JPH0479161B2 - - Google Patents

Description

[Detailed Description of the Invention] [Technical Field of the Invention] This invention provides a method for arranging a plurality of element antennas,
According to the given array density function, the side lobe level, which is important as an antenna characteristic, can be reduced by exciting or not exciting the element antenna depending on the array position of the element antenna. The present invention relates to a method for thinning out array antennas.

[Prior art]

An array antenna that is excited or not excited by element antennas is hereinafter referred to as a density taper array. There are two types of conventional density taper arrays, one in which excitation or non-excitation is determined definitively, and the other in which the excitation or non-excitation is determined based on statistical probability.

An example of the former will be explained using FIG. In FIG. 1, 1 is an element antenna arranged in a straight line,
It constitutes a linear array antenna. The excitation amplitude of each element antenna 1 is uniform. In a uniformly distributed linear array antenna, if the element spacing is constant, the sidelobe level is approximately _-13.5dB , making it unsuitable for antennas that require low sidelobe characteristics, such as radar antennas. .
Normally, in order to achieve low sidelobes in an array antenna, distribution (or taper, weight) is applied to the excitation amplitude of each element antenna. However, in active array antennas where each element antenna is equipped with a relatively low-output semiconductor oscillator and synthesizes large power in space, in order to effectively use the output of each oscillator, the maximum output and Therefore, the excitation amplitude of the array antenna must be equal. For this reason, conventional efforts have been made to provide a density distribution to the array of element antennas in order to reduce side lobes.

In Fig. 1, depending on the given array density function f (used in array antennas with amplitude distribution. For example, Taylor distribution is used), as shown in Fig. 1, the area A surrounded by the function is The array axis of the array antenna is divided into equal parts, and element antennas 1 are arranged in each divided range. As a result, as shown in FIG. 1, the density of the element arrangement is denser in the center and thinner toward both ends.

However, the conventional density taper array shown in Fig. 1 is superior in practice in that the arrangement position of the element antenna 1 is uniquely determined, but as is clear from Fig. 1, the arrangement interval of the element antenna 1 is are unevenly spaced. In an array antenna that is actually used, the physical shape and dimensions of each element antenna 1 are usually the same, and the element arrangement spacing is usually the same from the viewpoint of uniformity of characteristics, mass production, and economic efficiency. First, generally, the array positions or array grid points where the element antennas are to be placed are given at equal intervals. Therefore, in order to approximate the arrangement position of the element antenna, which has been optimally determined as shown in FIG. 1, by moving it to an arrangement position that can actually be placed, the arrangement position of the element antenna is shifted from the optimum position.

In addition, in the second type of conventional density taper array, which is based on a statistical method, the probability density proportional to the desired density function value is set as the probability of existence of an element antenna at each element antenna array position, and the probability density is calculated by probability operation. , the excitation or non-excitation of each element antenna (instead of non-excitation, the element antenna may be thinned out in order to reduce the weight of the antenna), and the desired low side lobe is determined. It has the disadvantage that the characteristics can only be obtained by statistical average.
That is, although the density taper array of this statistical method may exhibit relatively good characteristics for large-aperture array antennas with several thousand to tens of thousands of element antennas, it is not suitable for small-aperture array antennas.

In addition to these typical formats, in the density taper array obtained in the above format, the thinning positions of the element antennas are artificially changed, the radiation pattern is calculated each time, and the characteristics are improved. There is also a trial-and-error type density taper array that is obtained through repeated operations, but it is impossible to determine the optimal arrangement of element antennas, and it requires a huge amount of design time.

[Summary of the invention]

In order to eliminate these drawbacks, this invention
By cumulatively adding the values obtained by normalizing the maximum value of the array density function to 1 according to the order of the array positions, and converting this into an integer value and evaluating it, a thinning function with a value of 0 or 1 can be uniquely determined. Based on this thinning function, excitation or non-excitation of the element antenna at a given element antenna array position is determined to configure the element antenna array.The drawings will be described in detail below.

