JPS6353726B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a slot array antenna device having a feeding section near the center of the array.

Generally, in order to put a center-fed non-resonant spaced array antenna into practical use, it is necessary to equalize the power fed to the left and right arrays, make the equiphase planes of both arrays continuous at the feeding point, and further reduce the input reflection coefficient. requires special consideration. However, in the past, the behavior of electromagnetic waves at the feeding point was not sufficiently studied, and the distance from the feeding point to the left and right arrays, which was an important design factor, was overlooked. For this reason, it has not been possible to realize the above-mentioned antenna that can be put to practical use.

The present invention makes it possible to design an antenna device with target antenna characteristics by a reliable method that does not involve trial and error in a non-resonant spaced slot array antenna having a feeding section near the center of the array, and is thereby put into practical use. The purpose of this invention is to provide a slot array antenna with antenna performance that can provide the desired antenna performance.

Therefore, in the present invention, by strictly solving the behavior of reflected waves and traveling waves at the feeding section in the slot array antenna, a limit is set on the distance from the feeding section to the left and right arrays, By satisfying the feed power ratio and the equal phase plane shape, a slot array antenna device with almost target antenna characteristics is realized.

The antenna device according to the present invention will be referred to as a centrally fed non-resonant spaced slot array antenna (hereinafter abbreviated as NCFA) because the array element spacing is not the so-called resonance spacing.

That is, in the antenna device of the present invention, a plurality of slot elements are bored at non-resonant intervals in the wall surface of a waveguide, and the plurality of slot elements are divided into approximately 1/2 of the physical length of the waveguide. In a non-resonant slot array antenna device in which a feed point is arranged, the distance D between the adjacent slot elements sandwiching the feed point is 0.45λ g to 0.45λ _g for a channel wavelength λ _g .
Beam squint angle α = sin ^-1 {(λ _{0 /} λ _g )・δ/(1+δ)} (where: δ is determined as a parameter representing non-resonance in the above equation so that λ ₀ is the free space wavelength) is a value near the design beam width angle of the antenna, and one slot element adjacent to the other across the feed point is The distance D from the feeding point is D={(D/2)・[(1+2δ)/(1+δ)]
}, and the distance D between the other slot element adjacent to the feed point with the feed point in between is D={(D/2)·[1/(1+δ)]}, and the distance between the feed point and the slot element is The spacing between the plurality of slot elements on the side where the spacing between the feed point and the slot elements is D is set as (λ _g /2)・(1+δ), and the spacing between the plurality of slot elements on the side where the spacing between the feed point and the slot elements is D is set as (λ g /2)・(1+δ). The spacing between the slot elements is set as (λ _g /2)·[(1+δ)/(1+2δ)].

Figure 1 shows the structure of the NCFA near the power supply section. In FIG. 1, , are left and right arrays, each with a constant interval S,
It has a number of slots drilled in the shape S, and the left and right arrays are fabricated on a single waveguide 1. F is a feeding point, and the distances from this feeding point F to the first slot elements A ₁ and B ₁ of each array are D and D, and the distance between the slot elements A ₁ to B ₁ is D. A transmission line for power supply is connected to the power supply point F. P and P are equal phase planes of the left and right arrays, and the angles formed between these equal phase planes and the arrays are α and α.

The array antenna is configured such that the equal phase plane is flat over the entire array plane. In order to satisfy this condition in NCFA, there are restrictions on the above-mentioned intervals S, S and distances D, D. According to the traveling wave approximation, the angles α and α must be equal, so if one of the intervals S and S is set larger than 1/2 of the internal wavelength of the waveguide 1, the other must be set smaller. It won't happen. That is, for a non-resonant antenna, the non-resonance parameter δ is δ
≠0, and if one interval S is S = (λ _g /2) (1 + δ) > λ _g /2, then in order to satisfy the above condition, the other interval S is A simple calculation shows that S=(λ _g /2) [(1+δ)/(1+2δ)].

The above-mentioned intervals S and S are determined once δ is determined.

