Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
2027~56
--1
SAT~T.T.TTE BEAM-FORMING NETWORK SYSTEM
HAVING IMPROVED BEAM SHAPING
RACK~DOUND OF THE lNV~.llON
1. Field of the Invention
This invention relates to an antenna for satellites
to communicate with ground stations and, more particularly, to
an antenna system for satellites incorporating an antenna array
for producing a communicating beam which has improved side lobe
characteristics around the fringes of the ground area covered by
the antenna.
2. Description of the Related Art
Satellites are now employed for providing
communication, such as land mobile telephone service, between
distant points on the surface of the earth. One embodiment of
such a system of considerable interest is that in which the
satellite travels in a geostationary orbit about the earth. The
satellite may be located at a fixed position above the United
States and carries an antenna having a sufficient beam width in
the north-south direction and in the east-west direction to
permit the reception and transmission of communication signals
between any two points in the United States. The beam width in
the north-south direction can be enlarged to include both United
States and Canada, if desired. A beam width of approximately
4.5 in the north-south direction is sufficient to cover both
Canada and the United States. The beam width in the east-west
~i_
_ -2- 2027~56
direction should be approximately 8 to provide the desired
coverage. A problem arises in that the use of an antenna having
the foregoing beam width in the north-south and east-west
directions has less signal gain than is desired. This
necessitates larger power amplifiers for driving radiating
elements of the antenna.
In previous satellite communication systems, such a
wide beam width antenna has employed at least two overlapping
beams to provide the coverage. The generation of such beams
with a desired overlap until very recently required the use of
separate large reflectors each having a diameter of about 16
feet. In the construction of communication satellites, however,
it is desirable to reduce physical sizes, weights, and power
requirements to facilitate the construction and launching of
such satellites.
There has previously been disclosed
a system for communicating via satellite between
disclosed a system for communicating via satellite between
ground stations. The system comprises a set of ground stations
spaced apart along an arc of the earth's surface and a satellite
positioned above the earth in view of the arc. An array of
radiating elements is deployed on the satellite, and a frequency
responsive beam former connected to the radiating elements is
provided for forming a beam of electromagnetic radiation. The
beam is steerable in response to a carrier frequency of the
radiation to intercept individual ones of the stations in
seriatim. The frequencies of an uplink carrier and of a
downlink carrier respectively associated with respective ones of
the ground stations vary monotonically with position along the
arc to permit automatic positioning of a beam from the satellite
to a ground station upon energization of a carrier frequency
assigned to the ground station.
-
~3- 2027456
So that the present invention may be better
understood, the satellite communication system disclosed and
claimed in the aforementioned application will now be discussed
in some detail by reference to Figs. 1 through 5. As shown in
Figs. 1 and 2, the satellite 24 employs a simplified antenna
structure 30 comprised of two confocal parabolic reflectors, one
of which is a large main reflector 32 and one of which is a
small subreflector 34, and a 4 x 2 array 40 of eight radiating
elements 42, all of which are supported by a frame 44. A front
10 view of the array 40 is shown in Fig. 2. The array 40 of
radiators 42 is rigidly secured in front of the subreflector 34,
and the subreflector is located within the satellite 24. The
main reflector 32 is substantially larger than the subreflector
34 and due to the larger size is folded during launch and
15 subsequently unfurled when the satellite or spacecraft 24 has
been placed in orbit. Upon being unfurled, it extends outside
of the satellite 24 as shown. Also shown in Fig. 1 within the
frame 44 is other spacecraft equipment such as rocket engines
and fuel tanks, thereby to demonstrate that the antenna system
20 30 can be easily carried by the satellite 24.
The arrangement of the components of the antenna
system 30 provides a significant reduction in weight and
complexity for a satellite antenna over that which has been
employed before. This is accomplished by fabricating the main
25 reflector 32 and the subreflector 34 with parabolic reflecting
surfaces, the two surfaces being oriented as a set of confocal
parabolas having a common focal plane or point 48. Such
configuration of reflecting surfaces in an antenna is described
in C. Dragone and M. Gan~, "Imaging Reflector Arrangements to
30 Form a Scanning Beam Using a Small Array," Bell System Technical
Journal, Vol. 58, No. 2 (Feb. 1979), pp. 501-515. The
configuration provides a magnification of the effective aperture
of an array of radiating elements. In the preferred
configurations as shown in Figs. 1 and 2, the magnification
~~ _4_ 2027~5~
factor is 4.7. The eight radiating elements 42 of the array 40
represent a substantial reduction in complexity of the antenna
since, if a direct radiator of similar sized elements had been
employed, a total of 155 radiating elements would have been
needed to give the same antenna performance. As shown in Fig.
3, a hexagonally arranged antenna array 50 of seven primary
radiators 52 may be used if desired in place of the 4 x 2 array
of radiators mentioned above. The array 50 of feed elements 52
may be employed for both uplink and downlink communications.
Fig. 4 illustrates two exemplary spot beams 56, 58
produced by the satellite 24 (not shown in Fig. 4) in
geosynchronous orbit above the earth 60. Spot beam 56 extends
substantially along the eastern coast of the United States 62
and Canada 64, while spot beam 58 extends substantially along
the western coast of the United States 62 and Canada 64. The
satellite transmits and receives information-carrying radiation
to and from ground stations located within regions of the
earth's surface encompassed by the respective first and second
spot beams 56, 58. The coverage patterns of the respective spot
beams 56, 58 preferably are selected such that frequency bands
available for communications are concentrated in regions of the
surface of the earth 60 where the largest communications
capacity is necessary to optimize antenna gain usage by
substantially limiting the amount of antenna gain which is
incident upon regions wherein relatively little communications
capacity is necessary, such as in sparsely populated regions.
The antenna system of satellite 24 provide a
one-dimensional beam scan (which may be considered to be a
continuum of virtual spot beams) across the surface of the earth
60. Such a scan can be directed along an arc of the earth's
surface, such as a longitude or a latitude, or an arc included
relative to a latitude. The scanning can be accomplished most
efficiently for the geography depicted in Fig. 4 by scanning in
the east-west direction, providing a scan path which follows an
~5~ 2027~56
arc of a great circle of the earth. The sc~nning is preferably
implemented by using fixed delays (as will be described
hereinafter) among radiating elements of the antenna system and
by employing different frequencies for different geographical
locations on the surface of the earth. Thereby, the scanning is
accomplished by variation of the frequency of the radiation for
each position of the beam scan (i.e., for each virtual beam),
and in addition, a plurality of the beams (not shown) can be
generated simultaneously by the provision of different
frequencies of electromagnetic radiation in each of the beams.