[Embodiments of the invention]

FIG. 2 is a schematic diagram of a circular aperture array antenna showing one embodiment of the present invention.

In FIG. 2, 2 is a circular aperture in which the element antennas are arranged to form an array antenna, and 3 is an array lattice point provided in advance on the circular aperture 2, on which the element antennas are to be arranged. . (Only a part of the lattice points are shown in FIG. 2.) It is assumed that an array density function f to be realized is given on the surface of this circular aperture 2. When appropriately ordering each grid point 3 (how to order will be described later), the weight at position p _i of the i-th grid point is defined as f _i =cf(p _i ) (1). However, c is a constant for normalizing the maximum value of f _i to 1. Here, as shown in Figure 3, the values of f _i are cumulatively added according to the above order, that is, a cumulative weight is created, the integer value closest to this is selected, and this result is used to , a thinning function T(p _i ) as shown below can be created.
That is, T (p _i ) = [( _i 〓 ^j=1 f _j ) + 0.5] - [( _i-1 〓 ^j=1 f _j ) + 0.5] (2) However, [ ] is a Gaussian symbol. Yes, only the integer part of the number is taken. T(p _i ) in equation (2) takes a value of 1 or 0 (excluding unrealistic extreme discontinuous distribution functions), and when T(p _i )=1, p _i is connected to the element antenna. When T(p _i )=0, the element at the position p _i is not excited or is thinned out. In this way, the difference between the cumulative value of weights and the cumulative number of element antennas is always 1/2 or less, and an optimal density distribution can be achieved with an error of 1/2 or less.

As the simplest example of the order to give p _i ,
There is an order of f _i values from large to small. However, a more effective ordering method and arrangement format will be described below.

The radiation pattern within a certain observation plane is determined by the excitation distribution of the array antenna projected on the intersection line between that plane and the array antenna aperture plane, and by the distribution of the number of excitation element antennas (hereinafter referred to as element number distribution) in the case of a density taper array. Now, in Figure 2, if you want to make the radiation pattern in the observation plane including the X axis into a desired shape, the coordinates of _{p i are} (X, Y)
Then, the lexicographical order of (X, Y), i.e.
Order X from small to large, and for the same X, Y
may be ordered from small to large or from large to small. Or the order of X from large to small, and the same
In this case, Y may be ordered from small to large or from large to small.

Similarly, in order to bring the radiation pattern in the Y-axis plane in FIG. 2 closest to the desired pattern,
For example, the lexicographical order for (Y-X) may be used by replacing Y with Y. In this way,
The radiation pattern closest to the ideal can be obtained in the plane or the Y plane (generally, any one observation plane).

Next, we will describe the array format of the element antennas that makes the radiation patterns on the two planes, the X plane and the Y plane, as close to ideal as possible.

That is, first, regarding the X plane and the Y plane,
A thinning function for making the radiation pattern closest to the ideal, that is, using the above-mentioned ordering, is determined by the second equation. These are respectively T _X ,
Let it be T _Y. Ideal distributions P _Xj and P _Yk on the X and Y axes are determined from these T _X and T _Y. P _Xj represents the number of excited element antennas at the j-th grid point on the X-axis. On the other hand, the element number distribution Q _Yk on the Y axis on the opposite side is obtained from the thinning function T _X obtained by considering the optimization of the X plane. Here, D _k =Q _Yk -P _Yk (3) is the difference between T _X and T _Y , and the sum is zero.
Here, these matters will be explained using FIG. 4. FIG. 4 shows the distribution of excitation (1) and non-excitation (0) of the element antenna on the (X, Y) plane when focusing on the X plane and idealizing the distribution of the number of elements on the Y axis. The resulting distribution of the number of elements on the Y axis is indicated by Q _Yk . On the other hand, focusing on the Y plane, the distribution of the number of elements on the Y axis is shown as P _Yk . Here, focusing on the difference D _k between Q _Yk and P _Yk is 1 and -1,
For the 1 row and the -1 row, for p _i with the same X value, find a pair in which the former has 1 and the latter has 0, and exchange the former's 1 with the latter's 0. In FIG. 4, there are two such sets (indicated by arrows), but in this case, only one set needs to be replaced. By this exchange, without changing the distribution in the X direction,
The distribution in the Y direction can be made into an ideal distribution.

