EP0273923A1

EP0273923A1 - Combined uplink and downlink satellite antenna feed network

Info

Publication number: EP0273923A1
Application number: EP19870902969
Authority: EP
Inventors: Donald C. D. Chang; Wilbur J. Linhardt
Original assignee: Hughes Aircraft Co
Current assignee: Raytheon Co
Priority date: 1986-05-19
Filing date: 1987-03-30
Publication date: 1988-07-13
Also published as: WO1987007439A1; CN87103632A; JPH01501190A

Abstract

Réseau combiné (20) d'alimentation à liaison montante/descendante pour satellites de communications. Le réseau combiné (20) élimine la nécessité d'utiliser des réseaux séparés à liaison montante et descendante ainsi que des diplexeurs séparés dans chaque élément, ce qui permet de réduire la complexité, le poids et le coût du réseau d'alimentation.Combined uplink / downlink power network (20) for communications satellites. The combined network (20) eliminates the need for separate uplink and downlink networks and separate diplexers in each element, thereby reducing the complexity, weight and cost of the power network.

Description

COMBINED UPLINK AND DOWNLINK SATELLITE ANTENNA PEED NETWORK BACKGROUND OF THE INVENTION

The present invention relates to microwave transmission networks, and particularly to satellite feed networks. Satellite antenna feed networks control the amplitude and phase distribution used to excite the radiating elements of the antenna feed array and thus control the antenna radiation pattern.

Communication satellites are employed to provide a communication link between ground stations. In most applications, the satellite acts as a communication repeater, receiving an "uplink" signal from one station, and retransmitting the signal as a "downlink" signal to another ground station comprising the link.

The satellite typically comprises a receiver coupled to an antenna system for receiving the uplink signals, and a transmitter coupled to the antenna system for transmitting the downlink signals. To minimize interference between the uplink and downlink signals, separate and distinct uplink and downlink frequency bands are employed. The satellite also includes frequency translating means for converting the uplink signal from a frequency within the uplink frequency band into a downlink signal at a frequency within the downlink frequency band.

It is understood that in the design of prior communication satellites, the practice has been to employ separate feed networks for the uplink and downlink coverage beams. Thus, typically a separate uplink network is employed to couple the receiver to the antenna system, and a separate downlink network is employed to couple the transmitter to the antenna system. When separate feed networks attach to a common antenna feed array, it is necessary to locate a frequency-sensitive- diplexer at each feed element input to route the uplink and downlink signals to the respective feed networks.

The use of separate uplink and downlink feed networks is costly, space consuming, and adds to the weight of the spacecraft. The weight as well as the size of the spacecraft is limited due to launch requirements.

It would therefore be advantageous to provide a relatively lightweight, space efficient, and low cost feed network for communication satellites. SUMMARY OF THE INVENTION

An improved microwave feed network for coupling a satellite antenna system to the satellite receiver and transmitter is disclosed. In accordance with the invention, a single wideband uplink and downlink feed network is employed. The feed network comprises a single, frequency-sensitive diplexer at the feed network input for separating the transmit and receive signals and coupling them to the receiver and transmitter. The feed network further comprises a single wideband corporate feed network for coupling to the respective antenna radiating elements comprising the antenna system. The diplexer and elements of the corporate feed network, as well as the antenna radiating elements, are adapted for wideband operation over both the uplink and downlink frequency bands. The invention is particularly useful for those applications in which the satellite receive and transmit beam coverages are coincident and provides the advantages of lower cost and enhanced weight and space efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

These and other features and advantages of the present invention will become more apparent from the following detailed description of an exemplary embodiment thereof, as illustrated in the accompanying drawing, in which:

FIG. 1 is a simplified schematic diagram of a combined uplink and downlink feed network in accordance with the invention. FIG. 2 is an end view of a wideband coupler which may advantageously comprise the feed network of FIG. 1.

FIG. 3 is a plan view of the coupler of FIG. 2 sectioned along line 3-3 of FIG. 2.

FIG. 4 is a longitudinal sectional view of the coupler taken along line 4-4 of FIG. 2.

FIG. 5 is a longitudinal sectional view of the couple taken along the line 5-5 of FIG. 2.

FIG. 6 is a graph of phase shift versus frequency for each of two phase shifting sections of the coupler of FIG. 2.

FIG. 7 is a top schematic view of a typical horn antenna.