By use of this virtual beam technique, users at any point within
the coverage of the beam scan are close to the center of one of
the virtual beams. Therefore, users will typically receive 2 or
3 dB more power than they would from a comparable satellite
using fixed beams.
To minimize the required electromagnetic power and
provide for simplicity of antenna structure, the preferred
antenna system provides beams with a generally circular
cross-section and a width of 4.5 by use of the hexagonal array
50 of radiating elements 52 as shown in Fig. 3. The elements 52
preferably are cup dipole feed horns one wavelength in diameter.
As an example of its use, the satellite
communications system may be designated for land mobile
telephone service, sometimes referred to as the Mobile Satellite
(MSAT) system. Two frequency bands are assigned for such
service: 866-870 MHz for the downlink band and 821-825 MHz for
the uplink band. The 4 MHz width of each of these bands may be
subdivided into approximately 1000 frequency slots which are
individually asæignable to individual ground station~ on the
surface of the earth 60 for compounded single sideband voice
communication. Other frequency bands may be utilized, for
example, such as the L-band. If the stations were uniformly
positioned from east to west, with each station being at a
different longitude, approximately twelve assignable channels
-6- 2027456
comprising an uplink and a downlink would be available within a
scan angle of approximately 0.1 degree.
Since the channels would be uniformly spaced apart
in frequency, a beam would be uniformly stepped in the east-west
direction as the downlink (or uplink) frequency was shifted from
one channel to the next channel. In other words, the operating
frequency of the ground station is preferably selected to match
the frequency of a beam directed from the satellite to the
ground station. For a uniform distribution of the stations in
the east-west direction, the beam could be centered with respect
to the east-west component thereof upon each of the stations.
However, as a practical matter, the stations tend to be
clustered in various geographic areas of the United States 62
and Canada 64 providing a nonuniform distribution of the
stations along the east-west scanning path of the beam.
Consequently, a peak signal amplitude cannot be obtained for all
of the stations.
By way of example, assuming that 25 ground stations
are located within a scan angle of 0.1, the corresponding
reduction from peak signal amplitude is less than 0.01 dB
(decibels). This represents a significant impLov~ t over
previously available satellite communication systems employing
separate fixed beams wherein the average loss in signal gain
relative to peak signal gain in the east-west direction was
approximately 0.8 dB~ As noted above, such previous satellite
communication systems employed antenna systems having a
plurality of large antenna reflectors, measuring approximately
16 feet in diameter, while the antenna system described in the
aforementioned patent application requires only one such large
reflector and a much smaller confocal subreflector as will be
described hereinafter. Thus, the disclosed system provides for
improved uniformity of signal gain with a simplified mechanical
structure of the antenna system.
-
~ ~7~ 2027456
Fig. 5 presents a diagram useful in explaining the
frequency scanning operation of the antenna system. A set of
four radiating elements 42 are arranged side by side along a
straight line and face an outgoing wavefront 66 of
electromagnetic radiation. The angle of incidence of the
wavefront or beam scan angle is measured relative to a normal 68
to the àrray 40 of elements 42. A frequency scan is generated
in a planar array antenna by introduction of a progressive time
delay into the array. The progressive time delay provides for a
10 - difference in the phases of signals excited by adjacent ones of
the elements 42 such that the phase difference is proportional
to the frequency of the radiated signals. This explanation of
the operation assumes an outgoing wavefront, it being understood
that the operation of the array of elements 42 is reciprocal so
that the explanation applies equally well to an incoming
wavefront. It should be noted here that the desired frequency
dependent port-to-port phase progression may be achieved by
using techniques other than time delays, such as through the use
of an all-pass network. The relationship of scan angle to
frequency, element spacing and time delay is given by the
following equations:
2~D sin ~ = 2~f~T (1)
therefore,
sin ~ = ~ f~T (2)
D
wherein:
D = spacing between elements,
= beam scan angle,
= wavelength of radiation,
~ = phase increment between adjacent elements,
f = frequency relative to band center, and
~T = time delay increment between adjacent
elements.
_ -8- 2027~5~
The radiating elements 42 are energized via a source
70 of microwave energy and a series of delay units 72 coupled to
the source 70. Each of the delay units 72 provides the time
delay increment referred to above in Equations (1) and (2). The
source 70 is connected directly to an element 42 at the left
side of the array while the next element 42 is connected by one
of the delay units 72 to the source 70. The signals applied by
the source 70 to the third and the fourth of the elements 42 are
delayed, respectively, by two and three delay increments of the
delay units 72. This provides the linear phase relationship to
provide the scan angle for the outgoing wavefront 66. The phase
increment between two adjacent ones of the radiators 42 is
proportional to the product of the frequency of the radiation
and the delay increment. When this product is equal to 360,
the wavefront propagates in a direction normal to the array of
element 42. Increasing values of frequency produce greater
phase shift to direct the wavefront to the right of the normal
68 as shown in Fig. 5, while decreasing amounts of frequency
produce less phase shift and drive the wavefront to the left of
the normal. Accordingly, the wavefront can be scanned
symmetrically about the array of elements 42.
It has been set out in the previous disclosure that
for the case of the foregoing uplink and downlink frequency
bands, and for the case of the radiating elements 42 having a
diameter of approximately one wavelength, a suitable value of
differential delay, as provided by the delay units 72 of Fig. 5,
is 185 nanoseconds for the case of substantially uniform
distribution of ground stations on the surface of the earth 60.
To provide the east-west coverage of 8, the uplink and the
downlink beams are scanned through an arc from -4 to +4. In
view of the magnification factor of 4.7, the scan angle of the
array 40 of radiating elements 42 must be enlarged by the same
magnifying factor, 4.7, from that of the output scan from the
main reflector 32. Therefore, the beam produced by the
radiating elements 42 must be scanned through an arc of 18.8 to
~ _9_ 2027456
either side of a normal to the array 40. The foregoing value of
differential delay, namely 185 nanoseconds, provides the 18.8
scan to either side of the normal to the array 40. In the ideal
situation of uniformly distributed ground stations between the
east coast and the west coast of the United States and Canada,
the number of channels per degree has a constant value of
1000/8 = 125.