Since δ≠0 in a non-resonant antenna, the main beam direction of the antenna deviates from the mechanical center axis of the array, as is well known. Technical Report No. AD63600 by Hughes Aircraft Co., distributed by the National Diet Library and accepted by the National Diet Library on March 11, 1960.
348 “Waveguide Slot Array Design (WAVEGUIDE
SLOT ARRAY DESIGN).
According to this, the size of the beam quint is the angle α between the main beam direction and the mechanical center axis of the array.
If λ ₀ is the free space wavelength and λ _g is the waveguide wavelength, then the beam squint angle α is α=sin ^-1 {(λ ₀ /λ _g )−(λ ₀ /2S)}. be. Here, in the case of a resonant antenna, S is equal to 1/2 of λ _g , and in the case of a non-resonant antenna, the above-mentioned parameter δ indicates how much this is shifted, so in the above formula, the above-mentioned S = (λ _g /2) By substituting (1+δ) and expressing the beam squint angle α using the non-resonance parameter δ, the following equation is obtained.

α=sin ^-1 {(λ ₀ /λ _g )・δ/(1+δ)} As stated in the above document, in this type of antenna, the beam squint angle α should be approximately the design beam width angle. It is generally designed to If this α becomes too large, the equivalent aperture length of the array as viewed from the main beam direction will become short and the antenna gain will decrease. However, in general, if α is kept below, for example, 5 to 6 degrees, the gain will be sufficient for practical purposes. can be designed.

δ is basically a parameter selected in the design based on specific specifications such as the beam width of the antenna, but if you choose a value too close to zero for δ and get close to the resonant system, the reflection coefficients of the left and right arrays will change. Γ、
As Γ increases, the phase and amplitude errors at the central branch tend to increase. reflection coefficient Γ, Γ
Since the size of is also related to the number of slot elements in the array, δ is automatically determined by the design specifications mentioned above once the length of the array is determined, but if δ≠0
The purpose of this is to prevent the reflection coefficient of the array from becoming too large, and generally, as mentioned above, δ gives a beam quint about the width of the beam.
Practical reflection characteristics can be obtained by selecting the values. For example, in a commonly used rectangular waveguide, δ ₀ /λ _g = approximately 0.7, so if α = 5°, δ
is about 0.142, and when α=3°, δ is about 0.08, and usually δ is a positive value much smaller than 1.

On the other hand, regarding the above distances D and D, the feeding point F
There is a limit so that there is no step difference between the equiphase planes of the left and right arrays. This limit can also be easily calculated using the following equation.

D/D=1+2δ...(1) However, in this equation (1), even if δ is determined, the distance D,
D still remains optional. That is,
As mentioned above, δ is determined from the requirements of the antenna beam pattern, but even if δ is determined, there is only one condition as the ratio shown in equation (1) for the two unknowns of distances D and D. This is because it is not given to you.

Conventionally, distances D and D have been determined almost arbitrarily under the restrictions of equation (1), or a certain arbitrary distance D has been set without considering equation (1), and The method used was to determine the most desirable position based on the results of actual measurements at different positions. In either method, in the past, much of the process had to be done by trial and error, and although it required a great deal of effort, it was rare to obtain satisfactory results.

Conventionally, NCFA could not be put into practical use because
This is because even if the value of the parameter δ is determined as described above, the aforementioned arbitrariness exists in determining the distances D and D.

In the present invention, new restrictions are derived by examining the behavior of electromagnetic waves at the power feeding section in detail, distances D and D are determined based on these new restrictions, and a practical NCFA is provided.

In the present invention, when the parameter δ is determined according to the required beam pattern and frequency characteristics of the antenna as described above, the distances D and D are uniquely determined, and both the phase and amplitude conditions are simultaneously satisfied, and the input The reflection coefficient can also be reduced. The principle of the present invention will be explained in detail below.

Figure 2 shows an equivalent circuit near the power supply section of the NCFA. In FIG. 2, V and V are the voltages applied to the first slot elements A ₁ and B ₁ of each array,
Γ, Γ are the first slot elements of the array,
Input reflection coefficients, ai and bi at positions A ₁ and B ₁
(where i=,,) is Ti of each port (where i
= 1, 2, 3), the amplitude of the wave at the reference plane, Θ,
Θ represents the electrical angle between T ₁ ′ ~ T ₁ and T ₂ ′ ~ _{T 2} ,
Θ=D・2π/λ _g and Θ=D・2π/λ _g , respectively. Now, assume that the E-plane T branch is adopted as the power feeding section. At this time, if T ₁ ′ and T ₂ ′ are taken as the plane of symmetry of the T branch, and T ₃ ′ is taken arbitrarily, the S matrix can be written as the following equation (2), taking into account the symmetry of the circuit. Can be done.