FIG. 8 is a plot of the horn phase delay for two horn antennas of different aperture sizes, as a function of horn length at selected high and low frequencies.

FIG. 9 is a plot of the phase delay as a function of horn length for two horns of different aperture sizes.

FIG. 10A depicts a simplified representation of a reference horn antenna having an overall length of 12 inches and a two-inch aperture.

FIGS. 10B and 10C depict simplified representations of a horn antenna having a 12 inch length and a 4-inch aperture, respectively optimized (dashed lines) at two different frequencies within a frequency band of interest. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT

A simplified schematic of a combined uplink and downlink feed network employing the invention is shown in FIG. 1. The satellite antenna system comprises a plurality of wideband, non-frequency dispersive horns 60-67 employed for reception and transmission. The antennas 60-67 are coupled to a diplexer 5 by a corporate feed network 20, comprising a plurality of wideband non-frequency dispersive couplers 21, 25, 30, 32 34, 36 and 38.

The diplexer device 5 is a frequency sensitive device adapted to couple substantially all the signal power in the uplink frequency band to receiver 10, and to couple virtually no signal power in the downlink frequency band to the receiver. Thus, the function of the diplexer device 5 is to separate the uplink signals from the downlink signals. The diplexer is adapted to provide very good isolation between the receive and transmit signals and also to carry the relatively high signal power supplied by the transmitter 15.

It is noted that the network disclosed in FIG. 1 employs a single diplexer device 5. This contrasts with the prior art designs in which separate feed networks for the uplink and downlink signals are employed. While the horn antennas are shared by the uplink and downlink signals, separate diplexers are typically employed at each horn antenna to separate the uplink and downlink signals tα feed the respective uplink and downlink networks.

The diplexers in these prior art designs need not be designed to carry all the transmitter power, as in the disclosed embodiment, however, but rather need only carry a fractional part of the power, since it is distributed over the horns in accordance with the antenna pattern. For example, the diplexer 5 in the disclosed embodiment may be required to carry signal powers on the order of a kilowatt, while the individual diplexers deployed at each horn in the prior art design may be required to carry signal powers on the order of only 10 watts. The design of diplexers for carrying the higher power levels is known in the art.

The diplexer 5 is coupled to the corporate feed network 20, which performs a power distribution function between diplexer port 6 and the antenna feed ports 60a-67a coupling the respective antenna horns 60-67 to the network 20. As is well known to those skilled in the art, the relative, respective phase delays and attenuation introduced between network port 6 and the respective antenna ports affects the antenna system beam pattern. The network 20 operates reciprocally to divide the downlink energy inputted to network 20 on line 6 from the diplexer 5 and the transmitter 15 among the respective antenna ports 60a-67a, and to combine the uplink energy received at the ports 60a-67a from the antennas and provide the combined energy at port 6 to be coupled to the receiver 10 through the diplexer 5. Thus, although the coupler devices comprising the network 20 will be described in the following description as power dividers, it is to be understood that the devices also operate in the reciprocal sense as power combiners. In the exemplary network shown in FIG. 1, the corporate feed network 20 comprises the wideband phase compensated couplers 21, 25, 30, 32, 34, 36, 38 to divide the signal power between port 6 and the antenna ports 60a-67a. The network 20 further comprises phase adjustment trombones 41, 43, 45, 47, 49, 51, 53, 55 to provide additional phase compensation.

The isolated ports of the respective couplers 21, 25, 30, 32, 34, 36, 38 are each terminated in a matched load. The couplers, as described above, divide the input power provided the input port between the through and coupled ports in accordance with a coupling factor.

The input port 21a of coupler 21 is coupled by transmission line 6 of the corporate feed network. 20 to the diplexer 5. The through port 21b of the coupler 21 is coupled by transmission line 23 to the input port 30a of coupler 30. The coupled port 21c of coupler 21 is coupled by transmission line 22 to the input port 25a of coupler 25. The through port 34b of coupler 34 is coupled through phase compensating trombone 47 to the antenna port 63a by transmission line 46. The coupled port 34c of coupler 34 is coupled through trombone 45 to antenna port 62a by transmission line 44. The through port 3.2b of coupler 32 is coupled through trombone 41 to the antenna port 60a by transmission line 40. The coupled port 32c of coupler 32 is coupled through trombone 43 to antenna port 61a by transmission line 42. The through port 36b of coupler 36 is coupled through trombone 49 to antenna port 64a by transmission line 48. The coupled port 36c is coupled through trombone 51 to antenna port 65a by transmission line 50.