In the situation wherein the differential delay
provided by the delay units 72 are independent of frequency,
then an optimal direction of the scanned beam is obtained for
the ideal situation of uniform distribution of ground stations.
In the more likely situation of a nonuniform distribution of
ground stations, the scanned beam may be displaced slightly from
its designated ground station. As has been noted above, such a
beam-pointing inaccuracy reduces the signal level by less than
0.01 decibels for a beam-pointing error of 0.1 degree.
The previous disclosure set out that
the scanning can be adapted to accommodate the foregoing
nonuniformity in ground 6tation distribution by introducing`a
frequency-responsive component to the differential delay. It
gives an example of nonuniform distribution where the
differential delay between columns of the array 40 of radiating
elements 42 (see Fig. 4) should vary, at least for the forming
of the downlink beams, between 262 nanoseconds at the low
frequency end of the transmission band to 131 nanoseconds in the
high frequency end of the transmission band. Other values of
delay may be employed in the beam-forming operation of uplink
beams provided by the receiver of the antenna system (30).
The values of delay used in the different frequency
bands, namely the uplink and downlink frequency bands, are
inversely proportional to the center frequencies of these bands
as is apparent from Equations (1) and (2). A reduction in the
differential delay results in a reduced amount of phase shift
- 2027456
--10--
between successive beams with a corresponding reduction in
displacement of beam position on the surface of the earth 60
from one channel to the next channel. Thereby, the beam can be
more accurately positioned in a region of high density of ground
stations. In a corresponding fashion, an increase in the
differential delay results in increased movement of the beam as
the frequency is shifted from one channel to the next channel,
thus accommodating positions of the beam to a less dense
distribution of ground stations. The channel number corresponds
to a specific frequency in either the uplink or the downlink
band. With respect to the positioning of ground stations along
an arc of a great circle of the earth 60, as disclosed with
reference to Fig. 4, it is seen that the frequencies selected
for the various stations vary monotonically with position along
the foregoing arc.
In view of the foregoing description, it is seen
that the above-described communication system provides two-way
communications between ground stations and a geosynchronous
satellite. The assignment of specific frequencies to respective
ones of the ground stations, in combination with frequency
scanning of both uplink and downlink beams of the satellite
(24), permits a simplification in the circuitry of the system.
In addition, the use of the two confocal parabolic reflectors
provides a multiplicative factor which reduces the number of
elements required in the array of radiating elements. The use
of a scanned beam also reduces the physical size of the antenna
system by reducing the number of reflectors, resulting in a
lighter weight, more efficient satellite communications system.
It has been f~und that certain technical impediments
exist to the commercial implementation of the above-described
confocal reflector system. One such impediment is the system's
lack of efficiency. Due to spacecraft size limitations, the
subreflector 34 cannot be constructed large enough (in terms of
wavelengths) to perform with acceptable efficiency. These size
-11- 202745~
limitations also restrict the size of the main reflector and the
focal lengths that may be used in the confocal arrangement.
Another impediment is the beam distortion which is
present when the beam scans toward the edge of the coverage
area. This beam distortion is undesirable because it reduces
the overall capacity of the system to transmit information.
It would be desirable, in order to achieve a further
weight-saving and simplification of the aforementioned satellite
communications system, to eliminate the subreflector altogether
while still utilizing a relatively low number of radiating
elements. It would also be very advantageous to be able to
combine the power of output signals from several individual
amplifiers operated in parallel into an individual one or small
group of the radiating elements so as to produce a stronger spot
beam in any given location along the area of the earth being
swept by the scanning beam. It would further be desirable to
use as many elements as possible as common element in an antenna
system for the transmitter antenna system and receiver antenna
system of a communications satellite so as to save weight, space
and cost. Still, another desirable feature would be to reduce
beam distortion which presently occurs when the beam which is
transmitted from the antenna is swept towards the edge of the
coverage area. The present invention is directed to achieving
one or more of these and other desirable objects.
SUMMARY OF T~E lNV~llON
In light of the foregoing objects, there is provided
in one embodiment of the present invention an antenna system
including a reflector having at least one focal point associated
therewith and an antenna array having a plurality of feed
elements for transmitting beams of electromagnetic radiation
onto a target area. The antenna system comprises: means
operatively connected to the antenna array for performing an
~ -12- 2027~56
approximate spatial transformation on the amplitude and phase
distribution of input signals provided thereto; at least one
additional feed element positioned along with the array of feed
elements, the additional feed element being used to produce a
beam which has a center portion that falls outside of the target
area and a fringe portion which overlaps a portion of the target
area, and wherein the antenna array, the additional feed
element, and the reflector are positionable relative to each
other such that the antenna array and the additional feed
element are operatively disposed near the focal point of the
reflector when the reflector is in its intended operating
position. The spatial transformation which is performed is
selected from the group of transformations consisting of
discrete Fourier transforms and inverse discrete Fourier
transforms. The transformation performing means includes a
Butler matrix having a plurality of input ports and a plurality
of output ports. The antenna array, including the one
additional feed element and the Butler matrix, is preferably
used for both transmission and reception of signals. When the
antenna system is used for reception, the input signals provided
to the spatial transformation means are signals obtained from
electromagnetic radiation focused by the reflector onto the
antenna array for reception by the feed elements, and the
spatial transformation is an inverse discrete Fourier
transform. When the system is used for transmission, the
spatial transformation means produces signals provided to the
antenna array and the additional feed element, and the spatial
transformation is a discrete Fourier transform. When used as a
transmitter, the system preferably further comprises means for
feeding the input ports of the Butler matrix with a set of
signals having a predetermined phase relation from input port to
input port.
.,
~ 12a 202745~
Another aspect of this invention is as follows:
An antenna system comprising:
a reflector having a focal point associated
therewith;
an antenna array having a plurality of radiating
elements and responsive to signals provided thereto for
producing beams of electromagnetic radiation covering a target
area, at least certain of said antenna beams including
undesirable side lobes;
at least one additional radiating element positioned
along with said array, said array and said additional radiating
element being operatively disposed near the focal point of said
reflector, said additional radiating element being responsive to
said signals to produce a first additional beam of
electromagnetic radiation having a center portion falling
outside of said target area and a fringe portion overlapping at
least a first portion of said target area, said fringe portion
of said first additional beam combining with said certain of
said antenna beams to reduce said side lobes; and
means operatively connected with said antenna array
and said one additional radiating element for performing an
approximate spatial transformation on the amplitude and phase
distribution of said signals provided to said antenna array and
said one additional radiating element.
