EP0275303B1

EP0275303B1 - Low sidelobe solid state phased array antenna apparatus

Info

Publication number: EP0275303B1
Application number: EP19870905342
Authority: EP
Inventors: Jar Jueh Lee; Raymond Tang
Original assignee: Hughes Aircraft Co
Current assignee: Raytheon Co
Priority date: 1986-07-29
Filing date: 1987-07-21
Publication date: 1993-10-13
Anticipated expiration: 2007-07-21
Also published as: EP0275303A1; JPH01500476A; DE3787797D1; DE3787797T2; WO1988001106A1

Abstract

A low sidelobe, solid state array antenna apparatus comprises a large radiating aperture divided into a large number, N, of small, closely spaced radiating apertures, each small radiating aperture having associated therewith a radiating element and a linearly polarized solid state power module. The large radiating aperture is divided into M, preferably between (3) and about (10), differently sized, elliptically shaped, concentric radiating zones superimposed, for analysis purposes, upon another. Each such zone has an output voltage amplitude, Ei, and semi-major and semi-minor axes of respective lengths, ai and bi, each zone being considered separately in the far field equation: G(,PHI) = [f(PHI) (â cos PHI - âPHI sin PHI cos )]2, wherein f(,PHI) = (I), ui = (II), J1(ui) is the first order Bessel function, â and âPHI are unit vectors in the spherical coordinates and Ko is the wave number associated with the radiated field. Using the far field equation, values of Ei, ai and bi for each zone are computed which result in the far field sidelobe peak gain being a minimum or being a specified number of dB, for example, at least about 30 dB, below the far field mainlobe gain. The values of Ei in overlapping zones are summed to establish the required voltage amplitudes of the underlying power modules associated with the N radiation apertures.

