JPH0457980B2 - - Google Patents

Description

Detailed Description of the Invention (Technical Field) The present invention receives a signal using an array composed of a plurality of elements, forms a beam using phase compensation means, and calculates the direction cosine of the signal source from its maximum point. The present invention relates to a method for correcting an error in estimating the cosine of a direction, which is caused by a phase response error of each element in an array, in an angle measuring device that determines the angle in the direction of a sound source by estimation.

(Background Art) In sonar, acoustic positioning, and radar, a narrowband beamformer using an array consisting of a plurality of spatially arranged elements and a phase compensation means is widely used. An angle measuring device is also widely used which calculates the angle of the signal source direction by estimating the direction cosine of the signal source direction with respect to the reference axis on the array by adding a maximum point detection means for determining the angle.

In such an angle measuring device, the direction cosine is estimated based on the phase information of the output signal of each element of the array. On the other hand, if there is a phase response error in the phase response of each element of the array, the direction cosine is estimated based on the phase information of the output signal of each element of the array. Therefore, the phase response error causes an error in the estimation of the direction cosine.
Furthermore, since an amplifier, bandpass filter, etc. is usually added to each array element, if there is a phase response error in the phase response of the amplifier, bandpass filter, etc., the phase response error is also the direction cosine. If the estimation error of the direction cosine caused by these phase response errors cannot be ignored, a means for correcting the estimation error is required.

Conventionally, as a method for such correction, a method has been used in which a phase error corrector for removing the phase response error is added to each element.

An example of a conventional correction method for direction cosine estimation error is shown in FIG. In Figure 1, 1 _1, 1 ₂ ..., 1 _M
are array elements, 2 ₁ , 2 ₂ ..., 2 _M are amplifiers added to the elements, 3 ₁ , 3 ₂ , ..., 3
_M is a bandpass filter added to each of the elements, 4
₁ , 4 ₂ ..., 4 _M are each phase error correctors, 5 is a narrow band beamformer, 6 is a maximum point detector, 15 is a converter that converts the direction cosine into an angle, and 7 is an output of the estimated angle value. It is an output terminal. Note that the influence of the phase response error on the directional cosine estimation error is determined by the relative phase response error of other elements with respect to the phase response of one arbitrarily selected element, so the phase error correction shown in Figure 1. It is sufficient that the number of containers is M-1. M elements 1 ₁ , 1 ₂ ,... spatially arranged
..., 1 Signal input to _M x ₁ (t), X ₂ (t), ..., x _M
(t) is amplified to an appropriate level by amplifiers 2 ₁ , 2 ₂ , . . . 2 _M , and bandpass filters 3 ₁ , 3 ₂ , .
_3M , unnecessary frequency components are removed and converted into narrowband signals y ₁ (t), y ₂ (t), ..., y _M (t), and the narrowband signals are passed through the phase error corrector 4 ₁ , 4 ₂ , ... 4 _M , the phase displacement Δφ ₁ , caused by the phase response error of the element, amplifier, and bandpass filter.
Δφ ₂ , ..., Δφ _M are removed and the signals Z ₁ (t), Z ₂
(t), ..., Z _M (t).

The signals Z ₁ (t), Z ₂ (t ₎ , . The direction cosine estimate α^ is determined from the spatial frequency at which the beam output signal has a maximum value, and the transducer 15
The direction cosine estimate α^ is converted to an angle estimate.

FIG. 2 shows an explanation of the narrowband beamformer, and FIG. 3 shows an explanation of the direction cosine α when the array is a two-dimensional (planar) array. 20 ₁ , 20 ₂ in Figure 2
. . . , 20 _M are phase error correctors 4 ₁ , 4 ₂ , . . .
...4 Input terminals into which the output of _M is input, 21 ₁ , 21
₂ ,..., 21 _M are phase compensators for realizing phase compensation means, 22 is an adder, 23 is an envelope detector, 24 is an output terminal outputted to the maximum point detector 6, and 30 is a planar array, and θ _x and θ _y are the direction co-source angles in the signal source direction with respect to the X and Y axes of a rectangular coordinate system in which the origin O and the X and Y axes are placed on the plane array surface; The direction cosine is αΔ =
[cosθ _x , cosθ _y ] ^T. Figure 3 shows the case of a two-dimensional array, but in the case of a one-dimensional (linear) array, αΔ = cosθ _x , and in the case of a three-dimensional array, αΔ =
[cosθ _x , cosθ _y , cosθ _z ] If ^T is used, it can basically be treated in the same way as a planar array, so only the planar array will be explained below. Therefore, in Figure 1, α^Δ =
[cosθ^ _x , cosθ^ _y ] ^T. Note that here, the subscript T indicates the transposition of the vector.