The antenna system of the present invention is
preferably used in a satellite for communication with
ground stations. In such an application, the system
typically is
~ 13- 202745~
further comprised of a satellite frame to which the reflector
and antenna array are attached. The reflector and antenna array
are operatively arranged with respect to one another to enable a
steerable beam produced by electromagnetic radiation emanating
from said array to be reflected off the reflector when the
reflector is in its intended operating position.
Another aspect of the present invention is a method
of operating a steerable beam antenna system comprising the
steps of: (a) providing a set of first signals having a
predetermined phase relationship with respect to one another and
cont~;ning information to be transmitted; (b) generating a set
of signals from said set of first signals by at least performing
an approximate spatial transformation on the amplitude and
distribution of said set of first signals by at least performing
an approximate spatial transformation on the amplitude and
distribution of said set of first signals; (c)creating a
scanning beam by transmitting preselected ones of said second
set of signals towards the reflector by passing said preselected
ones of said second set of signals through a plurality of
radiating elements; and (d) creating at least one additional
beam by transmitting at least one signal in the second set of
signals towards the reflector by passing said at least one
signal through an additional radiating element located along
with the plurality of radiating elements, the additional beam
being directed so that the center portion of the beam falls
outside of the target area, and the fringe portion of the beam
falls upon said target side area, said fringe portion of said
additional beam combining with said scanning beam to reduce said
side lobes to a desirable level. The method may further comprLse
providing a Butler matrix in order to generate said set of
second signals from said set of first signals and wherein said
spatial transformation is a discrete Fourier transform.
The method is preferably used in satellite
communications systems for communicating with multiple ground
-14- 2027456
stations through the use of the steerable beam associated with
the antenna system. In such applications, the reflector is a
main reflector and is mounted on the satellite.
These and other aspects, objects, features and
advantages of the present invention will be more fully
understood from the following detailed description and appended
claims, taken in conjunction with the drawings.
BRIEF DESCRIPTION OF TFE DRAWINGS
In the accompanying drawings:
Fig. 1 is a side elevational view of a
communications satellite, showing an array of radiators, an
imaging reflector and primary reflector;
Fig. 2 is a front view of the rectangular array of
radiators in the Fig. 1 satellite;
Fig. 3 is a front elevational view of the antenna
subsystems shown in Fig. 1, which employs an alternative
hexagonal arrangement of radiators;
Fig. 4 is a stylized pictorial view of spot beams
formed on the surface of the earth using the Fig. 1 satellite;
Fig. 5 is a diagram showing a relationship between
an outgoing wavefront and the elements of a line array of
radiators;
Fig. 6 is a simplified diagrammatic representation
of an antenna system of the present invention usable in a
satellite;
-
-15- 2027~56
Fig. 7 is an optical diagram showing a dual lens
system;
Fig. 8 is a simplified electrical diagram of a
four-port Butler matrix usable in an antenna system of the
present invention;
Fig. 9 is a plot, as a function of input phase, of
the amplitude of the output signal of port A of the Fig. 8
Butler matrix when ports A', B', C' and D' are fed with a
specified set of input signals;
Fig. 10 is a plot, as a function of input phase, of
typical amplitudes of all of the output ports in the Fig. 8
Butler matrix when the matrix is fed with a specified set of
input signals;
Fig. 11 is a simplified electrical block diagram
showing a set of diplexers and a Butler matrix used in common by
transmitter and receiver networks in an antenna system of the
present invention;
Fig. 12 is an electrical diagram of one embodiment
of a receive network of the present invention which introduces
port-to-port phase differences into a received set of signals
through the use of progressive time delays or
frequency-dependent phase shifts;
Fig. 13 is an electrical block diagram of one
embodiment of a transmit network of the present invention having
functional similarities to the receive network ~hown in Fig. 12;
Fig. 14 is an electrical block diagram of a
dual-frequency, dual-signal transmit network of the present
invention for simultaneously summing and preparing for
transmission a plurality of distinct frequency signals;
2027~56
, .
Fig. 15 is a block diagram of a four-port Butler
matrix;
Figs. 16A-E are graphs of amplitude distributions
into and out of a four-port Butler matrix; and,
Fig. 17 is a combined block and pictorial view of
spot beams formed on the surface of the earth by a satellite
antenna array in accordance with the present invention.
DF.TA n.Fn DESCRIPTION OF TFE PREFERRED EMBODIMENTS
The present invention comprises a novel antenna
system for communicating with multiple ground stations typically
distributed over a large geographical area of the earth. The
following description is presented in conjunction with the
technical description set forth above to enable any person
skilled in the art to make and use the invention and is provided
in the context of a particular application and its
requirements. Various modifications to the preferred
embodiments will be readily apparent to those skilled in the
art, and the generic principle defined herein may be applied to
other embodiments and applications without departing from the
spirit and scope of the invention. Thus, the present invention
is not intended to be limited to the embodiments shown, but is
to be accorded the widest scope consistent with the principles
and features disclosed herein.
As shown in Fig. 6, the antenna system 80 of the
present invention includes a main reflector 32 and an array 40
(or 50) of radiators 42 (or 52) of the type described in Figs. 1
through 5 above. The antenna system 80 is preferably located
on satellite 24 by being mounted on a suitable frame 44. In the
antenna system 80 of the present invention, the subreflector 34
(shown in Figs. 1 and 3) is removed and the array 40 (or 50) of
-17- 2 D 2 7 4 5 6
primary radiators 42 (or 52) or feed horns is placed at (or
near) the focal point or plane 48 of the offset feed reflector
32. The array 40 (or 50) is fed by a Butler matrix 82 which is
arranged "backwards" with respect to the traditional use of this
type of beam-forming matrix. Connected to the matrix 82 are the
transmitter and receiver networks represented by block 84. The
use of a Butler matrix in this manner produces an excitation
sequence for the antenna system 80 which is the spatial discrete
Fourier transform of the excitation sequence input to the
beam-forming array of primary radiators in the antenna system 30
of Figs. 1 through 5. In this way, the far field pattern
produced by the array and signal reflector is identical (in the
ideal case) to that of the previously described confocal
arrangement. There will be some difference in the non-ideal
case due to the effects of spatial sampling and the physical
limitations on the size of the array that may be placed at the
focal point.