Description

The present invention relates generally to a solid state, phased array antenna according to the preamble of claim 1.
Radar antennas are well known to radiate microwave radiation in a broad pattern which, for directed antenna, includes a narrow mainlobe and wide sidelobes of radiation. By common definition, the mainlobe is the central lobe of a directional antenna's radiation pattern, the sidelobes referring to the lesser lobes of progressively decreasing amplitude on both sides of the mainlobe and often extending rearwardly of the mainlobe.
Radar antenna aperture configuration generally determines the extent and relative magnitude of the associated sidelobes; however, the gain of the strongest one of the sidelobes is typically only about 1/64 that of the mainlobe. In terms of decibels, the strongest sidelobe gain is typically down about 18dB from the associated mainlobe gain. Gains of the other sidelobes are usually considerably smaller than that of the strongest sidelobe. Although sidelobe gain is typically much smaller than mainlobe gain, because of the large solid angle into which sidelobes radiate, as compared to the small solid angle into which the mainlobe radiates, typically about 25 percent of the total power radiated by a uniformly illuminated radar antenna are in the sidelobes.
Ordinarily, sidelobe radiation provides no useful function and in addition to representing wasted radiating power has other serious disadvantages. For example, radar clutter from sidelobe returns increases the difficultly of discriminating targets from background. Another very significant disadvantage of sidelobe radiation is that such radiation can, in a military environment, be utilized by hostile forces for electronically jamming the radar and can also be used for positionally locating and for guiding munitions to the radar. In this regard, although mainlobe radiation is ordinarily much greater than sidelobe radiation, its relatively small solid angle of radiation and its directionality makes mainlobe jamming, radar location and munitions direction more difficult.
For these and other reasons, the reduction or suppression of radar sidelobe radiation is, particularly in military radar, important and military procurement documents establishing rigid limits on sidelobe radiation are not uncommon.
According to the preamble of claim 1 the present invention is starting from a prior art solid state phased array antenna as it is known from "IEEE MTT-S International Microwave Symposium Digest", 15-17 June 1982, Dallas, Texas, IEEE (New York, US), D.N. McQuiddy Jr; "Solid state radar's path to GaAs", pages 176-178. This known solid state phased array antenna is comprising an antenna aperture formed of a large number of N individual, closely spaced radiating apertures and a number of N individual radiating elements, each of which is operatively associated with a corresponding one of the a.m. radiating apertures for radiating microwave energy therethrough. There are further provided a number of individual solid state power modules in the form of "transient balanced power amplifiers", each of which is operatively associated with a corresponding one of the radiating elements for providing power thereto depending on the output voltage amplitude of these power amplifiers.
The use of an individual solid state power module for each radiating element or for at least a subgroup of radiating elements - thus forming a so called "active" array - offers the advantage of an improved cooling ability over "passive" arrays in which there is provided only one central power supply; in addition, within a large active array, a comparatively large number of power modules can fail or malfunction without impairing the effectiveness of the antenna.
A general problem of phased array antennas of that type is the suppression of sidelobe radiation; that is to say, the gain of sidelobe radiation should be much less than the gain of the mainlobe radiation. This is because the sidelobe radiation not only is a waste of radiating energy but also has other serious disadvantageous such as increased difficulty of discriminating targets. It, therefore, has already been proposed to suppress such sidelobe radiation in passive array antennas by "tapering" the illumination over the aperture so that individual radiating elements near the side edges of the array radiate less energy than elements which are closer to the center.
It has already been proposed to adopt the a.m. scheme of tapering to active arrays by using a very large number of power modules which each have a different power output and are arranged to simulate a tapering. In practice, however, it has revealed to be impossible to use this method because there are needed much more than 20 different types of power modules; such a great number of different types of power modules, however, is increasing substantially the production costs of the array and, what is even more disadvantageous, causes subsequent maintenance and logistical support problems.
It therefore is the object of the present invention to improve an active phased array antenna according to the preamble of claim 1 in such a way that there can be achieved an excellent suppression of sidelobe radiation without substantially increasing production costs or causing maintenance and logistical support problems.
According to the present invention this object is solved by the advantageous measures indicated in the characterizing part of claim 1.
Hence, according to the present invention, the power modules are subdivided in a number of M groups, the number M being significantly less than the number N, these M groups of the power modules furthermore being arranged in a concentric pattern (see Fig.4) around a central point of the array, wherein the output voltage amplitude of the power modules is equal within each group, but different in different groups. It has been found that by the a.m. measures it is possible to easily reduce the sidelobe gain to be at least 30 dB below the mainlobe gain by selecting the output voltage amplitude of each group of power modules and by in combination selecting a pattern dimension of each group. As in this way a sufficient sidelobe suppression can be obtain by merely using between 3 and 10 different types of power modules, neither production costs are substantially increased nor maintenance and logistical support problems are caused.
Advantageous developments of the present invention are indicated in the subclaims 2 through 5; a preferred process for configuring an array antenna of that type is the subject-matter of claims 6 through 10.
In US-A-3 760 345 there is disclosed a transducer array for receiving or transmitting acoustical or electromagnetic signals by means of uniform square transducer elements which each are coupled to associated shading resistances. This known array, however, is of the passive type and, consequently, has not a plurality of power modules in the sense of feature [c] of the preamble of claim 1.
A better understanding of the present invention may be had by considerating the accompanying drawings in which:

FIG. 1 is an exploded perspective of an exemplary solid state, active array antenna with which the present invention may be used to advantage;
FIG. 2 is a pictorial drawing of the radiation pattern of a typical airborne radar, showing mainlobe and sidelobe portions of the radiation pattern;
FIG. 3 is a diagram depicting the coordinate system used to specify the coordinatee of the far field relative to an radiating antenna;
FIG. 4 is a diagram depicting the manner in which a generally rectangular solid state active array antenna is divided into a series of M concentric, overlapping elliptical power module zones, each such zone having a different power level;
FIG. 5 is a diagram showing, relative to an array cross-section taken generally along line 5-5 of FIG. 4, how the aperature illumination taper is provided by superimposing different voltage levels of power modules in the different module zones of FIG. 4;
FIG. 6 is a diagram, similar to right hand portions of the diagram of FIG. 5, showing, for a particular array configuration and sidelobe radiation requirement, normalized power levels for five power module zones, the corresponding, normalized zone boundary dimensions being also indicated;
FIG. 7 is a graph plotting far field mainlobe and sidelobe gain vs angle from broadside axis for the conditions shown in FIG. 6; idealized, elliptical aperture zones being assumed; and
FIG. 8 is a graph plotting far field mainlobe and sidelobe gain vs angle from broadside axis for conditions in which stepped zone boundaries corresponding to actual module lattice configuration are assumed.

