JPS59154379A - Measuring system of position error of array element - Google Patents

Abstract

PURPOSE:To measure the position error of an array element with a signal source in an unknown position by performing time or phase delay compensation, which corresponds to a forecasted position of the signal source, for the output of the array element to determine a true forecasted position and attaining the position error. CONSTITUTION:Outputs of array elements 111, 112... which are operated in accordance with the signal source in an unknown position are subjected to time or phase delay compensation corresponding to the forecasted position by delay compensators 211, 212.... The delay quantity where the envelope of the output sum of compensators 211, 212... is maximum is detected by the first signal source position estimator 24 to determine the first estimate position of the signal source. The true estimate position is determined in the second signal position estimator 25 by interpolation near the first estimate position. In accordance with these estimate positions, mutual delay errors of individual array elements due to delay detectors 272, 273..., constants corresponding to geometric shapes, amplitude responses, etc. of array elements due to registers 302, 303..., etc., position error detectors 342, 343... calculate the position error of array elements with the signal source in an unknown position and outputs the position error.

Description

DETAILED DESCRIPTION OF THE INVENTION (Technical Field) The present invention relates to an array element position error measurement method for measuring the deviation (position error) of each array element from its rated position using a signal source whose position is unknown. .

Linear, planar, or three-dimensional arrays made up of a plurality of elements arranged in space are widely used in sonars, acoustic positioning devices, and radars.

The purpose of this array is to maximize the signal-to-noise ratio of a received signal and to estimate the direction of the signal source that generates the signal, both of which are realizations of spatially matched filters. In arrays used for such purposes, if there is a deviation from the rated position of each element in the array, that is, a position error, mainly due to the manufacturing technology of the array, the position N of each element will change.
The error causes a delay or phase difference between the output signals of each element (hereinafter referred to as "delay error" unless otherwise specified), and causes delay or phase mismatching in the array as a spatial filter.

Therefore, the positional error of each element in the array is an important factor for evaluating the performance of the array, and many efforts are currently being made to understand the positional error.

(Prior Art) Conventionally, the method of measuring the delay error of each element of the array includes:
Using a signal source whose position is known, each element of the array receives a signal from the signal source, detects the reception time difference or phase difference between the output signals of each element of the array, and calculates the time difference or phase difference. By converting to the positional deviation of each element of the array,
Methods have been used to determine the positional error of each element in the array.

The principle of this measurement method is that the signal source position with respect to the reference coordinate axis of the array is known, and therefore the output of each element (the reception time or phase difference between (Since r is known, the delay error of each element of the array can be found by subtracting the known quantity from the detected value of the reception time difference or phase difference. By converting each delay error into a position error, It is assumed that the position error of each element in the array can be determined.

Therefore, in this position error measurement method, it is necessary to accurately know in advance the position of the signal source with respect to the reference coordinate axis of the array. For example, if the signal source position direction is set to an accuracy of ゛01
In the case of arrays used in sonar or acoustic positioning devices where estimates are required to be on the order of degrees, prior to a+++ determination of the position error of the array elements, an estimate with respect to the array reference coordinate axis with an accuracy of at least 0.1 degrees is required. ^- It is necessary to decide the direction of the bow source position. Moreover, if the position error of each element of the array is a two-dimensional position error, the position of the signal source at two different points determined with such precision can still be compared to the case where the position error is a three-dimensional position error. It is necessary to determine in advance the original position with three or more points. However, arrays used for sonar and acoustic positioning must be measured underwater at a certain depth or above, taking into account multi-space effects on signal propagation. Since the signal is measured while being attached to a floating or underwater structure, it is difficult to accurately determine the position of the signal source, especially its direction, with respect to the reference coordinate axis of the array.

On the other hand, the performance of sonar and acoustic positioning has been increasing in recent years, and along with this, the demand for delay matching accuracy of arrays as spatially matched filters has also been increasing. Therefore, the conventional array element position error measurement method, which is based on the premise that the signal source position is known, cannot sufficiently meet the requirements for accuracy, and in addition, the cost required for measurement is high.

Figure 1 is a theoretical JJ diagram showing the relationship between the positional error of each element in the array and the delay error of the output signal of each element caused by the positional error in the case of an A+/-1'F plane array. be. 1o is a planar array; 11. ．． 112.

..., l 1. , , , I ZM are the elements of the array, X and Y, respectively.

Z is the reference rectangular coordinate axis with the X and Y axes placed on the array surface, θy, (k), θy(k), and C7, (k) are the coordinate axes X, Y.

Cosine angle in the direction of the th source position with respect to Z, β(k)
is a three-dimensional direction cosine vector β(1,)4(:cos(θx(k)), C05(θy
(k)), Ct)S(C7(k)))', P is the previous i-th element 11. The three-dimensional position vector including the position error is Pl-[Xl, yl, Z1]. However, k
is a number representing the signal source position (k=1.2...2k). Further, the subscript T represents transposition. In the planar array shown in FIG.
c ” H', r'! 4.0) Represented by T,
In addition, the direction cosine vector that is directly measured by the array at the rated position is flY(k) = [IC03(θ
x(k)), cos(θy(10))".Here, the position error included in the position vector P is generally Δ
P1-[ΔX1. Δy1. It is expressed as ΔZ1:]T.

Therefore, due to the position error ΔP1, the signal propagation distance is
If the distance from the array to the signal source is sufficiently long, β
Since it is displaced by T(k)·ΔP1, the time delay error of the output signal of the first element of the array is βT(k)·ΔP,/C, where C is the propagation speed of the signal.

If 1βT(k)·ΔP, l is sufficiently small compared to the wavelength λ0 of the signal, the phase delay error of the output signal will be 2πβ·ΔP, /λ0.

Figure 1 shows only the case of a planar array, but if the array is a straight line, P, = [: x, , O, O:]", 0
1(k)4os(θ(k)), 3-dimensional 7v(D
If K'd, ,P, 4cx, ,y, ,zr:
]”11gk), 4[C03(θX(k)), c
os (Z, (k)), cos (θ7(10)
1:]T, there is no essential difference from the case of a flat array, and therefore only the case of a flat array as shown in FIG. 1 will be explained below. Still, without losing film properties, the positional error will be explained below with reference to the array element A11. .