The through port 38b of coupler 38 is coupled through trombone 55 to antenna port 67a by transmission line 54. The coupled port of is coupled through trombone 53 to antenna port 66a by transmission line 52.

The particular coupling factors and phase compensation of the couplers and the phase compensation of the trombones comprising the network 20, as well as its general configuration, is dependent on the particular application, as will be appreciated by those skilled in the art.

Using the same network for both the uplink and downlink frequencies requires that the network components be more broadband than if separate uplink and downlink networks were employed. An exemplary set of uplink and frequency bands is 13.75 Ghz to 14.25 Ghz for the uplink band, and 11.75 Ghz to 12.25 Ghz for the downlink band. Thus, for this example, each component in the feed network is required to operate over the range of frequencies between 11.75 Ghz to 14.25 Ghz, an operable bandwidth of about 2.5 Ghz.

With reference to FIGS. 2-5, an embodiment of a hybrid coupler 110 is illustrated which may be advantageously employed in the network 20 illustrated in FIG. 1. This coupler 110 is the subject of the copending patent application entitled "Phase Compensated Hybrid Coupler," by M.N. Wong and W.J. Linhardt, docket PD-84060, serial number 782,677, filed October 2, 1985, which has a common assignee with the present application. The coupler 110 is formed of a first waveguide 112 and a second waveguide. 114, each of which have rectangular cross-sectional form wherein the ratio of a long wall to a short wall is 2:1. For operation at a microwave frequency of 12 GHz (gigahertz), waveguide type WR-75 is employed. Each of the waveguides has two long walls, namely a top wall 116 and a bottom wall 118, which are joined by short walls, namely outer sidewalls 120 and a common wall 122 which serves as an inner sidewall for each of the two waveguides 112 and 114. The coupler 110 is a very broad band device which, in the preferred embodiment of the invention, has an operating range extending from 11.7 GHz to 14.5 GHz.

The coupler 110 provides the dual functions of hybrid coupling plus phase compensation of electromagnetic energy between the two waveguides 112 and 114. The coupling of the electromagnetic energy is accomplished by a gate 124 located in the common wall 122. For 3 dB (decibels) coupling, the gate 124 is always open and has a fixed length approximately equal to one free-space wavelength of th electromagnetic energy, as measured along a longitudinal axis of either waveguide 112 or 114. For lesser amounts of coupling, the length of the gate 124 is reduced, for example, to 0.8 waveguide for 6 dB coupling. The coupler 110 has two output terminals, shown as a through port 126 and a coupled port 128, and located at ends of the waveguides 112 and 114, respectively. The coupler 110 further comprises an input port 130 located at an end of the first waveguide 112 opposite the through port 126 and an isolation port 132 located at an end of the second waveguide opposite the coupled port 128. The isolation port 132 is shown connected schematically to a resistor 134 which represents a nonreflecting load having an impedance matched to that of the second waveguide 114. Such a load (not shown) is constructed typically in the form of a well-known wedge which absorbs electromagnetic energy at the operating frequency of the coupler 110, an is conveniently mounted within a section of waveguide (not shown) connected to the isolation port 132 by flanges (not shown). In use, the coupler 10 could be connected to components of a microwave circuit such as the network illustrated in FIG. 1; such components may include waveguide fittings which would be connected in a conventional manner, as by flanges (not shown) to the ports 126, 128, and 130 of the coupler 110.

The arrangement of the coupling gate 124 in the common sidewall 122 of the two waveguides 112 and 115 provides the configuration of a quadrature sidewall short slot hybrid coupler. Microwave signals coupled between the two waveguides via the gate 124 undergo a lagging 90° phase shift, this phase shift being inherent in the well-known operation of a quadrature sidewall short slot hybrid coupler. In many microwave circuits, including the disclosed embodiment of the combined uplink and downlink feed network illustrated in FIG. 1, such phase shift is unwanted, and some sort of phase compensation is required to equalize the phase between the microwave signals of the two waveguides 112 and 114.