DESCRIPTION OF THE PREFERRED EMBODIMENT

There is shown in FIG. 1, in exploded form, an exemplary, solid state, active array antenna 10 of the general type with which the present invention may be used to advantage. Comprising antenna 10, which is shown as an aircraft-mounted type, are an aperture assembly 12, a cooling liquid plate assembly 14, a solid state power module assembly 16 and a stripline feed assembly 18. Included in aperture assembly 12 is a large number of small radiating elements 24, each of which has disposed therein a dielectric filler 26. Defined in a face 28 of aperture assembly 12 is a large number of openings 30, each of such openings being associated with one of radiating elements 24. Mounted on cooling plate assembly 14 are a number of loop assemblies 32, each of which is also associated with one of radiating elements 24. A large number of solid state power modules 34 comprise power module assembly 16, each such module preferably, but not necessarily, powering only a single associated radiating element 24.
The present invention is principally directed towards providing preselected voltage operating levels of power modules (corresponding to modules 34) and the physical arrangement of such modules in an assembly (corresponding to module assembly 16) so that the far field radiation from the antenna exhibits very low sidelobes. With respect to sidelobes, FIG. 2 illustrates a typical radiation pattern 38 associated with a radar carried by an aircraft 40. The airborne radar involved may, for example, comprise a solid state active array similar to array 10 depicted in FIG. 1. As shown in FIG. 2, radiation pattern 38 comprises a narrow, beam-shaped mainlobe 42 and smaller, fan-shaped sidelobes 44 on each side of the mainlobe. Sidelobes 44 comprise several different lobes 46 which fan out at different angles, α, relative to a main beam axis 48; typically the sidelobes diminish in intensity as the angle, α increases. It can further be seen from FIG. 2 that some of lobes 46 extend rearwardly relative to mainlobe 42, the angles, α, associated therewith being greater than 90°.
As more particularly described below, the present invention relates to a process for configuring a solid state, active array so that the far field sidelobe gain is at least 30dB down, from the far field mainlobe gain. In general, the reduced sidelobes provided by the present invention is accomplished by tapering the radiating illumination in a relatively few, precisely determined steps.
For purposes of further describing the invention, the more general case of a rectangular, solid state active array 60, depicted in FIGS. 3-5, is considered. Array 60 corresponds generally to array 10 (FIG. 1), insofar as general construction is concerned.
Also, for purposes of illustrating the invention, it may be assumed that array 60 has rectangular dimensions 2a and 2b, and has R rows and C columns of linearly polarized, rectangular radiating elements 62. Associated with element 6z is a power module 64 (shown in phantom lines).
It is, however, assumed, for purposes of simplifying the following computations, that array 60 has an elliptically (instead of a rectangular) radiating aperture 66, it having been determined by the present inventors that array corner regions 68 contribute only negligibly to sidelobes. For purposes of the following description, the far field, G, associated with radiating aperture 66 is considered, the far field at any point defined by angles ϑ and φ being generally identified as G(ϑ,φ) in FIG. 3.
A principal feature of the present invention is the dividing, for analysis purposes, of radiating aperture 66 into a relatively few, superimposed elliptical zones around a central point "A", and the selection of zone boundary axes a_i, b_i and the zone voltage amplitudes, E_i, associated therewith in a manner providing a tapered illumination of the aperture which assures very low, far field sidelobes.
Preferably the number of elliptical zones selected varies between 3 and about 10 and more preferably between 3 and only about 7. Insufficient illumination tapering is considered to be provided using less than 3 zones and although smoother tapering can be provided by use of more than about 7 zones, the cost of using more than that number of different types of power modules is costly and has moreover, been found by the present inventors to be unnecessary for achieving very low sidelobes. For specific purposes of illustrating the invention, the number of zones shown and described is 5; however, any limitation to the use of about 5 zones is neither intended nor implied.
First through fifth concentric, progressively larger elliptical zones 74, 76, 78, 80 and 82, respectively, are thus selected, the zones having semi-major and semi-minor axes equal, respectively, to a₁, a₂, a₃, a₄, and a₅ and b₁, b₂ b₃, b₄, and b₅ (FIG. 4). First zone 74 is the smallest zone and fifth zone 82 is the largest zone and completely fills aperture 66, dimensions a₅ and b₅ being, therfore, respectfully equal to aperture dimensions a and b (FIG. 3).
As can be seen from FIG. 5, which corresponds to a transverse output voltage cross-section of array 60, zones 74, 76, 78, 80 and 82 are, for analysis purposes, considered as stacked (or superimposed) upon one another, with the fifth, largest zone 82 at the bottom and the first, smallest zone 74 at the top. Associated with each zone 74, 76, 78 and 80 and 82 is a different voltage amplitudes, E_i, amplitude E₁ being associated with zone 74, E₂ with zone 76, E₃ with zone 78, E₄ with zone 80 and E₅ with zone 82. In regions where two or more zones 74-82 overlap, the voltage amplitudes, E_i, are added to establish power module voltage. For example, in a central, elliptical region 84, defined by first zone 74, the combined voltage amplitude of the stacked zones 74-82 required to be provided by underlying power modules 60 is equal to $E₁ + E₂ + E₃ + E₄ + E₅$
. In an annular region 86 of second zone 76 outside of first zone 74, the voltage amplitude required to be provided by underlying power modules 60 is equal to $E₂ + E₃ + E₄ + E₅$