The operation of the antenna system 80 shown in Fig.
6 and the foregoing explanation of same may be better understood
by considering the equivalent optical model of the conformal
reflector configurations, shown in Figs. l and 3 previously
described above. Fig. 7 shows the equivalent optical model 90
of this earlier antenna system 30 employing two lenses 92 and 94
which correspond in function to the main reflector 32 and the
subreflector 34 respectively. The focal plane x represented by
line 96 includes the focal point 48. The focal length of lens
92 is F2, while the focal length of lens 94 is Fl. The
magnification factor M of the system 90 in Fig. 7 is given by:
M = F2/Fl (3)
An amplitude and phase distribution of an image, F(x), at the
image plane x represented by line 98, is magnified by lenses 92
-18- 2027~5~
and 94 such that at the magnified image plane x" represented by
line 100:
f(x") = F (M X) (4)
From optical theory, it is well known that the amplitude and
phase distribution at the focal plane x' is the spatial dlscrete
Fourier transform of the amplitude and phase distribution at the
image plane X. That is to say:
f(x') = F[f(x)] (5)
By removing the first lens and producing f(x') directly at the
focal plane x', the same amplitude and phase distribution will
result at the magnified image plane x". The antenna system 80
of the present invention is based upon this idea.
Returning to Fig. 6, it may be seen that the Butler
matrix 82 in the system 80 performs the spatial discrete Fourier
transform of the excitation sequence generated by the
transmitter in block 84. It may also be seen that the Butler
matrix 82 performs the spatial inverse discrete Fourier
transform, F-l[f(x")], of the far field beam reflected off
reflector 32 and focused onto the antenna array 40 (or 50) for
reception by the feed elements 42 (or 52) and subsequent
processing by the receiver in block 84.
Fig. 8 illustrates a four-port Butler matrix 110
which has a set of four inputs and a set of four outputs. The
Butler matrix 110 includes four 90 phase lead hybrids 112 and
two negative 45 phase shifters 114 interconnected to one
another and to the two sets of four ports as shown. The
four-port matrix 110 is considered here for simplicity, but as
those in the art know, Butler matrices can be designed with any
number of desired ports. In this regard, much work has been
done in developing design technique for Butler matrices, see,
-19- 2027~56
e.g., M. Ueno, "A Systematic Design Formulation for Butler
Matrix Applied FFT Algorithm," IEEE Trans. Antennas and
Propagation, Vol. AP-29, No. 3, May 1981. In the traditional
use of this matrix, ports A, B, C and D would be the input
ports, and ports A', B', C' and D' would be the output ports and
would be attached to radiator elements in an antenna system
which does not use a reflector. When the antenna system 80 of
the present invention is used to transmit, ports A', B', C' and
D' are used as the input ports, and ports A, B, C and D are used
as the output ports. In the system 80 used as a transmitter,
the ports A', B', C' and D' are fed with a set of signals that
have some predetermined phase relationship from port to port
that is a function of frequency. If the same signals were fed
to a planar antenna array, different spot beams, each with a
different beam direction, would be formed for the different
frequencies. We sometimes refer to these spot beams as virtual
beams, since in theory a continuum of beams exists over the
entire beam width defined by the lowest frequency to highest
frequency spot beams. The different phase distributions
resulting from different frequencies are combined in the matrix
110 and constructively or destructively combine on different
output ports. The effect is the creation of a virtual phase
center in the array of signals at output ports A, B, C and D for
each frequency. In other words, the phase center of an antenna
array 40 (or 50) having a plurality of radiator elements 42 (or
52) with one such element attached to ports A, B, C and D will
scan as a function of frequency. A particular frequency may
result in a signal at one and only one port, or it may result in
signals at two or more ports whose amplitude and phase
correspond to a spatial phase center somewhere between the ports.
Curve 120 of Fig. 9 shows the amplitude response of
port A in Fig. 8 when input ports A', B', C' and D' of matrix
110 are fed with a set of input signals defined by:
Xn (t) = sin (~t + n~) n = 0, 1, 2, 3 (6)
-20-
2027456
where ~ is varied from 0 to 2 pi. (In the usual frequency scan
technique, the input phase value ~ is some function of frequency
and is not necessarily constant.) Port A' corresponds to n = 0,
port B' to n = 1 and so forth. Note that at a particular phase
distribution a ~~; , signal level occurs at port A. Figure 10
shows the magnitudes of the output signals on all ports when the
matrix 110 is fed with the same type of signal sequence
described above. Curves 122, 124 and 126 are the output signals
on all ports when the matrix 110 is fed with the same type of
signal sequence described above. Curves 122, 124 and 126 are
the output signals of ports B, C and D, respectively. Note that
each port has a maximum output value for a different relative
input phase value ~. Note also that of a - ;- amplitude for
a particular port, the outputs of the other ports are zero. For
example, when curve 120 associated with port A is of its ~~i ~
at point 128, curves 122-126 are at zero amplitude at point
130. Further analysis shows that the phase center of the
antenna array attached to the output of the Butler matrix (as
illustrated in Fig. 11) will scan the length of the array 40 (or
50) as a function of ~. Different frequencies result in
different phase centers of the antenna array.
The performance of an antenna system employing a
Butler matrix is affected by the number of elements used. The
more elements used the better the spatial sampling of the input
and output signal sequences. Thus, it will be appreciated that
a Butler matrix with relatively few ports performs a rough
approximation of a discrete Fourier transform (or inverse
discrete Fourier transform) on signals passing therethrough. As
the number of ports increases, the quality and accuracy of the
transformation performed increases.
Fig. 11 illustrates in greater detail how the Butler
matrix 110 may be used in an antenna system 138 of the present
invention which includes transmitter and receiver subsystems.
-
-21- 2027456
The system 138 includes an array 140 of feed elements or horns
142 which are used both as radiating elements and receiving
elements. The horns 142 may be of any conventional or suitable
design, such as a cup dipole one wavelength in diameter. In
practice, the horns 142 function in the same basic manner as the
5 radiating elements 42 (or 52) described earlier and are located
at or very near the focal plane or plane of an offset feed
reflector 32 as in the Fig. 6 arrangement.