112,..., I 1st element 11 of IM.
Relative position error Δmachi, 4ΔP1−ΔP1−[
ΔX1.4+Δyi, 1'Δzi, 1)", however, Δ,,
4ΔX1−ΔX+1Δyi, 4:=Δy1−Δy1,
It shall be defined as Δz 1.14 Az +′−Δz1.

In the conventional array element position error measurement method, it is assumed that the direction cosine vector β(k) is known, and the #(k)
The reception time difference -βT(k)(P, -P,) of the output signal of the other element 11 with respect to the output signal of the first element J11, which can be calculated from
By subtracting (k)・(P, −P,), the delay error Δ−,, (k)=−β(kL(ΔP, −Δp,)/
The position error ΔP1 is determined from the delay error Δτ1.1(k). Therefore, the position error 1 ΔP1, 1 or 2-dimensional In the case of position error, l(≧2, that is, two or more different signal source positions are required, and the Δ
If P1,1 is a three-dimensional position error, ≧:3, that is, three or more signal source positions are required.

(Objective of the Invention) An object of the present invention is to eliminate the drawbacks of the prior art described above, and to provide a constant force formula for array element position error 411j that does not require prior information on the position of the signal source.

(Structure of the Invention) The present invention uses a signal source whose position in space is unknown.
In an array element position error measurement method that measures the deviation from a rated position from an array element consisting of a plurality of elements, discrete time or Phase delay compensation [■ Determine the time or phase delay difference between the delay compensator that applies the output signal of each element of the array J, and the output signal of each of the remaining nine delay compensators based on the output signal of one of the delay compensators. a delay detector for detecting, an accumulator for accumulating the output signals of each of the delay compensators, and an envelope for detecting the output signal of the accumulator) 61 wave a (!: ,M envelope a first signal source position estimator that selects a discrete value corresponding to temporal or phase delay compensation that maximizes the output of the zero detection 5; a second signal source position estimator for determining a signal source position that maximizes the output of the envelope detector; This is a position error measurement method for array elements, characterized in that the position error of each element in the array is determined from a constant of the array determined by the above-mentioned constant and the outputs of the first and second signal source position estimators.

(Example) FIG. 2 is a block diagram of an embodiment of the invention.

11. 112, ・, I 1. is an array element,
21. .

212,-. 21M is a delay compensator, 22 is an accumulator, 23
is an envelope detector, 24 is a second signal source position estimator, 25 is a second signal source position estimator, 26 is a vector adder, 272, ``' + 27+1 (is a delay detector,
282, . . . , 28M is a delay amount corrector, 292.

..., 2~U delay error calculator, 3oZr'..., 3~
31 is a vector converter 1. ? 2u Renostar, 33 is an inverse matrix calculator 1. ? 42, -, 34
M is a position error calculator, 35°, and 735 is an output terminal for calculating the position error X:.

Regarding the signal from the signal source, the positional force i is unknown, but it is located at the first (th) position in different directions, and α(k) (
A signal from a source that is a two-dimensional vector (at each element in the array 11..11° in FIG. 2).

, the received signal is received by the delay compensator 2.
Ha, 212, . 27. , , you will receive extended compensation.

The delay compensators 21□+212, . 21. When the array is a planar array as shown in Fig. 1, the amount of delay compensation that can be taken by is a discrete value αn corresponding to n−1,·,N1)
T(k)I) 1/c, delectrick)l' 2/c
2. ”, '(k)-1'7C. However,
p ; I ``'' 1 +...1M is the rated position of the first element 1 of the array, x, Y-v! - Assume that the two-dimensional vector of improvement is FA [ma, , ,,)T.

1 1 1 The delay compensators 211, 212, . 21M output words are accumulated by an accumulator 22, and the output signal is outputted for the N directions, and is converted into an envelope signal by an envelope 0 detector 23. The level of the envelope signal is determined by the discrete set value of the direction cosine (k): n-1, -
, H. The first signal source position estimator 24 calculates the discrete set value σ of the direction cosine: n ” 1 +..., N
A discrete value that maximizes the level of the envelope signal within
It is called a discrete estimate of . ), and the second (Fi source position estimator 25 uses the level value of the envelope signal corresponding to the direction cosine (t (k)) in the vicinity of the discrete estimated value cr (k) to perform interpolation, for example. By the means, the direction cosine (hereinafter this will be expressed as Gt(k)) that maximizes the level of the envelope 0 signal, the direction cosine (1-(]<)
is called the negative estimate of . ).

The Vero I/L adder 26 receives the discrete estimated value α. (k)
and the true estimated discount k), the difference Gln(k)−Gt(k) is calculated.

In the delay detectors 272 1, 27y, the delay compensators 21□, 22, . . . , 21. The amount of delay △ compensation "nT(k) 'f' i/C, α."'(k
) F2/C,・−,”,′(k)pM/C delay amount Δδ′(k),・,Δδ,7 (k) is detected and input to the delay amount corrector 28z, -, 28M.The delay amount corrector 28 detects the delay amount Δδ2'(k).
.

Δa Q<) and the output Gtn of the vector adder 26
From (k)-σ(k), the delay compensator 211+21
2 +...

2's delay compensation is (k)・l)+i's 9''(k) F
2/c,・.

Other delay compensators 212 based on the output signal of the first delay compensator 211 when assuming that dT(k)p, 〆C, 21. The delay amount of the output signal (hereinafter referred to as the true delay amount) Δδ2(k), 46
Find M(k). True delay amount Δδ2 (k) −, 4
6M(k) is the delay error calculator 292. ..., 28, and the delay error calculator calculates each element of the array using the true delay amount and an array constant determined by the geometrical shape of the array and the load related to the amplitude of each element of the array. Relative position error ΔP 2. ,,・,Delay error Δτ2(k) caused by Δtown,1 , −,Δ7. (1<)
, However, L ΔT, (k) −−βr(k) −ΔP,
, , /C; i=2. ...1M is calculated, ゛/nostar 302. ....

30M is input. Said Reno Star 30, Hani=1, 2
.