; in an annular region 88 of third zone 78 outside of second zone 74, the voltage amplitude required to be provided by the underlying power modules is equal to $E₃ + E₄ + E₅$
. In turn, in an annular region 90 of fourth zone 80 outside of zone 78, the voltage required to be provided by underlying power modules 60 is E₄ + E₅; outside of zone 80, in an annular region 92 of fifth zone 82, underlying power modules 60 are required to provide a voltages amplitude equal only to E₅. However, by known principles of superposition, each zone 74-82 can be treated separately as providing only a single, corresponding voltage amplitude E₁-E₅.
The present process treats all zone axis dimensions, a_i, b_i, and zone voltage amplitudes, E_i, as independent variables. At least one set of values for these variables is computed which will provide, as may be required, either minimum sidelobes or a sidelobe gain which is a preselected number of dB less than the corresponding mainlobe gain. These independent variables a_i, b_i and E_i are computed, for numerous G(ϑ, φ) points, by the equation:

${G(ϑ,φ) = [f(ϑ,φ) (â}_{ϑ} {cos φ - â}_{φ} sin φ cos ϑ)]², (1)$

$u_{i} {= (k}_{o} a_{i} sin ϑ) √ \bar{{cos²φ + (b}_{i} {²/a}_{i} ²) sin² φ,} (3)$

and further wherein J₁ (u_i) is the first order Bessel function, k_o is the wave number associated with the radiation and â_ϑ and â_φ are the unit vectors in the sperical coordinate system.
To determine the optimum set of parameters (a_i, b_i, E_i) for low sidelobes, standard techniques of gradient search can be employed. In the optimization process an initial set of parameters is chosen as a starting point, and a present maximum sidelobe level (such as -30 dB) is selected as a performance criterion. Then the antenna far field pattern with the initial set of input parameters can be calculated by using Equation (1). Next the total power of all the sidelobes that exceed the present level, being defined as the error, is computed. After this a small variation of one of the parameters, either a positive or negative increment, is introduced and the error is recomputed. By examining the trend of the error, and hence the gradient (rate of change), one can decide which way the following step of variation should be implemented. The process is repeated for this parameter until a local minimum in the error is obtained. By the same procedure the iteration process is carried out for all other parameters until the error is reduced to an acceptable level. This optimization process can be readily accomplished by using a computer. By way of specific example, again with no limitations being thereby intended or implied, the present inventors have determined for M equal to 5 (that is, for five aperture zones), the optimum zone boundaries, a_i, b_i, and output voltage amplitudes, E_i. These values are shown below in Table 1, wherein $a = a₅ = 1.3 meters$
and $b = b₅ = .87 meters$
, the sum of $E₁ + E₂ + E₃ + E₄ + E₅$
is normalized to 1.0 and the radiation frequency is 3.25 GHz. Furthermore, for simplicity of mathematical derivation, the aspect ratio, b_i/a_i, for each zone is identical to that of each other zone. TABLE 1