All power applied to the port is
Therefore, S ₃₃ =0 is a necessary condition when used as the power feeding section of the NCFA. This condition can be achieved by mounting matching elements. If this condition is satisfied and the circuit is lossless, then due to the unitarity of the S matrix, |S ₁₁ |= |S ₁₂ |=1/2 ...(3a) |S ₁₃ |=1/√2 ... (3b) becomes. Since T ₃ ' is arbitrary, there is no problem in moving to position T ₃ where argS ₁₃ =0. thus,
The S matrix between T ₁ ′, T ₂ ′, and T ₃ is becomes. Further, by moving T ₁ ′ and T ₂ ′ to the positions T ₁ and T ₂ immediately before the first slot elements A ₁ and B ₁ of the array, an S matrix between T ₁ , T ₂ , and T ₃ is obtained, It becomes as shown in the following equation (5).

(However, φ is the transmission delay between the ports.) In Figure 2, the relationship between the output wave amplitudes from T ₁ , T ₂ , and T ₃ and the incident wave amplitudes to T ₁ , T ₂ , and T ₃ is ,
From the definition of scattering matrix, It is expressed as Also, a and b, and a and b are related by the reflection coefficients Γ and Γ of the load of each port, a=b・Γ, a=b・Γ
...(6b) It is expressed as.

Substituting equation (6b) into equation (6a), we get Substituting equation (5) into equation (6c), b, b on the left side
Performing the product on the components yields the following:
However, in this case, since a is the incident wave from the signal source, it is normalized as usual and set to a=1.

Here, if we create (6d) × e ^j 〓〓〓 + (6e) × e ^j 〓〓, the second term on the right side of each equation disappears, and we get b[e ^j 〓〓〓−Γ/2・e ^j( 〓 ^- 〓 〓 ⁾ 〕＋b
[e ^j 〓〓〓−Γ/2・e ^j( 〓 ^- 〓〓 ⁾ ] ＝b〔Γ／2・e ^j( 〓 ^- 〓〓 ⁾ 〕＋b
[Γ／2・e ^j( 〓 ^- 〓〓 ⁾ ] Therefore, b [e ^j 〓〓−Γ・e ^j( 〓 ^- 〓〓 ⁾ ]=−b
Obtain [e ^j 〓〓−Γ・e ^j( 〓 ^- 〓〓 ⁾ ]. If we create b/b from this, b/b=-e ^j 〓〓-Γ・e ^j( 〓 ^- 〓〓 ⁾ /e
^j 〓〓−Γ・e ^j( 〓 ^- 〓〓 ⁾ =−e ^j 〓〓{1−Γ・e ^j(
〓 ^-2 〓〓 ⁾ }/e ^j 〓〓{1−Γ・e ^j( 〓 ^-2 〓〓 ⁾ } By adjusting this, the following equation (7) can be obtained.

b/b=-1-Γ・e ^j( 〓 ^-2 〓〓 ⁾ /1-Γ・
e ^j( 〓 ^-2 〓〓 ⁾・e ^j( 〓〓 ^- 〓〓 ⁾ ……(7) To make the power supplied to the left and right arrays equal, the amplitude, that is, the power supplied, must be |b/b|=1. There needs to be. There are many Θ and Θ for which this condition holds, but here, Γ・e ^j( 〓 ^-2 〓〓 ⁾ = {Γ・e ^j( 〓 ^-2 〓〓 ⁾ } ^* ……(
8) Consider the case. In formula (8), * indicates complex conjugation. Since it is known that Γ and Γ are close to complex conjugate to each other in an actual array, the assumption in equation (8) can be approximately satisfied. In equation (8), Γ=Γ ^*
Then, if we rearrange the angles, Θ+Θ=φ+nπ, (n=0, 1, 2,...)
...(9) is obtained. φ represents the transmission delay between the ports and is generally small, so it can be ignored. In this way, we finally obtain Θ+Θ=nπ, (n=0, 1, 2,...). Expressing this using the tube wavelength, D≡D+D=n·λ _g /2 (10). From these equations (10) and (1), it is usually better that D is smaller, so if n=1, D=λ _g /4 (1+2δ/1+δ), D=λ _g /4
(1/1+δ) ...(11).