The calculated value Δτ of the delay error in 3, (1), Δ,
(2).

Δτ(3) is stored, and a delay error vector Δr1 is formed] [Δτ, (ILΔτ, (2, 1, Δτ, (3))”, and each delay error vector Δτ2.·, Δτ is a position Error calculator 34□, .

34. Input to A. In the following description, it is assumed that the positional error ΔP1,1 is a three-dimensional positional error, and is set to =3.

First, the vector converter 31 estimates the second signal source.
2, the true estimated value %(k) of the direction cosine obtained from %(k) is converted into the estimated value 9(k) of the three-dimensional direction cosine vector I-/β(k). In the case of a planar array and signal source position as shown in Fig. 1, the true estimate stand (k)-C~('Q,
(k++.

[Surrounded by (θx(k months, CQS(θ, (k+) , , x
Find (? (k))ゝ〕′, 9(k) is △
△ Manpower is provided by Renostar 32. The renostar 32 is 1(-1
, , 2, 3, the above estimation ij*, 9 (k)
The data is memorized to form a 3/3 matrix, and the data is manually input to the inverse matrix calculator 33.

The inverse matrix calculator 33 calculates the inverse matrix of the matrix day, and the retrograde day -1 is calculated by the position error calculator 34□, . . . , 34M.
is input.

Position error calculator 342. . . , 34M, the delay error vector Δτ1 . ..., Δ寥8. and each element 112. of the array from the inverse matrix a-+. ..., 1 position error Δp2. ,
,...

q-out ΔPM, 1, ΔP 2. ,,...,Δh
, 1 are output to each output terminal 352, . . . + 35 p, 1. In the case of this embodiment, the retrograde day -1 is K=3
The direction cosine of the acoustic position - ctor e(1), 9(2
), 9(3) are linearly independent, it can always be calculated.

FIG. 3 shows the delay amount corrector 282 of FIG. . , 28M is a 3D diagram showing an embodiment of the first delay amount corrector 28. 40 is an input terminal to which the output of the first delay detector 27 is input, 4 is an input terminal to which the output of the vector adder 26 is input, 42 is a register, and 43 is an inner integration for calculating a vector inner product. Output device, 44, is adder, 45
is the output of the adder 441 to each delay error calculator 292 .
. . . is an output terminal for outputting to 29M. The soster 42 has constants (P, -Pl) related to the rated position of the array.
/C is stored. The output of the register 42 is input to the inner product calculator 43 from the input terminal 41, and the output of 43 is the delay amount Δδ1(k) input from the input terminal 40 to the adder 44□. The true delay amount Δδ1(k)−Δδ1(kL−(p
, −1i, )T(Qn(k)−Q(k)+/C) is determined, and Δδ(k) is outputted to the output terminal 45.

FIG. 1 shows the delay error calculator 29,', i=1 in FIG.
It is a professional diagram showing an embodiment of ゜...9M. 50□2
゜50, . . . , 50 are respectively delayed by the correctors 282, 2
85° 13 1M ·, 28M output or input terminal, 51. □.

51 ] 3 +..., 51. M (respectively 41 calculators, 5
2□2, 52,5.

..., 52. M is each register, 53□ is an accumulator, 54
, are output terminals for outputting the output of the accumulator 53 to the accumulator 30. Each register s2,2,52
,5,...

52.9 stores array constant numbers Si2'Si3'..., S, M which are determined by the load related to the geometrical shape of the array and the amplitude of each element of the array. -Fure constant number S12.S13...' 83.
and the delay amount corrector 282. ..., the true delay amount Δδ2Q<) which is the output of 28M.

Δδ5(k), ·, Δδ, (k) means each multiplier 511
After the products are multiplied by 2 + 51151..., 51, the accumulator 53□ accumulates the results in the position error calculator 34□, .

FIG. 3 is a block diagram showing an embodiment of the 34M. 60 is an input terminal to which the output of the first register 30 is input, 6 is an input terminal to which the output of the inverse matrix calculator 33 is input, 62□ is a register, 6311, 63□2゜631
, is an inner product calculator that calculates the inner product of each vector, 64
64. 64.3 are the first output terminals 11 I
12 This is an output terminal that outputs the x, y, and z components of the position error that are output to the child 35.

In Cistero 2, the retrograde resemblance is CdI+ for each row vector d1T, d2T, and d3T of the matrix, that is, d
'd1□, a13]", d2=4 a21.a
2□, d23]”, d3=[d3.

d3□, d33] is stored for each d, , d2.
The delay error vector τ1 is inner-producted by each inner-product q-output unit 63□1°63.2, 63.3, and then outputted to the first element 11. Position error ΔP, 1-CΔX1
1. Δy11. ΔZ1. ) Each component Δx1. -d11
11. Δy11=d2T-+r□, Δz11-d3゜τ
1 is detected, and each output terminal 64□1, 64□2゜6
4□6・Furthermore, the true delay amount Δδ1(k) and the delay error Δτ1(k)
, position error Δmachi, 1 will be explained.

When the direction cosine of the signal source position is /(k) and the signal source is sufficiently far away, the delay amount of the output signal of the first element in the array shown in FIG. 1/T(k)・(Pl-P, )/c =-βT(k)
・ΔP, 4-α”(k)<6.-6.)/c-Δτ, (
k)-JT(k)(p,-1:), )/C(4). However, Δτ1(k) is the value of array element 11. This is the delay error caused by the position error ΔP3,1 of .

△ On the other hand, the discrete estimated value α (corresponds to 10) of the direction cosine
The delay amount Δδ1'(k) detected by the detection section 7□ can be calculated by Jj using the following equation.

Δδi'(k)−Δτ,(k)−(”(k)(I),−
1:),)/C+'CnT(k)(E),-47),)
/C−Δτ1(k) +(stand(l()−α(k))T(
βi-p, )/C-t (4nn(k)-Q (k)
) T(1:), -1p1)/c-Δτ1(k)+Δ
αT(k) (p, -p, )/C where Δα(k) is 1 for each element in the array. , , 1 position error ΔP2,1.
..., to Δ, each delay error Δτ2(k) by 1,
・True estimation of the direction cosine in the second signal source position estimator 25 caused by −2ΔτM(k) (if meeting (k
), which is the estimation error of Δα(l()) σ(k)−α(k
).