a₁ .44 m

a₂ .68 m

a₃ .88 m

a₄ 1.01 m

a₅ 1.3 m

b₁ .30 m

b₂ .46 m

b₃ .60 m

b₄ .68 m

b₅ .87 m

E₁ 0.26

E₂ 0.22

E₃ 0.16

E₄ 0.16

E₅ 0.20
FIG. 6, directly corresponds to the righthand half of FIG. 5 and depicts, relatively to scale and for the b_i dimensions normalized to $b = b₅ = 1$
, the corresponding, computed voltage amplitude, E_i, for each of the five zones 74, 76, 18, 80 and 82. Also shown in FIG. 6 is the dB value associated with the difference in power level across each boundary: 2.62 dB with zone 74, 3.06 dB with zone 76, 3.1 dB with zone 78 and 5.11 dB with zone 80.
For the computed a_i, b_i, E_i values listed in Table 1, there is plotted in FIG. 7 antenna pattern gain (in dB) against elevation angle, ϑ as measured from the broadside axis. From FIG 7 it can be seen that the gains of all sidelobes 46 (shown shaded) are down at least about 36dB from the peak (0°) gain of mainlobe 42 over the entire visible radiation range.
In the foregoing, it has been assumed, for computations involving Equation 1, that the boundaries of the five elliptical zones 74, 76, 78, 80 and 82 are perfectly elliptical, as would be the case if there were an infinite number of infinitely small power modules 64 distributed over antenna elements 62. In reality, however, each radiating zone intersects a finite, though usually large, number of radiating elements 62 so that the zone boundaries are more accurately approximated by a discontinuous, stepped shape, (FIG. 4). The question then arises as to which of two adjacent zones the intersected radiating elements 62 (and corresponding power modules 64) should be allocated and also whether allocation to one zone or another makes any significant difference with respect to sidelobe gain reduction.
To answer this question, a specific array pattern, with actual element spacing and lattice structure taken into account, was used by the present inventors to compute aperture zone parameters a_i and b_i and voltage amplitudes, E_i. For such purposed, the actual geometric configuration of a proposed solid state radar array, having an array size of 2.6 by 1.75 meters and having 1188 rectangular radiating elements, was assumed. It was futher assumed that the zone boundaries followed actual boundries of the radiating apertures. Values of a_i, b_i and E_i for minimum sidelobes were obtained for such an array configuration by operation of Equation 1. The computed gain VS elevation angle is plotted in FIG. 8 which shows that the highest sidelobe gain is down at least about 37 dB from the peak mainlobe gain. A comparison of FIGs. 7 and 8 thus reveals that although the sidelobe pattern is slightly different in actual conditions (FIG. 8) as compared to that of the idealized conditions (FIG. 7), the sidelobe gains are nevertheless about the same in both cases.

Claims

A low sidelobe, solid state, phased array antenna apparatus having a far field mainlobe and a sidelobe radiation pattern, comprising:
[a] an antenna aperture (12) formed of a large number of N individual, closely spaced radiating apertures (30);

[b] a number of individual radiating elements (24), each of which is operatively associated with a corresponding one of said N radiating apertures (12) for radiating microwave energy therethrough; and

[c] a number of individual solid state power modules (34), each of which is operatively associated with at least one of said N radiating elements (24) for providing power thereto depending on the output voltage amplitude (E) of the respective solid state power module (34);
characterized in that

[d] said power modules (34) are subdivided in a number of M groups, the number M preferably being between 3 and 10 and being significantly less than said number N;

[d.1] said M groups of said power modules (34) are arranged in a concentric pattern (Fig.4) around a central point (A) of said array;