Furthermore, in the above procedure, if the desired set does not exist, as shown in Figure 5, perform the exchange shown by the arrow that does not affect the distribution in both the X direction and the Y direction, and then repeat the above procedure. It is also possible to find pairs that meet the conditions.

Through the above procedure, the element antennas provided at each lattice point 3 of the array antenna of the circular aperture 2 shown in FIG. A density tapered array antenna can be constructed.

Next, another configuration of the density taper array according to the present invention will be described. In the configuration of the invention described above, it is assumed that the excitation amplitude level of each element antenna is equal. This uniform amplitude condition has always been used in conventional density taper arrays. but,
In this way, in a density taper array consisting of element antennas with equal amplitude levels, the directivity gain of the array antenna decreases due to non-excitation (or thinning out) of the element antennas. Therefore, in order to improve this, it is effective to provide multiple excitation amplitude levels for the element antennas, but until now, density taper arrays with multiple amplitude levels have not been considered, and therefore the design No examples or configurations are disclosed. Here, we will show a configuration of a density taper array antenna with multiple amplitude levels, which is a further extension of the new density taper array antenna configuration in the case of uniform amplitude levels described above.

Now, suppose that there are L types of amplitude levels of element antennas, and let each amplitude level be g ₁ , g ₂ , g ₃ , ..., g _L , and g ₁ < g ₂ < g ₃ <... < g _L , Let g _L = 1 (4). In Fig. 2, when the weight for the i-th grid point (element antenna position) p _i is f _i , in Fig. 6, p _i to be filled with an element antenna having a small amplitude level g ₁ is determined by f _i g is in a range smaller than ₁ . Also, in FIG. 6, the amplitude levels g ₁ and g ₂
The range to be filled with element antennas is the range where f _i is g ₁ and g ₂
The range is between. Therefore, consider the following weights for each range.

However, g ₀ =0, and the maximum value of f _i is 1. A thinning function T _li having a value of 1 or 0 is determined by applying the above-described deterministic construction method for determining excitation and non-excitation of the element antenna to this weight f _li . Then, if T _i = _L 〓 ^l=1 (6), the amplitude level of the element antenna at the i-th grid point p _i becomes g _Ti . However, when T _i =0, the element antenna is not excited or is thinned out.

In this way, a density taper array in which the element antenna has multiple amplitude levels can be constructed on the circular aperture plane shown in Fig. 1, thereby achieving low side lobes and low directional gain. A high density tapered array antenna is realized.

The characteristics of the density taper array of the embodiment constructed based on the deterministic element thinning method according to the present invention described above will be explained with reference to the drawings.

Figure 7 shows the calculated value of the radiation pattern of a density taper array with equal amplitude levels.
This is the characteristic of the point wave source circular aperture array antenna shown in the figure. The solid line is the one according to the present invention, and the broken line is the characteristic of the density taper array antenna constructed by the conventional statistical thinning method. The given array density function is a Taylor distribution of _−30dB , but the characteristics of the density taper array antenna according to the present invention are sufficiently low not only in the region close to the main beam but also over an angle sufficiently far away from the main beam. It exhibits sidelobe characteristics, and it is clear that its characteristics are much superior to conventional ones. The horizontal axis in Fig. 7 is the sine of the angle θ from the normal direction (Z-axis direction) perpendicular to the circular aperture surface (paper surface) in Fig. 2, and Fig. 7 is the sine of the angle θ in the X plane. It is a radiation pattern.