The requisite phase compensation may be provided by use of a set of four capacitive irises 136 located in the first waveguide 112 beyond the gate 124, and a set of four inductive irises 138 located in the second waveguide 115 beyond the gate 124. The configuration of the capacitive irises 136 in the waveguide 112 constitutes a phase shifter 140 which introduces a lagging phase shift of 45° at the through port 126. The configuration of the inductive irises 138 in the waveguide 14 constitutes a phase shifter 142 which introduces a leading phase shift of 45° at the coupled port 128. The combination of the -90° shift introduced at the gate 124 with the +45° shift introduced by the shifter 142 provides a net -45° at the coupled port 128 which balances the -45° shift introduced by the shifter 140 at the through port 126.

In order to use the coupler 110 in certain situations, such as the microwave network 10 handling two-way communications via an antenna carried by satellite, the coupler 110 is constructed with a bandwidth wide enough to accommodate a transmit channel and a receive channel spaced apart in the frequency domain by an empty band to prevent cross talk between the two channels . The increased bandwidth of the coupler 110 is attained by use of stepped abutments 144 located at the outer sidewalls 120 on a center line of the gate 124. The abutments 144 reduce the width of the waveguides 112 and 114 at the gate 124.

Each of the abutments 144 is composed of three tiers having steps 146A-E and risers 148A-E. The dimensions of an abutment 144 may be adjusted to attain a desired bandwidth. Typical dimensions in terms of the free-space wavelength are as follows. The overall lengths is 1-1/4 wavelength, the step 146C is 1/2 wavelength, the steps 146B and 146D are each 1/4 wavelength, and the steps 146A and 146E are each 1/8 wavelength. The risers 148A and 148E are each 0.050 inch, the risers 148B and 148D are each 0.045 inch, and the risers 148C on both sides of the step 46C are each 0.060 inch. It is noted that each of the risers is less than 1/10 of a wavelength so as to minimize reflections from the abutments 144.

With respect to the construction of the phase shifter 140, the two center irises 136 have an equal height of 1/8 wavelength, this being 0.110 inch at the operating frequency of the coupler 110. The remaining two irises 136, at the ends of the set of irises, have an equal length of approximately 1/16 wavelength, the lengt measuring 0.080 inch at the operating frequency of the coupler 110, this being shorter than the height of the central irises 136. The thickness of each of the irises 136, as measured along the axis of the waveguide 112, is 1/8 wavelength. The spacing on centers between successive ones of the irises 136 is 1/4 of the guide waveguide. The width of each of the irises 136, as measured in a direction transverse to the waveguide axis, is approximately 0.2 inch. The length of the segment of the wall adjacent the capacitive irises 136 is 1.7 inch. The capacitive irises 136 are centrally spaced between the two sidewalls 120 and 122. While the capacitive irises 136 are shown as extending upwardly from the bottom wall 118, it is noted that, alternatively, they maybe constructed as extending downwardly from the top wall 116. With respect to the construction of the phase shifter 142, the two center inductive irises 138 extend from the outer sidewall 120 a distance of 0.115 inch, and the remaining two irises 138 at the outer ends of the set of irises extend from the sidewall 120 a shorter distance; namely, 0.110 inch. The spacing between centers of the inductive irises 138 is 1/4 of the guide wavelength. The thicknesses of the inductive irises 138, as measured along an axis of the waveguide 114, is approximately 1/8 free- space wavelength. Other dimensions of the coupler 110 are as follows.

The section of the common wall 122 adjacent the input port 130 measured 0.7 inch. The spacing between the sidewalls 120 and 122 in each of the waveguides 112 and 114 is 0.75 inch, this being approximately 3/4 wavelength. The overall length of the coupler 110 is 3.6 inches.

Both the abutments 144 and the inductive irises 138 extend, the full distance between the top wall 116 and the bottom wall 118. The desired phase shift and bandwidth has been obtained in the preferred embodiment by constructing the capacitive irises 136 with a width, as noted above, which extends only part way the two sidewalls 122 and 120 of the first waveguide 112.

In operation, the coupler 110 operates as a Ku-band sidewall short slot hybrid coupler with phase compensation introduced into the output terminals 126 and 128. The phase compensation is non-dispersive in frequency, and the phase shift structures permit the construction of the coupling device in a compact light-weight assembly for use in broadband power division networks. The capacitive phase shifter 140 introduces a phase shift of -45° at the through port 126. The inductive phase shifter 142 introduces a +45° phase shift in the second waveguide 114, which phase shift is algebraically combined with the -90° phase shift introduced by the hybrid coupling. The algebraic combination of the -45º phase and the 90° phase shift in the second waveguide 114 produces a resultant phase shift of -45º at the coupled port 128, this resultant phase shift being equal to the -45° phase shift at the through port 126. Thus, upon the application of radiant energy to the input port 130, the resultant electromagnetic waves exiting the through port 126 and the coupled port 128 are in phase with each other.