△ Here, the true delay amount Δδ+ (k) is expressed as Δδ1(k)
Δτ1(k) Δσ”(k)(1), −Dl)/C(
6), the delay amount Δδ1'(k) detected by the delay detector 27 and the discrete value obtained by the first and second signal source position estimators 25 Estimate Q (k) and antilog 1), −pl to Δδ, (k) = Δδ,′(k)−(′Q.(k)−
(k))T(β1-11)/c (7).

Next, refer to the document “Igarashi: 5SBL Acoustic Acoustic Measurement Side Angle Error Due to Phase Error”, Institute of Electronics and Communication Engineers, Study Group 5ANE82-
15.July 1982'', the period of the center frequency of the signal is T. Toshigi, IΔτ, (k)/To1<
V2π;1-2. ...M, that is, period T. The delay error of the array Δτ2(k) , · , 77M
When (k) is sufficiently small, the estimation error Δα(k) of the true estimated value α(1()) of the direction co-chord caused by the delay error is given by the following equation.

However, bl;i=1. ...1M is the number of each element in the array
.

112, . . . is a weight related to the amplitude response of lIM,
1m) c, H are constants of the array, and if the array is a planar array as shown in FIG. 1, -1; a two-dimensional vector; a 2×2 matrix.

Therefore, by substituting the relationship in equation (8) for Δα(k) in equation (6), M-1 Δτ near(10,...,77M(
The following simultaneous equations with 1<) as unknowns are obtained.

−・・−(11) However, here; j=2. ...1M, 2.・-・M
is an array constant determined by the geometrical shape of the AJ loop and the load of each element of the array, and is given by the following equation.

■<,,, 'E: lb,, 12(p, -p.)''(
, protein, lb, 12H a' (p-p) (12JI

JC Therefore, the right side of the above formula 0]) (M-1
)X(M-1) if the next matrix is a regular matrix, an inverse matrix of the matrix exists, and each element 112. ., 11M position error ΔP2. ,..., Delay error Δτ, (k) caused by ΔPM,1; i =2. ...1M is given by the following equation from equations (9) and (11).

Δτ(k) = ΣS ・Δδ(10(14)j=2 Since each element 51J of the inverse matrix (r is given only by the constant KJ□ of the array, the element S is predefined by the constant and I~, You can leave it out.

Next, the delay error Δτ(k): i = 2,...
If 7M is calculated by the above formula 0'3 for 1(-1, 2, 3, τ] (1□ -1-~If)T(k
)Δl:l j , 1/C1ΔP, , -[ΔXi, 1.
Δyl, 1+ΔZl, 1) From the relationship of T, ΔX1, 1.
The following simultaneous equations with Δ3'5j and ΔZ1.1 as unknowns are obtained.

1-2. ...1M Note that the error due to the approximation of Equation (2) is a second-order or higher error and can be ignored if the position error ΔP1,1 of each array element is sufficiently small.

From the above equation (1$), if the matrix on the right side of the equation is regular, the inverse matrix of the chain equation, equation (3), exists, and therefore the position of each element of the array Tf''-112, . . . , 11 Error ΔP1.
, ;l-1°...2M are given by the following equations.

In addition, in the example, the number of different iL source positions is set to = 3.
However, it is also reasonable to take K>3, and in this case the position error ΔP1,1 is τ, (], ) , τ, (2), ・
....

It can be determined from τ1(k) by, for example, the least squares method.

(Effects of the Invention) In the present invention, (direction cosine β(k) of the position of the i source, that is, a signal source located sufficiently far away: k= 1.2,...
, even if K is unknown, the amount of distributed delay compensation is determined by the delay compensator 211, 21, . . .
7r, given, the delay amount [delay amount of the post-prize signal with respect to the reference signal Δ(k),..., Δδ; (10 is detected, and the direction of the signal source position corresponding to the decoupling delay compensation needle is The envelope detector 2 is calculated by means of interpolation or the like in the vicinity of cosine (estimated discontinuity value).
The estimated value of the direction cosine (true estimated value) Δ of the page that maximizes the output level of 3/'(k); k = 1. ..., the Jf rate tD53z number S22. given only by the geometrical shape of the array and the unloading of each element of the array.・・ls
MM, the relative position error ΔP29 of each play element with respect to the reference element is calculated using MM. .

. . , ΔPM, 1 is determined, so when determining the position error of the array element, the direction cosine of the signal source position # (k
); to=1. 1j, which does not require prior information on...
There is a point.

Therefore, when measuring the position of array elements used in fields such as sonar, acoustic positioning equipment, and 1-node devices, where it is difficult to accurately determine the signal source position, it is effective in improving measurement accuracy and reducing the cost required for measurement. but

[Brief explanation of the drawing]