Furthermore, FIG. 8 shows a list of radiation characteristics of a density taper array according to an embodiment of the present invention in which the element antennas have a plurality of amplitude levels. However, the calculated value is based on a circular aperture array with a diameter of 20λ, a density function of −35dB _Taylor distribution, a directivity of the element antenna of cos ^1.3θ , and an array grid shape of a regular triangular grid with a side of 0.6λ. be. It is clear from FIG. 8 that the directivity gain and sidelobe level are improved by providing a plurality of amplitude levels.

Next, other embodiments of the invention will be described.
In FIG. 9, eight hatched areas among the circular apertures are circular apertures that are missing due to reasons such as the antenna support structure or the apertures being shared with antennas of other frequency bands. FIG. 10 is an enlarged view of one missing portion 4b in FIG.
Among the grid points 1, 32, . . . , 39, the grid point 35 is removed by the missing portion 4b, and no element antenna can be placed at this grid point. Here, in this invention, when some grid points are removed in this way, the optimal weight is given as follows. That is, consider the occupied area (hereinafter referred to as element occupied area) Δs _i of the element antenna at each grid point. In FIG. 10, if the lattice point 35 was not removed, all the lattice points would have an area equal to the area of the square cells, but if the lattice point 35 is removed, Divide the element occupied area of lattice point 35 into four equal parts, and divide the area occupied by the lattice point 35 into four lattice points 3 that are closest to the lattice point 35.
2, 34, 36, and 38.
Therefore, the element occupation area of each grid point 32, 34, 36, and 38 increases by 1.25 times. In this way, the element occupied area Δs is recalculated for the remaining lattice points, taking into consideration the lattice points removed by the missing portions 4a and 4b. At this time, consider the weight at each grid point p _i as follows.

f _i =c・f(p _i )・△s _i (7) However, f is the density function given at the beginning, and c is f _i
This is a constant for normalizing the maximum value of . In this way, if we consider the element occupied area △s _i and consider the weight f _i , the weight given to the lattice points near the part where the lattice points have been removed increases, and the element antenna excited on those lattice points This increases the proportion of holes placed, and can compensate for the lack of excitation distribution due to partial apertures, and hence the deterioration of characteristics.

Next, still another embodiment of the present invention will be described. If a part of the antenna aperture is missing as mentioned above, or if the aperture is elliptical or rectangular, the weighting function that should be given to obtain low sidelobe characteristics, for example, is the Taylor distribution for a circular aperture. I can't find anything as convenient as this. Therefore, in addition to adding the element occupied area to part of the weight as described above, the optimal density function can be obtained by the following method.

Figure 11a shows a radiation pattern in which the main beam is directed toward θ = 0°, and point wave sources are placed at all remaining lattice points of the circular array antenna where parts of Figure 9 have been removed. It is obtained by excitation with equal amplitude and phase. Next, consider suppressing the side lobe in the θ=θ ₁ direction to the level of Δ in the radiation pattern of FIG. 11a. At this time, consider a radiation pattern directed in the _θ1 direction as shown in FIG. This pattern maximizes the gain for the array antenna shown in Fig. 9, that is, in the case of the planar array antenna shown in Fig. 9, which uses the same element antenna, it provides equal amplitude and an appropriate linear phase distribution. This can be easily accomplished. By overlapping these two patterns a and b with their relative amplitude ratio and phase appropriately selected, the desired sidelobe level of θ ₁ can be obtained as shown in Figure 11c. It can be made △. The complex amplitude of the point wave source at the i-th element position (excitation amplitude and excitation phase together expressed as a complex number) which gives the radiation pattern shown in figure a is A _0i , and the excitation complex amplitude which also gives the pattern shown in figure b. If A _1i , then c
The new complex amplitude giving the pattern is given by A _i = A _0i + αA _1i (8). Here, α is a complex coefficient that is easily determined from the condition that the sidelobe level of θ ₁ in FIG. 11c is exactly Δ. The amplitude of this new excitation distribution {A _i } can be used as an optimal density function to make the side lobe in the θ ₁ direction in the partially missing circular aperture array antenna of FIG. 9 a desired Δ.