In the conventional direction cosine estimation error correction method shown in _FIG .
4 ₂ , _. In addition, each array element 1 ₁ , 1 ₂ , . . . , 1 _M and an amplifier 2 ₁ , 2 ₂ , .
If the phase response errors Δφ ₁ , Δφ ₂ , _{. .} . , Δφ _M of the band filters 3 ₁ , 3 _{2 ,} . ₁ , ₄₂ ,... _4M , the phase correction values of all the correctors that require a change in correction value must be reset. Furthermore, when the received signal consists of a plurality of frequencies and the direction cosine is to be estimated for any of these frequencies, the array elements 1 ₁ , 1 ₂ , . . . , 1 _M , the amplifiers 2 ₁ , 2 ₂ ,...
_2M , the phase response characteristics of the band filters ₃₁ , ₃₂ , ..., _3M change depending on the frequency, and therefore the phase response error also changes depending on the frequency, so in this case, the phase error corrector 4 ₁ , 4 ₂ , . . . 4 _M must be switched in conjunction with the receiving frequency, which has the disadvantage that the device becomes more complicated and larger.

SUMMARY OF THE INVENTION The present invention aims to eliminate the drawbacks in the prior art as described above.

The angle estimation error correction method of the present invention consists of M pieces arranged in space (M is an integer of 2 or more).
an array including an element and an amplifier and a filter added to the element; a beamformer for the output signal of the element using phase compensation means; and maximum point detection means for determining the maximum point of the output of the beamformer in a spatial frequency domain. In a narrowband angle measuring device, the angle in the direction of the signal source is determined by estimating the direction cosine of the direction of the signal source on the reference coordinate axis of the array from the maximum point determined by the maximum point detection means, the input signal analytically estimate in advance an estimation error of the direction cosine caused by a phase response error of the phase response characteristic of each of the elements, store the estimated estimation error, and store the estimated error in the beamformer and the maximum point detection means. The estimated value of the direction cosine is corrected by subtracting the stored estimation error from the estimated value of the direction cosine.

(Description of Examples) Here, before entering into the description of the examples of the present invention, an outline of the present invention will be explained using equations.

See Figure 3. the geometric shape of the array,
That is, the array elements 20 ₁ , 20 ₂ , . . . shown in FIG.
..., 20 _i , ..., 20 _M at rectangular coordinates X, Y, respectively position vectors P ₁ , P ₂ , ..., P _i , ...,
P _M is given, and the phase response error Δφ ₁ ,
Δφ ₂ , ..., Δφ _i , Δφ _M are given, |Δφ _i ≪1;
If i = 1, ..., M, the document "Angle measurement error in SSBL acoustic positioning due to phase error; Institute of Electronics and Communication Engineers,
SANE82-15, July 1982",
The direction cosine estimation error caused by Δφ _i ; i=1,...,M ΔαΔ = [cosθ _x , cosθ _y ] ^T is Δαλo/2π( _M 〓 ⁱ⁼¹ | b _i | ² H) ^-1 _M ⁱ⁼¹ ｜b _i ｜ ² (P _i −P _c )Δφ _i (1). However, here, λ ₀ is the wavelength of the signal, and the matrix H and vector P _c are constants given as follows.