FIG. 6 shows frequency dispersive characteristics of the phase shifters 140 and 142. As is well known, the phase shift introduced by a phase shifter at one frequency differs somewhat from the phase shift introduced at another frequency. The coupler 110 is to be employed over a wide range of frequencies and, accordingly, any frequency dependency of phase shift must also be corrected to prevent distortion in the resultant antenna coverage pattern. While the nominal values of phase shift of the inductive iris 138 and the capacitive iris 136 are +45° and -45°, respectively, the actual values of phase shift vary from the nominal value as a function of frequency. As shown in FIG. 6, the inductive phase shifter 142 introduces a phase shift in excess of +45º at lower values of frequency, the value of phase shift dropping towards the nominal value for higher values of frequency. The phase shift introduced by the capacitive phase shifter 140 is smaller than the nominal value for lower values of frequency and increases to the nominal value at higher frequencies.

However, the difference between the phase shifts introduced by the series of inductive irises and the series of capacitive irises remains constant at 90º over the range of frequencies in the band of interest. Thus, the coupler 110 compensates for frequency induced variations in phase shift so as to provide for a broadband compensation of the inherent 90° phase shift associated with a hybrid coupler. As shown in FIG. 6, the upper trace for the series of inductive irises accurately tracks the lower trace representing the series of capacitive irises. Thereby, the phase compensation of the coupler 110 is free of frequency dispersion. This advantage is attained in conjunction with the mechanical benefit of reduced package size and reduced weight.

The horns 60-67 are also adapted for non-frequency dispersive operation over the effective antenna system aperture for the uplink and downlink frequency bands of interest. The horns 60-67 each introduce a phase delay to the received or transmitted signal whose value is a linear function of the signal frequency. To achieve the non- frequency dispersive operation, the horns 60-67 should each have a phase delay versus frequency function which is not only linear but of the same slope. One known way to achieve this performance is to employ equal-sized horns as the antenna system elements.

Horn antennas are well-known antenna array components. A typical horn antenna 10 is shown in the top view of FIG. 7 and has an overall length L_h equal to the sum of the flare length L_f and the waveguide length L_w. The horn aperture A measures the horn H-plane dimension. The throat of the horn has a dimension L_t. The axial length L_a of the horn is measured between the aperture and the intersection of the projected flared walls of the horn.

Co-pending patent application serial number_________, filed_________________, docket number PD-85175, entitled "Horn

Antenna Array Phase Matched Over Large Bandwidth," assigned to a common assignee with the present application, describes to an array of horn antennas having different aperture sizes in which the individual horns will phase track over a; wide frequency band. This array exploits the different phase slope characteristics of horn antennas and waveguide and may be advantageously used as the antennas 60-67 shown in FIG. 1.

For the rectangular aperture horn, the phase delay through the horn (its electrical length) is primarily determined by the H-plane dimension A, the horn length and the size of the horn throat opening. The phase slope characteristic is a measure of the phase delay of the horn per unit length of the horn. The phase slope is a constant for given aperture and throat dimensions irrespective of the horn length, and this characteristic is exploited by the horn optimization technique.

FIG. 8 illustrates the phase slope of two different horn antennas at two frequency boundaries (11.7 and 14.5 Ghz) of the frequency band of interest, one horn having a larger aperture, but each with the same overall length, bandwidth and center frequency. For purposes of description of the invention, the horn with the smaller aperture will be considered the reference horn. Line 220 illustrates the phase slope of the reference horn at the lower frequency, 11.7 Ghz. Line 225 illustrates the phase slope of the same horn at the upper frequency, 14.7 Ghz.

Lines 230 and 235 represent the phase slope of the second horn at the respective upper and lower frequencies, 11.7 Ghz and 14.5 Ghz. Because the aperture of the second horn is larger than the aperture of the reference horn, it has a longer electrical length than the first horn, and the phase delay through the second horn is larger than the phase delay through the reference horn.