In addition, the direction in which the sidelobe is suppressed can be selected not only in θ ₁ but also in many directions such as θ ₂ , θ ₃ , etc., and a new excitation amplitude can be obtained in the same way as above, so this can also be expressed as a density function. can be used.
Regarding the optimal excitation phase distribution when an asymmetric radiation pattern is desired, for the density taper array obtained by using the above density function and determining excitation and non-excitation in the manner described above, the equation ( An optimal phase distribution can be obtained by repeating the operation like 8).

Although the above description has been made regarding a circular aperture array, the present invention is not limited to this, and can be applied to an array of apertures having arbitrary notch shapes in addition to rectangular or elliptical shapes to obtain similar effects. Further, the shape of the array lattice is not limited to a square, but may be a rectangle, a triangle, or the like. Furthermore, any type of element antenna may be used.
Furthermore, as mentioned in the main text, element antennas that are not excited may be thinned out depending on the case. Further, the arrangement positions of the element antennas are not limited to a regulated arrangement such as grid points, but may be irregular arrangement points.

〔Effect of the invention〕

As described above, in the array antenna thinning method according to the present invention, the cumulative value of the weight determined from the given array density function is converted into an integer value and evaluated.
By uniquely determining a thinning function with a value of It is possible to realize a density taper array antenna with optimized radiation characteristics, or a density taper array antenna with optimized radiation characteristics when the element antenna has multiple amplitude levels. When used in required active type phased array antennas, the effect is extremely large.

[Brief explanation of the drawing]

Fig. 1 is an explanatory diagram of a density taper array antenna obtained by a conventional deterministic method, Fig. 2 is a schematic configuration diagram of a circular aperture array antenna used in an embodiment of the present invention, and Fig. 3 is a diagram of a given density taper array antenna. A diagram showing the relationship between the weight accumulation determined from the density distribution function, an integer value, and a thinning function. Figure 4 is an explanatory diagram of the distribution of the number of elements on two orthogonal X and Y axes. Figure 5 shows the excited elements. A diagram illustrating an example of how to exchange an antenna and a parasitic element antenna, FIG. 6 is an explanatory diagram when the element antenna has a plurality of amplitude levels, and FIG. 7 is a diagram illustrating a density taper circular array antenna according to an embodiment of the present invention. FIG. 8 is a diagram showing a calculation result of a radiation pattern, and FIG. 8 is a diagram showing a list of characteristics of a density taper circular array antenna according to an embodiment of the present invention when there are multiple amplitude levels.
Fig. 9 is a schematic configuration diagram of a circular aperture with a part of the aperture missing, Fig. 10 is an enlarged view of a part of the missing aperture, and Fig. 11 is a diagram of a circular aperture for suppressing a side lobe in the direction of θ ₁ . FIG. 3 is an explanatory diagram for obtaining a new amplitude distribution. In the figure, 1 is an element antenna, 2 is a circular aperture, 3 is a lattice point, 4a and 4b are missing parts, and 31 to 39 are lattice points. In the drawings, the same or corresponding parts are denoted by the same reference numerals.

Claims (1)