The system 138 also includes a group 144 of
diplexers 146, a receive network 148 and a transmit network 150
which are respectively connected to the diplexers 146 by groups
152 and 154 of electromagnetic conduits or conductors. These
components may all be of conventional or suitable design. For
the diplexers 164, however, we prefer to use diplexers of the
type fully described in commonly assigned U.S. Patent No.
4,427,953 to T. Hudspeth and H. Keeling entitled "MICROWAVE
DIPLEXER." The diplexers 146 serve to properly route incoming
signals in the uplink frequency band (received by the antenna
array 140, transformed by the matrix 110, and impressed upon
conductors of conductor group 156) to the receive network 148.
Similarly, the diplexers 146 serve to route the set of signals
in the downlink frequency band (generated by the transmit
network 150 and impressed upon conductor group 154) to the
Butler matrix 110 where they are transformed and impressed on
conductor group 158 for delivery to the antenna array 140. Note
that in Fig. 11 the output of ports B and C of the Butler matrix
110 is reversed in order to obtain a continuous scan of the
virtual phase center with frequency. This is accomplished by
having port B connected by conduit 158b to feed horn 142c and
having port C connected via conduit 58c to feed horn 142b as
shown. The need to reverse the outputs of the beam ports B and
C is clear when one observes that in Fig. 10 the curves 122 and
124 representing for the output signals of ports B and C are
reversed with respect to the ordering of the output signals of
ports A and D, from high to low values of input phase ~. The
-22- 2027456
two 180 phase shifters 159a and 159b correct for phase
reversals in the output signals of ports A and B, which occur on
account of the operation of the Butler matrix 110.
Fig. 12 is a block diagram for one possible
embodiment which shows the various components and signal paths
of the receive network 148 of antenna system 138 in Fig. 11.
The network 148 includes: a group 160 of preamplifiers 162 for
boosting the level of the received signals delivered by conduits
152; a group 164 of frequency translators 166 for reducing the
carrier frequency of received signals from preamplifiers 162 to
an intermediate or baseband frequency range; a group 168 of four
bandpass filters 170 for rejecting side lobes or other frequency
translation products outside of the desired frequency range; and
a group 172 of three shift-producing components or elements 172,
all connected together as shown to produce a baseband signal on
output terminal or port 176. The shift producing elements 174
may either be time delay elements or frequency-dependent phase
shifters. The receiver network 148 is tuned to the frequency
bands of the respective uplink communication channels, thereby
permitting simultaneous reception of signals from a plurality of
ground stations.
Fig. 13 is a block diagram of the transmit network
150 shown in Fig. 11. The network 150 includes: a group 180 of
shift-producing components or elements 182; a group 184 of four
frequency shifters 186 for increasing the carrier frequency of
signals imposed on group of conduits 188 to a higher frequency
range, a group 190 of bandpass filters 192 for removing unwanted
signals outside of the desired frequency range generated by the
o~er~t~on frequency translators 186; and a ~roup 194 of power
amplifiers 196 to boost the power of the signals impressed on
conductor group 154. In the transmit network 150, the signal to
be transmitted is imposed upon input terminal or port 198.
-
-23- 2027~5~
Figs. 12 and 13 also illustrate one possible method
for producing the port-to-port input phase value ~ as a function
of frequency through the introduction of time delays or
frequency-dependent phase shifts. These are introduced through
the shift-producing elements 174 in Fig. 12 and the
shift-producing elements 182 in Fig. 13. In Fig. 13, the time
delays or phase shifts are introduced at baseband (or some
intermediate frequency), and then each signal in the resultant
signal set is imposed on a conduit of conduit group 188 in order
to be frequency-translated in parallel by frequency translators
186 up to the desired frequency range. This is done so that a
particular bandwidth will produce the desired range of phase
distributions in the signal set applied to the Butler matrix 110
through conductor group 154 and therefore result in the scanning
of the phase center of the array across to the desired range.
This method can be advantageously applied, for example, in the
MAST system discussed in the background portion of the
specification.
The MSAT system satellite discussed above will
transmit in the UHF band at 866 to 870 MHz (or in the L-band if
desired). The change in phase of a sinusoid due to a time delay
such as those in Figs. 12 and 13 can be calculated by the
following formula where input phase value is expressed in
radians:
~ = n2~ (7)
In order to produce a sufficiently large beam scan angle when
using the Butler matrix 110 shown in Fig. 8, a fairly wide range
of phase distributions is required.
One approach for determining what time delays or
phase shifts are required to operate the system 138 of Fig. 11
in the desired manner is to choose the optimum range of phase
distributions and find a frequency at which the bandwidth in
-24- 20274S6
question will produce this range using a time delay or phase
shifting device. For example, the Butler matrix 110 shown in
Fig. 8 will provide the best sc~nning of the phase center of the
antenna array if the input phase distributions range between ~/4
and 7~/4 radians. For simplicity, assume a time delay will be
used. Hence, setting the conditions:
~/4 = 2~fl~ (8)
and
7~/4 = 2~f2~ (9)
then
I = 7~r/8~f2 = lt/8ttfl (10)
or
f2/fl = 7 (11)
Combining this relationship with the idea that
f2 - fl = 4 MHz (12)
i.e., in the bandwidth of downlink transmissions in the MAST
satellite, we can find that fl = 666.7 x 103 Hz and f2 =
4.6667 x 106 Hz. Working at this intermediate frequency range,
we can now find a time delay that will produce the desired range
of phase distributions, namelg ~ = 1/(8fl) = 187.5 x 10 9
seconds. Wor~in~ with this time delay at this intermediate
frequency band allows the bandwidth of the signal to produce the
desired port-to-port frequency-dependent phase relationships.
Each signal in the set can then be frequency-translated up to
the desired frequency range (in parallel) without changing the
-
-25- 202745~
port-to-port phase relationship introduced by the time delays at
the intermediate frequency band.
By using different time delays and different
intermediate frequencies, the signals from different
transmitters (or to different receivers) may be combined to use
the same Butler matrix and produce the same type of antenna
patterns, even if the transmitters are operating at different
frequencies and have different bandwidths. Using the technique
described above, different signals with different bandwidths may
be used to produce sets of input signals with the same range of
phase distributions. Applying the combination of these sets to
the Butler matrix feeding an antenna array allows both signal
bandwidths to produce the same frequency scanned virtual beam
patterns. This concept is illustrated in Fig. 14 which shows a
dual-frequency transmit network 210 capable of generating two
sets of output signals at different frequency bands. The
network 210 may be used in place of transmit network 150 in the
antenna system 138 shown in Fig. 11.