For purpose of this example, it is assumed that the first horn depicted in FIG. 8 has a waveguide section length L equal to zero.

The phase slopes of standard waveguide sections whose cross-sectional configurations match those of the throats of the reference and second horn antennas are also depicted in FIG. 8 by lines 240 and 245, for the respective lower and upper frequencies of interest. For illustration of the invention, the respective phase delays of the waveguide sections equal in length to the reference horn are shown to equal, or are referenced to, the phase delay of the reference horn at the upper and lower frequencies of interest. It is noted that line 240, representing the waveguide phase slope referenced to the phase shift of the reference horn at the lower frequency, intersects line 230, the lower frequency phase slope of the second horn, at point A illustrated in FIG. 8. Line 245, representing the waveguide phase slope referenced to the phase shift of the reference horn at the upper frequency, intersects line 235, the high frequency phase slope of the second horn, at point B. It is significant that the two points A and B occur at substantially the same value of length "X" along the horizontal axis. As will be described, the value of X represents the optimized flare length L_f of the second horn and the corresponding waveguide length L_w = L_h - L_f necessary to optimize the second horn to phase track the reference horn. Thus, FIG. 8 represents the analytic solution for the determination of the lengths L_f and L_w, given the parameters of the required total phase slope of the optimized horn and the phase slopes of the nonoptimized horn flared section and the waveguide section. The solution represents the intersection of the two lines 235 and 245, and the two lines 230 and 240.

With the second horn having the flare length and waveguide length selected as described above, the phase slope of the waveguide section changes as the frequency changes so as to keep the value of X substantially equal to the same constant. As the frequency increases, the ideal flare length of a given flare section decreases, while the ideal length of the waveguide section increases, thereby compensating for the change in electrical length of the two sections. With the lengths of the waveguide and flared sections chosen appropriately, this mutual compensation results in the horn having a substantially constant electrical length over a wide frequency band. Therefore, horns of various aperture sizes and restricted to a maximum overall length can be phase matched over a band of frequencies by reducing the flare length of each horn relative to the flare length of the horn with the smallest aperture, with the difference in the overall horn length being made up in waveguide sections.

The optimization may be further illustrated with reference to the specific example illustrated in FIG. 9. In this example, the reference horn antenna has a phase delay of 700° at the center frequency of the band between 11.7 Ghz. and 14.5 Ghz, an overall length of 12 inches and a two inch aperture dimension. The second non-optimized horn antenna would have flare length and a phase delay of 800º at the same frequency, the same overall physical length as the reference horn, and a four inch aperture. The goal is to optimize the second horn so that its electrical length equals that of the reference horn over a wide frequency range, while maintaining the physical aperture and length dimensions of the second horn.

The phase slope of the reference horn is depicted by line 250 between the points having coordinates (X₁, Y₁) and (X₃, Y₃). The phase slope of the larger horn is depicted by line 255 between the points having coordinates (X₁, Y₁) and (X₂, Y₂). This slope ml is equal to Y₂/X₂, for the case where X₁ and Y₁ are zero. The phase slope m₂ of a standard waveguide section is shown as dotted line 260 extending between the points having coordinates (X₄, Y₄), and (X₃, Y₃). The slope m2 may be written as equal to (Y₄-Y₃) / (X₄-X₃). This phase slope m2 is also equal to 360º/λ_g, where λ_g represents the waveguide wavelength. Solution of the two equations defining the lines 255 and 260 having the respective slopes ml and m2 shown in FIG. 9 results in the solution for the value x = L_f, defining the flare length of the optimized horn with the four inch aperture. The equation relating the value of y to x for the line 255 having slope ml is given by Equation 1.

y = (m1)x (1)

The equation relating the value of y to x for line 260 having the slope m2 is given by Equation 2.

y - Y₄ + x(m2) (2)

Since Y₄ = Y₃ - (m2)X₃, Equations 1 and 2 may be solved for their intersection point x - L_f:

Lf = 1 (3)

The length of the waveguide section needed to complete the phase compensation is simply the horn length L_h minus the flare length L_f, with the overall horn length being equal to the overall length of the reference horn. The above calculations may be readily implemented by a digital computer to automate the design process. An exemplary program for the Basic programming language is given in Table I.