[d.2] said output voltage amplitude (E) of said solid state power modules (34) is equal within each group, but different in different groups; and in that

[e] said output voltage amplitude (E) of each group and a pattern dimension of each group in combination are selected such that the sidelobe gain is at least 30 dB below the mainlobe gain.
The array antenna as claimed in Claim 1, wherein the number M is about 5.
The array antenna as claimed in Claim 1 or 2, wherein the outer boundary of each of said M groups of power modules is elliptically shaped, each of said boundarys having a semi-major axis of length a_i and a semi-minor axis of length b_i, wherein the subscript "i" refers to the ith boundary.
The array antenna as claimed in Claim 3, wherein the output voltage amplitudes and the arrangement of said M groups of power modules are selected by treating the M module group arrangements as comprising a superposition of M elliptically shaped, overlapping zones having the same boundaries as corresponding ones of said M groups of modules, each of said M zones having associated therewith a different voltage amplitude E_i, the voltage amplitude of the power modules in each of said M groups being selected by adding the different voltage amplitudes, E_i, of the corresponding overlapping zones, wherein the subscript "i" refers to the ith zone.
The array antenna as claimed in Claim 4, wherein the voltage amplitudes, E_i, and semi-axis lengths, a_i and b_i, are selected by application of the following far field equation to cause the sidelobe gain to be down at least about 30dB from the mainlobe gain:

${G(ϑ,φ) = [f(ϑ,φ) (â}_{ϑ} {cos φ - â}_{φ} sin φ cos ϑ)]²,$

$u_{i} {= (k}_{o} a_{i} sin ϑ) √ \bar{{cos²φ + (b}_{i} {²/a}_{i} ²) sin²φ,}$

where: J₁ (u_i) is the first order Bessel function, â_ϑ and â_φ are unit vectors in the spherical coordinate system and k_o is the wave number associated with the radiated field.
A process for configuring an array antenna according to the preamble of claim 1,
characterized by the steps of:
[1] subdividing said power modules (34) in a number of M groups, the number M preferably being between 3 and 10 and being significantly less than said number N;

[1.1] arranging said M groups of said power modules (34) in a concentric pattern (Fig.4) around a central point (A) of said array;

[1.2] making said output voltage amplitude (E) of said power modules (34) equal within each group, but different in different groups; and

[2] selecting said output voltage amplitude (E) of each group and a pattern dimension of each group in combination such that the sidelobe gain is at least about 30 dB below the mainlobe gain.
The process as claimed in Claim 6, wherein the number M is about 5.
The process as claimed in Claim 6 or 7, including the step of arranging said M groups of power modules so that the outer boundaries thereof are substantially elliptically shaped, each boundary having a semi-major axis of length a_i and a semi-minor axis of length b_i, wherein the subscript "i" refers to the ith boundary.
The process as claimed in Claim 8, including the steps of :
[1] treating said M groups of power modules as comprising a superposition of M elliptically shaped, overlapping zones having the same boundaries as corresponding ones of said M groups of modules, each of said M zones having associated therewith a voltage amplitude, E_i, and

[2] treating the voltage amplitude of the power modules in each of said M groups of power modules as an additive superposition of the voltage amplitudes, E_i, of the corresponding overlapping zones, wherein the subscript "i" refers to the ith zone.
The process as claimed in claim 9, including the step of using the following far field equation to obtain values for the zone voltages amplitudes, E_i, and the zone semi-major and semi-minor axis lengths, a_i and b_i, which cause the far field sidelobe gain to be down at least about 30dB from the corresponding far field mainlobe gain:

${G(ϑ,φ) = [f(ϑ,φ) (â}_{ϑ} {cos φ - â}_{φ} sin φ cos ϑ)]²,$

$u_{i} {= (k}_{o} a_{i} sin ϑ) √ \bar{{cos²φ + (b}_{i} {²/a}_{i} ²) sin² φ,}$

where : J₁ ^(ui⁾ is the first order Bessel function, â_ϑ and â_φ are unit vectors in the sperical coordinate and k_o is the wave number associated with the radiated field.