Fig. 1 is an explanatory diagram showing the geometric relationship of the array, Fig. 2 is a block diagram showing an embodiment of the present invention, Fig. 3 is a block diagram of delay amount compensation hemostasis shown in Fig. 6, and Fig. 4. is a block diagram of the delay error calculator in FIG. 2, and FIG. 5 is a block diagram of the position error calculator in FIG. lO; Planar array, 1111..., l1M near l/lO
Motoko, 211. ...+21y,: delay compensator, 22;
Accumulator, 23; Envelope detector, 24; First signal source position I′ estimator, 25; Second signal source position 4i (determiner, 26; Vector i·le adder, 272. ..., 27M
; slow 9L amount detector, 282. ．． 28M: Delay compensator,
292°, 29M; Delay error calculator, 30□,..., 3
0M; 1/nostar, 31; vector converter, 32;
Nosta, 33. Inverse matrix calculation profit, sl2. -, 34. ;Position error calculator, 42. ;Renostar, 431;Inner product calculator, 44;Adder, 51,2. ...'511M: Multiplier,
51] 2 +..., sl,,; Lesosta, 531;
Accumulator, 62; Nosta, 63. ,6.3,2
．． 6.3□U; Inner product calculator. 1 @ 3 Figure 4 Figure 4 51j+x Figure 5 Procedural Amendment (Resistance 59.2.14 Showa edition) 4 (-Monday' Samurai ``1 Gocho 1'-: To 1: Tono 1 Matter f'1's life 1981 Patent Application No. 28585 2 Name of the invention Array element position error measurement method; 3 Relationship with the supplementary iJE case Patent No. 1 (xr4
1 person Location (〒1.05') Tokyo Minato Toranomon 1
No. 7-12 Name (029) 5 Chuo Electric Industry Co., Ltd. Representative Director and President Minami Hashimoto: Masao 1
1: Location (105) Toranomon 1.T, Minato-ku, Tokyo
f: 1. No. 7 No. 12 G Contents of amendment Details as attached (“Title of invention” column remains unchanged)
11; 1. Name of the invention Array element position error measurement force formula 2. Claim space "2 position iI'i'; F is unknown a, "F %
jl'ri 4: Kawashi)(,i(,・7J Yu case i\"l',' of the element-J' of an array consisting of several elements)
i: position of the array element at which the deviation from “l” is measured, 1
14 difference 11ili', in the M method, 1 (discrete time y corresponding to the numerical value) regarding the predicted position of the signal source
&1st place O[]Delay sleeve compensation for the output signal of each element of the array -riero delay sleeve f21; Output part of the instrument - Between [1 and 5] (T1, a delay detector for detecting a phase delay difference, and an accumulator for accumulating the output signal of the above-mentioned zigawa flute, l) 1. 1 (1% cumulative/Jll:',
1-: The output of Shingetsu's envelope 1.1- The device to emit 1 moxibustion of zo, / bellow) "detector, 1) i, + envelope 0
A first method for selecting a discrete value for the position jKi', K that corresponds to the time or phase delay compensation that maximizes the output of the detector (input position estimation 8:; and the first step) Source position i
4. In the vicinity of the discrete value selected by the estimator, the output f of the envelope detector is ' L...
,, F'+: The constant of the 1 cedar array determined by the output of each slow cedar g output RH and the geometry and shape of the array, and the load of the width response of each element of the hi array. The first and second (iJ-
A method for measuring the positional error of each array element based on the output of a positioning device. :3 Detailed explanation of the invention [ 1 (Technical field) The present invention uses a signal source in which l):(, 111 Element position error measuring force formula for an array that determines Alternatively, a three-dimensional model is widely used.
Shinzuki Geni, which generates the signal in J°1☆of the fB ratio, ,f,! This is an estimation of the force surface, and both are realizations of spatially matched filters. In an array for such a purpose, the positional error of each element of the array within the rated position based on the manufacturing technology of the array is If present, the corresponding number of each element)",
Wrong? :b +=, delay or phase difference 1 between output signals of each element? (hereinafter referred to as "1 delay error" when no distinction is made between them), and the delay in this delay as a spatial filter causes phase mismatching. Therefore, The position of each element of the array is 1 yen, 1 error (bow 1, as a no-after to evaluate the performance of the play).It is cheap, and a lot of effort has been put into understanding the position error. is currently 7° (prior art). Conventionally, in the method of determining the iY hood error of each tree in the array, the position is known (the M sound is once a month, and the signal from the signal source is and detect the reception time difference or phase difference between the output signals of each element of the array.
Therefore, a method for determining the position error of the array element -r is used to convert the time difference or phase difference to the position error of the array element -r. The principle of the three-plane measurement force method is that the signal source position with respect to the axis is known,
Therefore, the reception time or phase difference between the output power of the element S at the rated position in the space of J, J' is known, and , the known quantity is i
``iiJkibutsu-suke difference or place 44, Detection of ``lzuri'' I1
〆By subtracting 2 from 1, find the delay ζII, error; 1 of each of the arrays A. It is assumed that the position error of the element can be calculated b'). Intermediate-C1 this. 1. In the error measurement force equation, the signal source position with respect to the E axis of the array is determined in advance (... whether two phases are necessary or not). For example, Shinsei Borrow - Placement; Ding j with shield 11" and 01 degree (!)
It may be necessary to make a correction to the order by using a non-ner or an acoustic measuring device (/Kan Jr.)
-- i, 4Aj combination)' 1 tatami 1, the position of the play element:; ↓ Prior to measuring the difference, the accuracy is at least within 01 degrees. , 7il'L jI mark/g) It is necessary to determine the direction of the bow source position regarding 11. Alternatively, the 6'L position error of each element of the array, the quadratic ji +'+-r position difference position charge (where = the signal source positions of two or more different points determined with such precision, -1-/In the case of three-dimensional equipment errors, it is necessary to determine in advance a signal sound f\'J device with three or more points.However, it is difficult to use it with sonar or acoustics411] Ichino "shi・-f□J1, signal train propagation ten multi-gu, su young song with 6 (・1!□11 and 7, '-: 4-depth or more underwater Okaoshi)' is 7111 fixed It must be done (7, (
・）ru. It's a trench.
K6: I was caught in the underwater J'M structure and measured: t
Therefore, it is difficult to accurately determine the position and direction of Nobutsuki () regarding the -) ゛ and -substrate seat 4 votes side]. - Performance of Tsuba sonar and acoustics ranked 4111th (
has become increasingly l″i,s in recent years111'i 1FiJ
Accordingly, the demand for delay matching accuracy of RALS as a spatially matched filter is becoming higher and higher. Its/(Me, Ij; 弼”4hara(I\'1・position of the conventional play element which assumes that i'J position is iJ knowledge: il:↓In the difference measurement method, the accuracy - To meet the above request, in addition, the two steps required for measurement are also 1-t(i.
ti, 2.3° Figure 1 (σ1 array or plane array) position error and position error of each element of the array;
This is an explanatory diagram of the relationship between the 10 delays of the Shokai A and the 10 points of delay between the respective GOs and Ryo. 1
0 is flat++'ti -7L/-111,,112°・,
111. ...Jha, each of the array (・two children, X,
"l'. Z has X and Y axes placed on the array surface/, - group to 1- target r; vote for +1Q11, θ (k), θ
(+, :), θ (k) U front n 1m
1ii1! IX, Y. Y Z [9,1] 1't1 state position force direction cosine angle, β(1()v:x 3 r missing L-force surface n chord'
p (k) 4 [cas (θx(kl4
(θ1, (k)), cos(θ7.(k))),
P, is 'mD i'je No. 1 1 ■ 1 stab completed 11
Including 1q error of;
X, y, 7]. However, k is the signal source position 1
1 1 l [1.4 liters of vinegar (k=1., 2, .
. . . is a number representing 1().・1/.・The subscript T represents trochanter 1. 1) The position vector [・
is P-Δ-1. X, y20] is the force 1, and the directional cosine vector of the direct total 11111 is α(k)JCcos(θ(1(moon,
cns(0(k))]. Here, -y, 1NiJ positional position] - positional error (d) contained in P is generally expressed as JP4[ΔX, Δy, ΔZ]. Therefore, the 1, -, , , l
1 1 position 1 ↓ By the difference ΔF, the propagation distance of the signal iqt, the −array force), and the distance from the source to the source − is sufficiently long.
Since the displacement is straddle P, the time delay error of the output foil of the first element of the array is the propagation velocity of the signal as C and [7
”(10・ΔP,/C.It is 1pT(k)・J