The network 210 includes a first frequency network
portion 212, a second frequency network portion 214, and a
common network portion 216. Network portion 212 is comprised of
a group 180 of three shift-producing devices 182 and a group 184
of frequency translators 186 which operate as previously
explained in Fig. 13. Network portion 214 includes a group 222
of shift-producing devices 224 and a group 226 of frequency
translating devices 228 for producing a set of signals at a
different frequency band from those produced by network portion
212. The shift-producing devices 182 and 224 may be time delay
units or frequency-dependent phase shif t units. Network portion
216 includes: a group 230 of sum-producing elements or mixers
232 (which combine the two different sets of signals from
network portions 212 and 214 delivered to the mixers 232 via
conductor groups 234 and 236, respectively); a group 190 of band
pass filters 192; and a group 194 of power amplifiers 196. The
-26- 202745~
various components of network 210 are connected as shown in Fig.
14 and result in the production of two sets of signals having
different frequency bands which are combined, amplified and then
simultaneously impressed upon conductor group 154 for delivery
to the remainder of the system 138 shown in Fig. 11.
The shift-producing units or devices 174, 182 and
224 may be conveniently fabricated of lumped parameter all-pass
networks employing well-known circuitry. These units or devices
are located ahead of the transmitting power amplifiers 184 and
196 in Figs. 13 and 14 so as to operate at relatively low power
and thereby minimize power loss.
To summarize at this point, the steerable beam
antenna system described above provides a significant reduction
in satellite weight by eliminating one of the two confocal
parabolic reflectors which are used to provide two-way
communications between ground stations and a geosynchronous
satellite. Also provided is a technique for combining the power
of output signals from several individual amplifiers operated in
parallel into an individual one or small group of feed elements
so as to more efficiently produce a stronger spot beam in a
given location within the target area. In addition, the system
described above provides for a technique wherein several antenna
elements are used for both receiving and transmitting
electromagnetic radiation. By providing two functions with
several common elements, weight, space and cost of the satellite
are reduced.
Although the antenna system described above is
highly effective, it produces beams which, at the outer edges of
the target area, have undesirably high side lobes. The present
invention, however, provides a solution to this problem, and in
this connection reference is now made to Fig. 15, wherein a
Butler matrix 110 is shown having four inputs a', b', c', and d'
and four outputs a, b, c, and d. The Butler matrix 110 is
- 2027~56
-27-
comprised of four 90 phase lead hybrids 112 and two negative
45 phase shifters 114 interconnected as shown. The outputs of
the Butler matrix a-d are routed to feed elements 116-119. The
feed elements 116-119 deliver the electromagnetic radiation to a
confocal reflector (not shown) which reflects and redirects the
radiation to the target area. Outputs a and d are routed
through 180 phase shifters 115 before they deliver the outputs
of their respective Butler matrix ports to their respective feed
elements. As previously disclosed, the Butler matrix 110 is
used to perform a spatial discrete Fourier transform on the
incoming signals a'-d' to produce a focal plane field
distribution equivalent to the common focal plane distribution
in a dual confocal reflector system. If the Butler matrix feed
arrangement as shown in Fig. 15 is used, some undesirably high
side lobes will be present in the scanning beam whenever it
scans towards the edge of the target area as mentioned above.
The presence of these undesirable side lobes can be explained by
referring to Fig. 16.
Shown in Fig. 16a is a hypothetical amplitude
distribution which may be applied to Butler matrix 110 via its
input ports a'-d'. This hypothetical amplitude distribution
will be used as a reference in discussing the output
distributions shown in Figs. 16b-16e. As can be seen in Fig.
16a, the input distribution maintains a constant amplitude from
port-to-port and a constant port-to-port phase gradient. Fig.
16b shows the output that would be present on feed elements
116-119 if the input to the Butler matrix 110 was fed by four
constant amplitude signals which shared a constant phase
gradient of -45. Note that for this hypothetical input phase
~radient, all of the power radiated from horns 116- 119 would be
output from port 2. Assuming that a phase gradient of 45 was
presented to the signals that were input to the Butler matrix
110, Fig. 16c shows that all of the power output from the Butler
matrix 110 would be output from port 3, and no power would
emanate from ports 1, 2 or 4. Fig. 16d shows that if the four
-28- 2027456
inputs to the Butler matrix a'-d' share a 0 phase gradient
(i.e. they are all in phase), each output port then contributes
to the total output power. Note that Fig. 16d shows there is a
180 phase reversal on the outside ports. Also shown in Fig.
16d is the electrical phase center 300 of the array for the
input phase gradient of 0. Fig. 16e shows the output of the
Butler matrix for an input phase gradient of 90. Note that
like Fig. 16d, each of the ports contributes some power to the
overall output of the array, and that there is a 180 phase
reversal on left-side port 2. Shown at 302 in Fig. 16e is the
approximate location of the electrical phase center of the array
for an input phase gradient of 90.
The case in Fig. 16e illustrates the presence of
undesirable side lobes for an input phase gradient of 90. In
this case, unlike the cases in Figs. 16b-16d, the excitation of
the horns 116-119 is asymmetric about the electrical phase
center of the array. This asymmetry in the feed excitations
causes an asymmetry in the beam patterns for these intermediate
beam positions which occur near the edge of the array. In other
words, as the electrical phase sensor of the array scans towards
the edge of the array, the beam shape degrades or dilates. This
dilation is undesirable because in order to accommodate it, the
system bandwidth and consequently system capacity must be
sacrificed. A novel approach has been developed and is herein
disclosed for eliminating this beam dilation. This novel
approach will now be described in conjunction with Fig. 17.
Now referring to Fig. 17, a standard eight-port
Butler matrix 110 is shown having eight input ports 310 and
eight output ports 312. Ei~ht-port Butler matrices of this
nature are well-known to those in the art of satellite
communications. Four of the outputs of the eight-port Butler
matrix 116-119 are shown terminated at output feeds 1-4. This
correlates to the arrangement used in Fig. 15. Two of the
L~- ?;n;ng outputs from the Butler matrix are terminated into
-29- 2027456
output feeds which flank output feeds 116-119 and are shown as
output feeds 314 and 316. The remaining two outputs of the
Butler matrix are terminated into the non-radiating load
elements 318, 320.
