for
the transmission of wave energy from a conduction wave into a
dielectric member. Accordingly, the exciter line has high radiation
damping so that reflections at the end of the exciter line are
hardly detectable even with short lines and the input impedance of
the exciter line is equal to the characteristic impedance of the
exciter line.
It is especially desirable if the wave traveling in the dielectric
of FIG. 3 is already traveling in the principal radiation direction
of the antenna, since then this wave need undergo no further change
of its direction. With the nonsymmetrical antenna of FIG. 2 the
dielectric should travel parallel to the conductor 2, i.e., the
wedge angle between conductor 1 and conductor 2 should be
approximately equal to the angle a given by Equation (2). The
desirability of this wedge angle has been confirmed by measurements
on antennas.
In FIG. 4 it is shown how the antenna is fed by the conducting base
surface 2 with a coaxial line 7, wherein the inner conductor of the
feed line is connected to point a of the exciter 1 and the outer
conductor is connected to the conducting base surface 2. If the
antenna is to be used for a very large frequency range, the
characteristic impedance of the exciter line will thus be made
equal to the characteristic impedance of the feed line so that the
feed line will have wide band matching to the imput impedance of
the antenna.
The dielectric antenna can be integrated with an amplifying
electronic circuit having a three-terminal character. A
three-terminal electronic circuit is illustrated symbollically in
FIG. 5 by three-terminal transistor T. The first terminal of the
transistor is connected to the exciter 1, the second terminal to
the base surface 2, and the third terminal to the inner conductor
of feed line 7.
It is desirable to have the components required for the electronic
circuit and its frequency band limitation mounted wholly or
partially on the surfaces of the dielectric. A frequency band
limitation is often necessary in order to prevent the radiation of
undesired harmonics during utilization as a receiving antenna.
Furthermore, it is known that optimization of an active antenna,
e.g., in regard to noise matching for receiving, is possible only
in a system with a bandpass filter action of the antenna.
Therefore, in many applications the input impedance of the antenna
in the complex impedance plane must with increasing frequency
traverse a loop curve having the character of the input impedance
of a two-circuit resonance-band filter. The loop should be close to
a given optimal value Z.sub.opt throughout the entire operating
frequency range. In passive systems, this Z.sub.opt is the
characteristic impedance of the connected feed line and, in active
systems, the optimal impedance for the electronic circuit.
An antenna impedance with a loop structure of the previously
described type can be obtained by the addition of a circuit at the
base of exciter 1 and/or by introducing certain suitable
modifications of the impedance of the exciter itself, in departure
from the triangular basic form. Flat conductors of a special shape
are applied to the outer surfaces of the dielectric member. It is
well known that thin and/or meander-shaped elongated conductors act
as conductances, while larger flat conductors act as capacitances.
Variations of conductor 1 on the surface 4 can result, for example,
as shown in FIG. 6, by making conductor 1 somewhat thinner in the
vicinity of the feed point and somewhat thicker at the other end
than in the simple triangular shape of FIG. 2. Such conductor
shapes yield impedances which in certain frequency ranges are very
similar to those of a series-resonant circuit. FIG. 7 shows
triangular conductor 1 having a symmetrical indentation. The
resulting constriction of the current path in the center of
conductor 1 acts as a series-connected inductance and likewise
generates resonance effects in interaction with other wider parts
of the conductor 1 which therefore acts as a capacitor.
As shown in FIG. 8, the surfaces of the dielectric member can be
attached to conductors 8 and 8a which are connected to the
conducting base surface 2. Parallel capacitances arise between the
conductor 1 and these additional conductors in a filter circuit. If
the edges of the conductors 8 and 8a lie near conductor 1 then
slots result, as shown in FIG. 9, which can have a
position-dependent width and, if long enough, can have slot
resonances which have a strong effect on the impedance of conductor
1 and can contribute to the formation of defined bandfilter
effects.
When there is effective coupling between the exciter and the
dielectric, additional measures are essential in order to produce
between the wave field in the dielectric and the wave field in the
adjoining free space a transition for directional radiation but has
little frequency dependence.
The wave entering into the dielectric from the exciting conductor
as shown in FIG. 3 strikes the wall of the dielectric and is
partially reflected into the dielectric, partially exits from the
dielectric into the outside air, and also travels as a wave into
the surrounding space. Therefore, inside the dielectric there is a
complicated superposition of waves in many directions. The
dielectric member acts similarly to a cavity resonator but as a
resonator of low Q, since its outer walls are not a completely
reflecting metal but rather a more or less transparent boundary
between dielectric and air. The higher the dielectric constant of
the dielectric, the greater are the reflections and the higher is
the Q of the resonator. There are two different engineering
embodiments of the directivity-generating dielectric member; first,
antennas with small bandwidth in which a suitable resonance of the
dielectric member at the operating frequency is intentionally
introduced and the filtering action of this resonance is used, and
second, antennas for a broad frequency band in which resonances of
the dielectric are substantially eliminated.
In the wide band case, the wave generated in the dielectric as
shown in FIG. 3 and in the configuration of FIG. 2 should emerge
into the adjoining space with as little reflection as possible and
should be radiated. For this purpose, the dielectric wedge requires
a low-reflection front as termination. Therefore, in the wide band
case, this front must create a continuous transition between
dielectric and air. In the examples of FIGS. 10 and 11, this
transition is achieved by the addition of several small wedges 9.
The transition regions shown in FIGS. 10 and 11 can have
differently shaped outer boundaries. As shown in FIG. 11, they can
also be a direct continuation of the wedge of the feed zone.
* * * * *