P11 is the wave λ of the month. If τ is ten points smaller than τ, the phase delay of the output tower is incorrect.11. (r) 2π7
β・ΔP17/λ0. Figure 1 shows only the plane array〇 field
'II'' + α0(), so (θ, (k)), in the case of a three-dimensional array, P40''+ + 3' +, 'T
J + (f(k)4 (cos(OX(k)), c
os (θ, (k) l , cos (θ2(k) l
3 and dimension rtic, 'F iA, i array case and this Jf4 (because there is no change, 1'),
1. In the case of the plane −f ray shown in FIG. -Element 111
. 112,..., relative position 5 difference Δp1, 14ΔP based on the first element I11 of l1M: 711I:
-JP, = [ΔXl+”Δ71.1+ΔZ4.+], However, ΔX, 4ΔX, -ΔX1・Δy1,1 迭Δy, -Δ
y1. It is assumed that ΔZ1,1 is defined as ΔZ1−ΔL1. The position error of the conventional array element is
, assuming that the direction cosine vector l3 (10) is known, and using the 13(k) and the output signal of the first element li1 that can be outputted from the rated position v1 as a reference,
IJ's appearance J1. ;-Sea J's 1g time difference-/”
(lri) (P 1P 1) Actual measured value - βT (k
)・(P , −P 1) by subtracting it from the delay 1 error corl(k)4−. 1: βCk) - (ΔP ΔP+ )/C-p ” (1< ) ' JPi
, 1/C fixed tree 17'), the delay, error Δτ,, 1(
k) to position error ΔP12. Therefore,
1ftl If the position error ΔPi,+ is a two-dimensional position error, k≧2, that is, two or more different signal source positions are required, and the ΔP1,1 is; In the case of an error, f/i k and 3, that is, a 3-point VV +. (1-) The purpose of the present invention is to eliminate the shortcomings of the prior art of MfJ ports (particularly, to eliminate the need for prior information on the position of the bow and bow source).
-・No, the position error measurement force formula of the un-element j'- is all provided 1° (-i' vinegar! JJ's 4'F'N configuration) The present invention 0 boo X! a The position j1'/f error of the array element is measured by using a signal source whose position at IHj is unknown, and the deviation from the nominal position, y+, from the element of an array consisting of several stages of elements. +! In the 411 power formula, the prediction of the signal source, )'111, the discrete time or phase delay complement corresponding to the gradient 1 is the output of each element of the array +, -i -'': Month t '4, j4i delay compensator and each of the remaining output signals of the delay compensator as a reference,
a 1j-1 delay detector for detecting a time or phase delay difference;
The output signals of each of the eight compensators are cumulatively summed by 4! ＞
i and the envelope of the output signal of the accumulator 0&: e
an envelope detector with Σ output 1 and the envelope 1-2-
The time or fd phase delay that maximizes the output of the 0-channel detector (
1; The signal source position i1 of tl which selects the discrete value corresponding to the compensation (・ζ・χ・J) is selected by the detector and the first signal source position estimator tF 'I: & the vicinity of the value The said envelope =
The output of each of the 8 wave detectors and the geometry of the array are and the constant of the array determined by the load ↑1j of the amplitude response of each element of the array, and the llj force of the signal source position estimator of the first and :j02. This is a force formula for measuring the position error of the Anne element, which has all the characteristics.
It + l 12 +・', 11M is -fV-
Element, 2 no 1゜212,..., 21M is a pure compensator, 22 is an adder, ゛! ? , σ1 Enovel rJ-P he; Base, 24 is the first signal source Ij'l:l Self 11゛ constant c(d,
? , 57 Second Satoru M Ryo CD device 114 constant device, 26
(・-effect “Guri 1・) V force 11 calculation′ crystal, 272,...
, 27. L is late furnace detection □, heat, 282, ・928M is delayed furnace IE device, 29□. 2 g, 'a, 'Zoen 1 difference A output device 1, No02,・
・,308ba1・・Zosta 1. 'j7 vector(・le transformer, 32 is Lenosf, 33 is degree matrix -I:9-output device, 3
4□ ,..., 34M6 'i\(!-position error ρ゛ output, 35□ ,..., 35M is the →output value of the position error. 1st line, H> from the source Regarding the box number, the position is unknown, but there is a sufficient cooling force (1'), 2L, the th IM in a different direction] i'i has a direct force of 1, and the direction cosine of that position is The signal from the signal source whose size is <lo (two-dimensional vector) is 21st m>:1 for each element 111.'
1 no,? , ・11, So, the Ukei 8th Shin-Yumi (
d) Delay compensator 211 r 212+..., 21
Receive delay compensation at M. Dazzling delay iri'ba'r'+'ii271 + 2
12 + ・, 21 Inotori state ru 1j〆en Mi compensation 1
1; is estimated when the array is a planar array as shown in Fig. 1. : (n!, q
j force direction direction, n-1-+-・+ N 1) i3
j, 7g, addition 11 value dnT(k) I) +
i c , α′1TOOP 2/Cl −′ rσ,
” (k) p, /C with 4)).
i-1+...r Mt is the quadratic 1 on the X and Y planes 11 at the rated position of the
-1 vector! ';,! , Cx4, y11]”
Suppose that The output signals of the delay compensators 211+212+ are accumulated in the accumulator 22, and the output signals are outputted for N times of directions: h, , envelope angle detector. 23, the signal is converted into an envelope 0 signal, and the estimator 24 calculates the angle of the direction cosine (
Interspersed 'A' + 1 value σ; n '' 1 y..., to the Enverono 01'M 413 in N - Ga 1 numerical value (hereinafter 1ζ, here) 1 to α. (
1) Expression 1, divergence estimation of direction cosine αte1() 1'1
The second signal source position estimator 25 calculates the direction cosine d in the vicinity of the estimated value αn(k), and the envelope corresponding to (1). Using Shinzuki's level value, for example, 1 Senma - Shallow ('Teyori, [!!I Envelope 0 signal leveling', I-, direction cosine (hereinafter this will be referred to as 1() to obtain the direction cosine (referred to as Mr.'s estimated value of J(k)). ．、
Difference of 9 degrees (10 - Steel k) Call I [Output. In the delay generators 27□, . ”
(k) ・p 2/C, −・ , Q, T(k) ・ 9
M/C'f! : Delay amount ``δ2'(
1()9. Δδya'(k) is used for bow and arrow, and 11j1 (4
The moth suspension is manually applied to the vessel 282, -..., 28M.
), . ΔδY' (k) and the output Q of the vector adder 26,
(k) 4+<) Measurement 11゛ (Tsuka, 13, the delay amount of the output 1 blind number of other delay forceps IA devices 212, . . . , 21M based on the output signal of device 21. Determine the true delay amount Δδ2 (102, 36M(k), 29□, 29. + tc human power tll, the delay amount. l
-1 No - In the error calculator, the true amount of delay and the number of arrays are
Using the Japanese language (l-shape and the thread width of each element of the array ((regarding force 1 [kawako, L: -)],
Array name, :・, Relative position error ΔP
, Δ9, I' Therefore, LI2.1 + delay error Δτ2(k), ・, ΔτMCk), I11 Δτ1(1g)・−1β″C1O・ΔP+′
, + /C: l = 2 +-... 2M C1 out 12, Renosuf 302,... 30. The force input to the The slow Q in the Lenosph, me()bani = 1.2.3; E ij Wu,
Write τ(2) and Δτ(3) as a kernel, and delay 1f! '4
．． ! , 1" spectral Δτ sound [Δτ, (1) , Δτ(2
), Δτ1 (3)] 11℃ and the name ZYI. 'iFA difference vector ΔF2. -, ΔF>4 i-position 1
i''+1 error output f'!: ``i''' 2. ・ , , 34.Inputted in . If the dimensional order 1〆11 error
Estimated value of the direction cosine r from the vessel 25 at 1-'4 fu'+α(1(
): cubic no, direction cosine vector − (k) estimate 9
′(k)/, ri). In the case of flat + i array and 18ji bow source position as shown in Fig. 1, the true estimate j3t4 '')l'k ) = [tBg (' or (
10+. Sufu, n1,? Shi'(-be-)Jsa1. The Lenosph 32 (il<-・], +2 I:'S!'
<-',,l +)i + estimated value β(1()), 3 'I: Ill
gi:J,'? can be done manually. The distance between the inverse matrices is 11゜, violet 3, pa is the inverse matrix of the matrix t3゜,
k '91i, and the retrograde group lj B + is at the position 1
Prefectural difference 1 [effect] K3.1□ , -, 34,, tc human power h~ru. Samurai 1〆7. -↓A-ζ "1 blasphemy 342, -.34"
is the output of the Lenosf 302, -, 30M; 7
% t no L1 error vector Δr1. . . . ΔEM and the inverse matrix, 3-1 to each element 112 of the array.・・・
, , 11M position error ΔP2 , I T "" ΔP
Calculate M, +, ΔF2.1,”' + Δ
To, 1 is each 1 force ζ 4 if 6,000 35 □ , + 35 M
Output with i. [) 11th color j color 11th column 13-1 is
In the case of the false male, K = 3 blind echoes ii':'', I no li'i-1 cosine (ri l l he 1), ha 2), ku (3)
are linearly independent.Table 1. You can always roll a 9. FIG. 3 shows the delay amount corrector 282 of FIG. +28M, JT, No. 1 [1 delay amount corrector 28. This is a diagram showing an example of the system. 40 is the first delay director 2
7 The output of Jl is input, the human power XTi, 41 r, the output of the vector adder 26 or the input human power>j'nl
``11', 42 (d Renosuf, 4.> id vector inner product division V) inner product calculator, 44 is square 1 calculator, 4
5 (r: Delay the output of the four adders listed in LFIJ,
9. ．．・l ”' ”M] This is an output terminal that outputs V. Lenosuf 42. VCl
Rated position of i-y-rule (te19-1 constant (page, go β1)/C
is written 1ζ. The output of the I/nostar 42 is 3 inches,
Inner product - zo) Usufruct 43. In the input terminal 41. Input from △ go to [)) j notation