P _c Δ = _M 〓 ⁱ⁼¹ ｜b _i ｜ ² P _i ／ _M 〓 ⁱ⁼¹ ｜b _i ｜ ² (2) HΔ ＝ _M 〓 ⁱ⁼¹ ｜b _i ｜ ² P _i P _i ^T ／ _M 〓 ⁱ⁼¹ | b _i | ² P _c P _c (3) The present invention corrects the direction cosine estimation error Δα given by equations (1) to (3) for the direction cosine estimation error caused by the phase response error of the array element. It is used for

FIG. 4 shows a first embodiment of the present invention, in which 8 is a register, 9 is an adder, and 16 is an output terminal that is output to the converter 15. The above formula (1) is written in register 8.
The direction cosine estimation error Δα obtained in ~(3) is memorized, and the uncorrected direction cosine estimation value α~Δ output from the maximum point detector 6 is calculated from [cosθ~ _x , cosθ~ _y ] ^T , by subtracting the direction cosine estimation error Δα by the adder 9, the corrected direction cosine estimated value α^=
[cosθ^ _x , cosθ^ _y ] ^T is output to the output terminal 16. In the first embodiment of the present invention shown in FIG. 4, the phase error corrector 4 ₁ for correcting the conventional phase response error,
4 ₂ ,...4 _M is not required and the array constant is set in advance.
If the estimated direction cosine difference Δα is obtained from the above formula (1) from P _c , H, and phase response errors Δφ ₁ , Δφ ₂ , ..., Δφ _M , then
With only one register 8 and one adder 9, the direction cosine estimation error caused by the phase errors Δφ ₁ , Δφ ₂ , . . . , Δφ _M can be removed.

As explained above, in this embodiment, the estimation error of the direction cosine caused by the phase response errors Δφ ₁ , Δφ ₂ , ..., Δφ _M is obtained in advance by an analytical method, and the phase response error Direction cosine estimate obtained from narrowband beamformer and maximum point detector for signals y ₁ (t), y ₂ (t), ..., y _M (t) containing Δφ ₁ , Δφ ₂ , ... _, Δφ M Since the direction cosine estimation error caused by the phase errors Δφ ₁ , Δφ ₂ , ..., Δφ _M is removed by subtracting it from There is an advantage that the circuits 4 ₁ , 4 ₂ , . . . 4 _M are not required, and only one register and one adder are required. In addition, the phase response errors Δφ ₁ , Δφ ₂ , ...,
When the value of Δφ _M changes, it is sufficient to calculate the direction cosine estimation error Δα again using the above formula (1) and replace the stored contents of the register 8, and the array elements 1 ₁ , 1 ₂ , . . . , 1 _M , amplifier 2 ₁ , 2 ₂ , ...2
It _is also possible to easily cope with changes in characteristics of band filters 3 ₁ , 3 ₂ , . . . , 3 _M due to aging.

FIG. 5 shows a second embodiment of the present invention, in which 8 ₁ ,
8 ₂ , . . . , 8 _M are registers, 10 is a multiplexer, 11 is a frequency-capable local oscillator, and 12 ₁ , 12 ₂ , . . . , 12 _M are multipliers. Local oscillator 11 has N types of frequencies.
F ₁ , F ₂ , . . . F _N can be generated, and the local oscillator 11 and multipliers 12 ₁ , 12 ₂ , .
..., 12 _M is one narrowband signal from among N narrowband signals having N types of center frequencies f ₁ , f ₂ , ..., f _N included in the received signals x ₁ (t), ..., x _M (t). _The output terminals of the band filters _{3 1} , 3 _{2 , .} (t), ..., y _M (t) is output.

Furthermore, the values of the phase response errors Δφ ₁ , Δφ ₂ , ..., Δφ _M at the frequencies f ₁ , f ₂ , ..., f _N are stored in the registers 8 ₁ , 8 ₂ , ..., 8 _M using the above formula. Direction cosine estimation error Δα ₁ , Δα ₂ , ..., obtained in advance in (1)
Δα _N is stored, and the multiplexer 10 selects the frequency F _j selected by the local oscillator from among the direction cosine estimation errors Δα ₁ , Δα ₂ , ..., Δα _N , that is, the center frequency f _j of the selected input signal. The corresponding direction cosine estimation error Δα _j is selected and output to the adder 9, and the following is the same as the first embodiment of the present invention.