Now, let us consider the excitation phase of the first slot element under the condition of equation (11).

Since it can be written as V/V=b(1+Γ)/b(1+Γ), using equation (7), argV/V=argb/b+arg1+Γ/1+Γ
=(Θ−Θ)+arg1−Γ・e ^j( 〓 ^-2 〓〓 ⁾ /1
−Γ・e ^j( 〓 ^-2 〓〓 ⁾ +arg [(1+Γ)/(1+Γ)] ...(12) In equation (12), the first term is the design value, and the second term is the power supply part
b due to the reflected waves of the array via T ₁ ′, T ₂ ′
, b, and the third term represents the phase difference based on the traveling wave of the first slot element excitation voltage caused by the reflected wave.

Now, consider φ−2Θ and φ−2Θ in the second term of equation (12). As mentioned above, it is usually possible to set φ=0. Θ and Θ are not equal because δ≠0, but in the present invention, as described above, δ is 1.
Since we are targeting cases much smaller than
From equation (11), D≒D≒λ _g , and 2Θ≒2Θ≒π ...(13). In this way, by substituting equation (13) into equation (12), argV/V≒(Θ−Θ) +arg[(1+Γ)/(1+Γ)] +arg[(1+Γ)/(1+Γ)], and the second term and The third term cancels each other out. Therefore, arg(V/V)≈Θ−Θ, and a phase difference close to the designed value can be achieved.

Next, when calculating the reflection coefficient at the port, we get
I get the following result:

|Γ| ^{= ||} X| was used. From this equation, it can be seen that |Γ| becomes smaller than Γ when Re[X] is positive. It is not difficult to design an array in this way, especially when |X| = cos(argX) in the above equation, the reflection is 0.
becomes.

As mentioned above, the non-resonant parameter δ
As usual, the left and right arrays are set to be much smaller than 1, an element is provided for matching the transmission line connected to the feed point, and the distances D and D are given by equation (11). This means that all the conditions necessary to realize NCFA, such as equalizing the power fed to the two and making the equiphase planes of both continuous at the feeding point, can be met. In the calculation example described above, the E-plane T branch was explained as the power feeding section, but the same conclusion can be reached even when an H-plane T branch, probe coupling, magic T, etc. are used.

Also, in the calculation example above, φ=0 and Γ=Γ ^*
However, if it becomes necessary to correct these two approximations, a correction term is added to the right side of equation (10). This correction term, approximately within ±5% of λ _g , is sufficient in almost all cases. In this way, by setting D≡D+D=λ _g /2 (1.1) to λ _g /2 (0.9) and D/D=1+2δ, the amplitude, phase,
It is possible to create a state in which the input reflection coefficient is satisfied.

[Brief explanation of drawings]

The figure shows one embodiment of the present invention.
The figure is a block diagram showing the vicinity of the center feed section of the center feed non-resonant spacing slot array, and FIG. 2 is an equivalent circuit diagram of FIG. 1. 1: waveguide, ,: array, A ₁ , A ₂ : first
Slot element, F: feeding point.

Claims (1)

[Claims] 1. A power feeding system in which a plurality of slot elements are bored at non-resonant intervals in the wall surface of a waveguide, and the plurality of slot elements are divided into approximately 1/2 of the physical length of the waveguide. In a non-resonant slot array antenna device in which points are arranged, the distance _{D between the slot elements adjacent to each other across the feeding point is 0.45λ g to}
Beam squint angle α α = sin ^-1 {(λ _{0 /} λ _g )・δ/(1+δ)} ( (where λ ₀ is the free space wavelength) is determined as a parameter representing non-resonance so that λ 0 is a value near the design beam width angle of the antenna, and one slot element adjacent to the feeding point and the feeding point are The distance D between the points is D={(D/2)・[(1+2δ)/(1+δ)]
}, and the distance D between the other slot element adjacent to the feeding point with the feeding point in between is D={(D/2)·[1/(1+δ)]}, and the distance D between the feeding point and the feeding point is as follows. The spacing between the plurality of slot elements on the side where the spacing from the slot element is set to D is set to (λ _g /2)·(1+δ), and the side where the spacing between the feed point and the slot element is set to D. A slot array antenna device characterized in that an interval between the plurality of slot elements is (λ _g /2)·[(1+δ)/(1+2δ)].