[, ¥! , (f)−α(1,) and the inner shuttle (j
Seven is taken. Calculation of internal debt: Output of ``i'rl 'i'' (
・”2. In the tactile device 44□, the above formula]-J)!
1 and 1 Rél() / Harajin 7] The delay amount Δδ'
(k) to 15! /c でノJさ:h、ru. Figure 1 V:1:, Delay error calculator 29 in Figure 2; l-
1°, lli'lt/) is a block diagram showing an embodiment. 50.2°, 50□5, "" 501M is valley delay amount correction', 528□. 283 +, 28MII") ii! Power input QNA completed, 5112 1 57 + 3 1
``'T 5 to M each'' 11 units, 52□2. 52.3 ''' 521M ``l each Nosta, 5
3 is an output terminal that outputs all the outputs of the cumulative horn [1 device, 5・1 is the aforementioned cumulative) J[l Niji''i 53, and VO. +, 521M·l (, is the geometry of the array) l'are constants S, S, S, M determined by the curved shape and the amplitude of each element.
2 +3' The constant number S12. S, 3, 2
S18 and i"lit IjY long sleeves 2B...
・, direct delay which is the output of 28M: ¥4Δδ2(1()
, Δδ3(k), -., 38M (1<) means each multiplier 51, 51, -, 5111. Take the product with 1.
After that, 12 i5 is calculated nl Δτ(1() is the output +'@i output to the child 54. , 1 Figure 5 shows the position error calculator 34 □ in Figure 2 ,... 34M actual vertex FIG. 2 is a block diagram illustrating an example.Ja 04j
:r) i + the output or input of the first Renosuf 30 i
:! 't a) Input terminal, 61 is the inverse matrix 'cotton use 3
3's output terminal is input terminal for human input, 62 is Renosuf,
/) 3□, , 6. i, 2゜] 63□3 are the inner integrals of the vectors), 641, , 64, 2.64□5 are the first outputs and □7[I5] children, respectively. This is an output terminal that outputs the x, y, and z components of the position error output to 351. In Lenosph 62, (the inverse matrix B-1 is the row of the matrix) vector d, T, d2T, d3T #7I,
That is, d+-C"+ +. d12 "13""2-[d:! I” 22
"23""3"" (d31 "32" 33
]Remember every T! i-1 each d, , d2゜d3 and the delay error vector