_The second embodiment of _the present invention shown in _{FIG .}
is selected to estimate the direction cosine of the signal source, and the phase response error Δφ ₁ , Δφ ₂ , ...,
This is an application example where Δφ _M is different for each of f ₁ , f ₂ , ..., f _N . The phase error corrector 4 ₁ , 4 ₂ , ...,
The conventional method using 4 _M requires M×N phase error correctors in such a case, whereas
In the present invention, N registers 8 ₁ , 8 ₂ , ..., 8 _N
, one multiplexer 10 and one adder 9 are required. Therefore, similar to the first embodiment, this embodiment also has the effect of significantly reducing the size and simplifying the device compared to the conventional system.

(Effects of the Invention) This embodiment corrects the estimation error of the direction cosine caused by the phase response error of the array element by estimating the direction cosine obtained in advance from the constant determined by the geometrical shape of the array and the phase response error. Since this is done using the error, there is no need for phase error correctors, which are roughly proportional to the number of array elements, and only one adder and a register proportional to the number of receiving frequencies are required, which greatly reduces the equipment size. It has the advantage of being miniaturized and simple, and can be used for angle measuring devices in sonar, acoustic positioning devices, and radar using narrowband beamformers.

[Brief explanation of drawings]

Fig. 1 is an explanatory diagram of a conventional direction cosine difference correction method using a narrowband beamformer, Fig. 2 is an explanatory diagram of a narrowband beamformer, and Fig. 3 is an explanatory diagram of a two-dimensional (planar) array and a direction cosine. FIG. 4 is an explanatory diagram of the first embodiment of the present invention, and FIG. 5 is an explanatory diagram of the second embodiment of the present invention.
It is an explanatory view of an example of. 1 ₁ , 1 ₂ , ..., 1 _M ; array element, 2 ₁ , 2 ₂ ,
...2 _M ; Amplifier, 3 ₁ , 3 ₂ , ..., 3 _M ; Bandpass filter, 4 ₁ , 4 ₂ , ... 4 _M ; Phase error corrector,
5; Narrowband beamformer; 6; Maximum point detector;
8; register, 9; adder, 8 ₁ , 8 ₂ , ..., 8
_N ; register, 10; multiplexer, 11; variable frequency local oscillator, 12 ₁ , 12 ₂ ,...
..., 12 _M ; multiplier, 15; converter.

Claims (1)

[Claims] 1. An array including M elements (M is an integer of 2 or more) arranged in space and an amplifier and a filter added to the elements, and an output signal of the element using a phase compensation means. and maximum point detection means for determining the maximum point of the output of the beamformer in a spatial frequency domain, and a direction cosine of the signal source direction on the reference coordinate axis of the array from the maximum point determined by the maximum point detection means. In the narrowband angle measuring device that calculates the angle in the direction of the signal source by estimating the direction cosine estimation error caused by the phase response error (Δφ1,...ΔφM) of the phase response characteristic of each element with respect to the input signal. (Δα) is analytically estimated in advance, the estimated estimation error (Δα) is stored, and the estimation is stored from the estimated value of the direction cosine estimated by the beam former and the maximum point detection means. A method for correcting an angle estimation error, characterized in that the estimated value (α) of the direction cosine is corrected by subtracting the error (Δα). 2 The estimation error (Δα) is estimated by the formula Δαλo/2π( _M 〓 ⁱ⁼¹ | b _i | ² H) ^-1 _M 〓 ⁱ⁼¹ | b _i | ² (P _i −P _c )Δφ _i , In the above equation, b _i is a coefficient indicating the i-th receiving sensitivity, λo is the signal wave, Δφ _i is the phase response error of the phase response characteristic of the i-th element with respect to the input signal, and P _i is the i-th The position vector of the th element, P _c is the barycenter position vector of the array when the coefficient b _i is assumed to be a mass point, and is expressed by the formula P _c Δ = _M 〓 ⁱ⁼¹ | b _i | ² P _i / _M 〓 ^{i= 1} ｜b _i ｜ ² , H is the second moment about the center of gravity of the array, and the formula HΔ = _M 〓 ⁱ⁼¹ ｜b _i ｜ ² P _i P _i ^T / _M 〓 ⁱ⁼¹ ｜b _2. The angle estimation error correction method according to claim 1, wherein P _i ^T and ^P _{c T} represent the transposition of vectors _{P i and} P _c ^.