The example of FIG. 9 is further depicted in FIGS. 10A, 10B and 10C, which respectively show simplified top views of the reference horn (with no wavelength section), the larger aperture horn optimized by the present method at the lower frequency of interest (11.7 Ghz) and the larger aperture horn optimized by the present method at the upper frequency of interest (14.5 Ghz).

The reference horn with a two inch aperture has a total calculated electrical length equivalent to phase shifts of 3894.67° and 5002.09° at the respective upper and lower frequencies. The phase shift of the horn (non- optimized) having the four inch aperture is calculated as 4090.95° at 11.7 Ghz and 5155.83° at 14.5 Ghz. Thus, the phase dispersion between the two horns (without optimization) is 198.25° at the lower frequency, and 156.28° at the upper frequency.

Using the computer program shown in Table I, the horn design is optimized at 11.7 Ghz and at 14.5 Ghz. At the lower frequency (11.7 Ghz) , the flare length and waveguide length are calculated as 9.444 inches and 2.556 inches, respectively. This is illustrated in FIG. 10B, where the non-optimized horn is depicted in solid lines, and the optimized horn is depicted in dashed lines. At 11.7 Ghz, the flared section of the optimized horn has a calculated phase delay of 3219.58°, and the waveguide section has a total phase delay of 675.11°. Thus, the total phase delay of the optimized horn at 11.7 Ghz is 3894.69°, exactly equivalent to the calculated reference horn phase delay. At 14.5 Ghz, the flared section of the optimized horn has a calculated phase delay of 4057.64°, and the waveguide section has a phase delay of 949.50°. The total phase delay of the optimized horn at 14.5 Ghz is 5007.14°, which differs from the calculated reference horn phase delay at the same frequency by 5.05°.

Also using the computer program of Table I, the horn design is optimized at 14.5 Ghz. This results in slightly different calculated dimensions for L_f and L_w, 9.357 inches and 2.643 inches, respectively. This design is illustrated in FIG. 10C, where the non-optimized horn is depicted by the solid lines, and the optimized horn is depicted by the dashed lines. At 14.5 Ghz, the flared section of the optimized horn has a calculated phase delay of 4020.26°, and the waveguide section has a phase delay of 981.82°. Thus, the total phase delay through the optimized horn at 14.5 Ghz is 5002.09°, exactly equivalent to the calculated reference horn phase delay at this frequency. At 11.7 Ghz, the flared section of the optimized horn has a calculated phase delay of 3189.92° and the waveguide section has a phase delay of 698.02°. Thus, the total phase delay through the optimized horn of FIG. 4C at 11.7 Ghz is 3887.94°. This differs from the calculated reference horn phase for this frequency delay by 6.75°. The mutual phase compensation provided by the horn optimization is further illustrated from the respective phase delays of the flare and waveguide sections at the upper and lower frequencies for the two horn optimizations. The 2.643 inch waveguide section has a calculated phase delay of 981.82° at 14.5 Ghz, while the 2.556 inch waveguide section has a calculated phase delay of 949.50°, a difference of 32.32°. The corresponding 9.357 inch flare section has a phase delay of 4020.26° at the 14.5 Ghz, and the 9.444 inch flare section has a phase delay of 4057.64° at the same frequency, a difference of -37.38°. Summing the two differences (32.32°-37.38°) yields a total phase dispersion between the two horn optimizations at 14..5 Ghz of only -5.06°. Thus, the two horns optimized at different frequencies have virtually equal electrical lengths at 14.5 Ghz.

A similar comparison at the lower band edge (11.7 Ghz) yields a phase dispersion of -6.75°.

The calculated results for the optimizations at the upper and lower boundaries of this bandwidth indicate that slightly better phase tracking performance over the entire band is achieved when the horn is optimized at the lower frequency boundary. In practice, the frequency at which the horn is optimized will typically be between the lower frequency limit of the band and the mid-band frequency.

As is known to those skilled in the art, to avoid antenna pattern deterioration, the flare angle of the horn should be chosen to minimize the phase error across the aperture. The phase error across a horn with aperture A and axial length L_a is given by Equation 4:

ΔΦ = (2π/λ) (((A/2)² +L_a ²) ^½ - L_a) (4)

The maximum phase error should not exceed 90°, using Reyleigh's criterion. This places a restriction on the amount of phase compensation which may be achieved by the horn optimization technique.