5The novel technique disclosed herein for eliminating
the undesirable dilation of the arrayed beam involves using
additional feed elements in the following unique manner. As can
be seen from Fig. 17, primary feed elements 116-119 adequately
cover the target area (continental United States) 32Z. Antenna
10feeds 314 and 316 are used to flank the primary feeds and to
create beam patterns that do not concentrate their energy on the
target area. This is illustrated in Fig. 17 by noting that the
beam pattern projected by feed element 314 does not in fact fall
upon the continental United States. Likewise, the beam pattern
15emanating from feed element 316 is shown off of the east coast
of the continental United States and predominantly falling
within the Atlantic Ocean. Note that the reflector has been
omitted from Fig. 17 for the sake of clarity and explanation,
but would in fact be employed in a normal working system.
20The primary beams 324-330 fall upon the target area
generally along a common axis. These beams emanate from the
feed elements 116-119 and are used to transmit information
throughout the target area. The outer feeds 314 and 316 are
used only for beam shaping and consequently are not used for
25information transmission. In Fig. 17, beam patterns 332 and 334
are drawn to show what the beam pattern would look like if a
sufficiently high amount of energy was delivered to feed element
314 and 316. However, in the disclosed system, beam 332 and 334
~re never used as primary beams, but rather they are used as
30secondary beams and consequently they are excited with only a
fraction of the energy which is normally delivered to the
primary beams~ Feeds 314 and 316 are used to produce secondary
beams for the purpose of beam shaping as the phase center of the
antenna array moves between intermediate beam positions. Beam
_30_ 2027~56
shaping is accomplished by virtue of the secondary beam
destructively interacting with the undesirable side lobes which
accompany the primary beams when the primary beams scan towards
the outer ends of the antenna array. The inclusion of these
feeds 314, 316 provides for a more symmetric array excitation,
and hence a better pattern shape results at the intermediate
beam positions. It is important to note that in addition to
destructively combining with the undesirable side lobes of its
adjacent neighbors, a secondary beam also acts to reduce the
side lobes which are produced by its more remote neighboring
beams of the array, although the further away from a given
secondary beam the side lobe occurs, the less the secondary
beam's effect on that side lobe.
Although two of the outputs of the Butler matrix 110
lS are shown in a terminated state 318, 320, they along with feed
elements 314 and 316 can be used to perform beam shaping. If
all eight outputs of the Butler matrix 110 were used, the
system's overall ability to eliminate beam dilation would be
even greater than it is when only six of the outputs of the
Butler matrix 110 are used. Terminating the two outputs as
shown in Fig. 17 introduces a small power loss. For the case
illustrated in Fig. 17, if the input phases are restricted such
that beams 322 and 330 represent the ; beam scan position,
the maximum power loss is -.33 dB. If one watt of power is
presented to feed element 116, then .0732 watt would be
dissipated in the loads shown at 318 and 320. If all eight
ports of the Butler matrix 110 are used for beam shaping, this
loss will not occur.
The foregoing embodiments of the present invention
have been described with respect to a mobile satellite
communications system for transmitting and receiving between
multiple ground stations at certain specified frequencies in the
L band. Those skilled in the art will appreciate that the
present invention may be readily adapted for use in land or
-31- 2 0274 5 6
satellite communication systems operated in other frequency
bands, such as the C or Ku bands, for example. The number of
additional radiating elements, which are added to eliminate side
lobes, is not limited to one or two. Any number of additional
radiating elements may be added, each of which will act to
further reduce undesirable side lobes. The size of the main
reflector, the arrangement and type of antenna arrays, and the
specific receive and transmit networks utilized in the present
invention may vary substantially without departing from the
scope of the broader aspects of the present invention. For
example, separate feed horns may be used to transmit and receive
electromagnetic radiation constituting the steerable beam.
Also, a conventional screen-type diplexer may be placed between
the antenna array and reflector so as to divert the incoming
electromagnetic radiation to be received to a separate receiver
array arranged at a substantial angle to the plane of the first
antenna array. Such an embodiment would thus have separate
transmit and receive antenna arrays. Alternatively, two
separate main reflectors could be provided, one to be used with
a separate transmit antenna array and the other to be used with
a separate receive antenna array. We presently do not favor
this latter arrangement for satellite antenna systems of the
present invention on account of the appreciable extra weight and
cost of providing two main reflectors. However, such an
embodiment may be quite suitable for systems of the present
invention constructed on land or on sea-going vessels.
In view of the foregoing description, it is seen
that the antenna system of the present invention is well-suited
for two-way communications between ground stations and a
geosynchronous satellite. The antenna system of the present
invention has the advantages of effectively combining the power
of the output signals of a plurality of power amplifiers
simultaneously operated in parallel. It also provides a single
reflector antenna system which, through the use of a spatial
transformation means such as a Butler matrix, is functionally
_ -32- 2027456
equivalent to the dual confocal reflector system described in
the background portion of the specification, including achieving
a magnification of the effective aperture of the elements. The
antenna system of the present invention eliminates the need for
the use of a subreflector without providing additional radiating
elements, thus saving weight, space and cost. Since the antenna
system of the present invention uses a scannable virtual beam
technique, it also reduces the physical size of the antenna
system by minimizing the number of radiating elements and
reflectors which must be used. The present system also acts to
reduce the undesirable side lobes that occur as the radiated
beam scans towards the edge portions of the target area. This
side lobe suppression is achieved by placing additional
radiating elements along with the main array and using the
additional radiating elements to produce beams of energy which
cancel the undesirable side lobes which accompany the primary
beams. Thus, an antenna system of the present invention results
in a lighter weight, more efficient satellite communication
system. Finally, the use of time delays or phase shifts at
baseband or intermediate frequencies allows the output of
multiple transmit networks to be applied to a single array of
radiating elements to produce the same antenna pattern at
different frequencies, thus enabling the antenna system to be
used in satellite communication systems requiring a multiple,
two-way, simultaneous communication channel between many widely
separated ground stations within the scAnn;ng angle of the
virtual beams.
It is to be understood that the above-described
embodiments of the present invention are illustrative only and
that modifications thereof may occur to those skilled in the
art. Accordingly, the present invention is not to be regarded
as limited to the embodiments or methods disclosed herein but is
to be limited only as defined by the appended claims.