[Claims] 1. An array antenna consisting of a plurality of element antennas provided at grid points of a given array grid,
In an array antenna thinning method that aims to reduce side lobes by exciting or not exciting the element antenna according to a thinning function, an array density function defined on the element antenna array plane is given, and each lattice is The points are ordered linearly or circularly and continuously, and the weight obtained by multiplying the array density function at the j-th grid point by a constant is f _j (however, the maximum value of f _j is 1) Then, the value of the thinning function at the i-th grid point is T _i = [( _i 〓 ^j=1 f _j ) + 0.5] − [( _i-1 〓 ^j=1 f _j ) + 0.5] =0 or 1 However, when [ ] is expressed as a Gaussian symbol, when T _i =1, the element antenna at the i-th lattice point is excited, and when T _i =0, the i
An array antenna thinning method characterized in that the element antenna at the th grid point is not excited. 2. When the array grid points are distributed on a plane and the two mutually orthogonal axes on this plane are the X axis and the Y axis, the order of attaching the grid points is as follows: On the other hand, a patent characterized in that the lexicographic order of (X, Y), that is, the order of X from small to large, and the order of Y from small to large for grid points where X is the same An array antenna thinning method according to claim 1. 3 When the array lattice points are distributed on a plane and the two mutually orthogonal axes on this plane are the X axis and the Y axis, the order of attaching the lattice points is as follows: On the other hand, a patent characterized in that the lexicographical order of (X, Y), that is, the order of Y from small to large, and for grid points where Y is the same, the order of X from small to large. An array antenna thinning method according to claim 1. 4 When the array lattice points are distributed on a plane and the two mutually orthogonal axes on this plane are the X and Y axes, consider the thinning function projected on the X and Y axes and find out that the function value does not change. 2. The method for thinning out an array antenna according to claim 1, wherein the excitation element and the parasitic element are exchanged in the following manner. 5. The array antenna thinning method according to any one of claims 1 to 4, characterized in that element antennas that are not excited are removed. 6 An array antenna consisting of a plurality of element antennas provided at the grid points of a given array grid,
In an array antenna thinning method that aims to reduce side lobes by exciting or not exciting the element antenna according to the thinning function, an array density function defined on the element antenna array plane is given, and each element antenna is The grid points are ordered linearly or circularly and continuously, and the weight determined from the arrangement density given to the i-th grid point is f _j ,
The types of excitation amplitudes of the element antenna to be excited are L types, and each amplitude is g ₁ , g ₂ ,..., g _L (however, g ₁ < g ₂ <
...<g _L , g _L =1), give a new weight at the i-th grid point for each l by the following formula, However, g ₀ = 0 Using this weight f _li , T _li = [( _i 〓 ^j=1 f _li ) + 0.5] − [( _i-1 〓 ^j=1 f _lj ) + 0.5] = 0 or 1 However, [ ] calculates the thinning function value obtained by the Gauss symbol formula, and when the added value of T _li for l is Ti= _L 〓 ^l=1 T _li , the element at the i-th lattice point An array antenna thinning method characterized in that the excitation amplitude of the antenna is given by g _Ti , and no excitation is performed when T _i =0. 7. The array antenna thinning method according to claim 6, characterized in that element antennas that are not excited are removed. 8 An array antenna consisting of multiple element antennas provided at the grid points of a given array grid,
In an array antenna thinning method that aims to reduce side lobes by exciting or not exciting the element antenna according to a thinning function, an array density function defined on the element antenna array surface is given, and the array lattice is Some of the points are missing, and each remaining grid point is sequentially or circularly ordered, and the position of the i-th grid point is determined as P _i , i
Assuming that the exclusive area of the element antenna at the th grid point is △S _i and the given array density function is f, f _i =C・f(P _i )・△S _i , where C is the maximum value of f _i 1. An array antenna thinning method characterized in that a thinning function value is determined by using f _i given by a constant expression normalized to as a weight of an i-th grid point. 9. The array antenna thinning method according to claim 8, characterized in that element antennas that are not excited are removed. 10 In an array antenna consisting of a plurality of element antennas provided at the lattice points of a given array lattice, low sidelobes can be achieved by exciting or not exciting the element antennas according to a thinning function. In the array antenna thinning method, consider a radiation pattern b having the main beam with the maximum gain in the direction of a certain sidelobe level of the radiation pattern a of the array antenna, find the excitation amplitude and phase corresponding to the radiation patterns a and b, and An array antenna thinning method characterized in that an array antenna excitation distribution resulting from appropriately selecting and overlapping two relative amplitude ratios and phases is used as an array density function defined on an element antenna array surface. 11. The array antenna thinning method according to claim 10, characterized in that element antennas that are not excited are removed.