The disclosed embodiment of the invention is useful for applications in which the transmit and receive beams have the same coverage area at the same time. Because only one feed network serves both the uplink and downlink system, the same level of performance available from separate networks may not be obtainable. This loss in performance results from the capability to separately optimize each uplink and downlink network for performance in the respective uplink and downlink frequency band. For many applications, however, the advantages of the single feed network in accordance with the invention outweigh the loss in performance. These advantages include the approximately fifty percent reduction in weight, fewer components, elimination of the need for separate diplexer devices at each horn, and consequent cost reductions.

A combined uplink/downlink feed system has been disclosed for coupling a satellite antenna system to the satellite receiver and transmitter. It is understood that the above-described embodiment is merely illustrative of the possible specific embodiments which can represent principles of the present invention. Other arrangements may be devised in accordance with these principles by those skilled in the art without departing from the scope of the invention.

Claims

CLAIMSWhat is claimed is:

1. An improved microwave feed network for coupling the elements of a satellite antenna system to the satellite receiver and transmitter, comprising: a frequency sensitive diplexer means coupled to said receiver and transmitter for separating the transmit and receive signals; and corporate feed network for coupling said diplexer means to the respective satellite antenna elements, said diplexer and said corporate feed network adapted for wideband operation over the satellite receive frequency band and the satellite transmit frequency band.

2. The invention of Claim 1 wherein the satellite receive and transmit beam patterns coincide.

3. The invention of Claim 1 wherein said corporate feed network comprises a plurality of antenna feed ports for coupling to the respective antenna elements and a diplexer feed port for coupling to the diplexer means, and wherein said corporate feed network is adapted to introduce predetermined phase shifts and attenuation between said diplexer port and the respective antenna feed ports in accordance with characteristics of the satellite beam coverage.

4. The invention of Claim 1 wherein said antenna system comprises a plurality of antenna elements adapted for non-frequency dispersive operation over said receive and transmit frequency bands.

5. The invention of Claim 4 wherein said antenna elements comprise a plurality of antenna horns whose electrical lengths are substantially equal over said satellite receive and transmit frequency bands.

6. The invention of Claim 1 wherein said corporate feed network comprises a plurality of phase compensated waveguide hybrid couplers which are adapted for non- frequency dispersive operation over said respective receive and transmit frequency bands.

7. In a communication satellite adapted to receive uplink signals in an uplink frequency bandwidth and to transmit downlink signals in a downlink frequency bandwidth over a satellite antenna system comprising a plurality of non-frequency dispersive antenna elements, the improvement comprising: a combined uplink/downlink feed network for coupling the satellite receiver and transmitter to the satellite antenna system, and wherein the elements of said network are adapted for broadband non-frequency dispersive, operation over said uplink and downlink frequency bands.

8. The improvement of Claim 7 wherein said feed network comprises a diplexer means coupled to said receiver and to said transmitter for isolating the uplink and downlink frequencies so that the uplink signals are not coupled into the transmitter and the downlink signals are not coupled into the receiver.

9. The improvement of Claim 8 wherein said feed network further comprises a corporate feed network having a plurality of antenna feed ports for coupling to the respective antenna elements and a diplexer port for coupling to said diplexer, said corporate feed network for distributing the respective uplink and downlink signals between said antenna ports and said diplexer port to achieve the desired satellite beam coverage.

10. The improvement of Claim 9 wherein said corporate feed network comprises means for introducing predetermined phase shifts and attenuation between said diplexer port and said respective antenna feed ports, said respective phase shifts and attenuation having substantially constant values over said uplink and downlink frequency bands.

11. The improvement of Claim 9 wherein said corporate feed network comprises a plurality of phase compensated waveguide hybrid couplers which are adapted for non-frequency dispersive operation over said uplink and downlink frequency bands.

12. A combined feed network for coupling the receiver and transmitter devices of a communications satellite to the satellite antennas in a non-frequency dispersive system, comprising: a frequency sensitive diplexer coupled to the receiver and transmitter for separating the satellite receive signals in an uplink frequency band and the satellite transmit signals in a downlink frequency band; and a wideband corporate feed network having a plurality of antenna ports for coupling to said satellite antennas and a diplexer port for coupling to said diplexer, and wherein said corporate feed network is adapted to introduce predetermined respective phase shifts and attenuation to signals in the uplink and downlink frequency bands between said respective antenna ports and said diplexer port.