[1 is Taniuchi 61 “5111”,
jH;=63.63, 631. The inner product is taken by
1. J, 'l' i 1+Δz11] Each component ΔX1 of 1
l-d1Tτ1. Δyl i −d2 Tτ+ + coy
, +' +-=ds ” r '+ is Σ' set i7L
1. So, each output width is 1゜64.1゜6412" 1i
3 Ilko・11'lThe power is Lru. S, i i, ·, ′, (“1’s slow electric power” δ(shi), delay error Δτ(1(), position true, 1; ΔP1, 1 each for one friend (3 will be explained. + ?, I' ”7 jll'11.I☆, f', ', j
Is the direction cosine of β(k)? ', the (M M i t'7S> 1qi Jj), rffj of the array as shown in FIG.
The delay amount of the chapter number is -1!5 ``j' /
1's output t; p''('i0 ・(
P'+-P1), /C-27J'T(10・IP 1
．． +/C(X”(k)(p +'-ya-,)/. -koτ1(l()-σ7θ()(脣1-I),)/C
(4) Ear X- et al. ilIL7, Δτ(1() is 11171 of array element II; ::'-<M
JP i,! ”・The delay error caused by this is △ One force, the distance in the direction of the force indicated by M'l is estimated to be α□ (L for l(), z; () field 27. 31 I fly amount Δδ,'(10t4th generation -riera no,ru. Δδ,'(k) -Jr,(k) -ffT(1()(
p 1−P, )/C+ff. ”U (1) i-p,
)/('−Δτ1(”) + (Q(1<) (1
'(k))T(stone, -P,)/C+('9.(k)
AlO)T(p, -p, )/C=ΔT,(k)→
-Δ(1'(k)(1),-1),)/C+('5.(
k)-'9(k))T(1)(1), )/C(5), Δ01(k) is the array capacitor's 12 r ·J
J hui. The position error ΔP , , Δ'M , I (fc
J: 7. ,) Name N Bait, jll'j '／)#'2
．． 1' Δτ2. (IOr ・, ΔτJ<) The second factor 1 caused by the true delay? Δδ(k) k
J δ1(1<)g JT, (10+, #T(f)
(1) □-1)+ )/C(6), then the Δ
δ1(k)! d riiJki NN extension detector,? From the delay amount Δδ,' (k) detected in step 7 and the above] and second transfer constants pi Pj, (k)=, J/i,'(1,) <y4ノ. (f') −'2.i<))'(p, --p, )/
c (7)i'(7,J': J:i tree)1)
You can do that. I: < (1(, sentence, ilI, r5I-Arashi: ″Phase error V Koyoru SS BLs'r7 Kataken I:', 'f (Q
ji'i'l 1'+Incorrect A・°°, 11L Communication Society, Study Group 5AND: 82-15.July 1982') (・ko, l:ni' in y3) Shi Zhou?
T for the number of skins, 1 and 1. As, (Δr, Q+), /'
rl]l<: l, "2yl:; 1=2 r-M
%zunaJ-)chizhou)υ11', 1. Compare with (-7,
Delay error Δτ2 (k) , , , of kernel a j,-. "τx (IO or 1-minute:,'
(1() estimation error, ``(芙・,1<)・″''::missing
[(te゛ j→ 2 people "ro) 1 ru. IE, l, t-, +, h';i" i, "',
M is the array element 111°/17 r ”' r
lhl's amplitude home, answer regarding the answer, size, /kop
, H is a constant of IA 'f t'' -, and the array is
,; two-dimensional vector; 2×2 matrix. Therefore, by substituting the relationship in equation (8) into JQ'(l) in equation (6), i+1, M-1 Δτ200. Since . (1]) Idanji, here, ;J22. ..., M, 1--2.
-..., M, 11 is the geometry of the array and the load 11 of each element of the array! It is a constant of the array determined by the following formula "0-rp.
121-1)-' (possible 1-pc) t, x2 same,
r'if+note j(+, 1]) (M-1,)X(
M-1) The next 11 columns are positive (the 11th '31/lj, and the inverse of the matrix exists in b), and the position of the -Furuno capacitive element 112...r l l,. The delay error Δτ (+<); + −2+ ・2M caused by the error JP7.1+ ”' + dP, , is:
The above formula (9) and (1η) are 4j gills 71. Δτp) = , '21 S 1.・Δδj(k)
+1 -2 The above series? The 3' catfish element s61 in column i is the constant of the array,
), only - Lich et al. No. 1. Rumo, the three elements s, 1,
is set in advance as a 5E number as -9V ill, l, . Next J, ・(, j)
1(k);I-2,2M or 1ku2I r 2, '3
(/4 and the above formula (1) is used to calculate the ratio of ``r'', (k) ---βT(10dP,
, , /C, ΔPH, +=CΔXH, + +”
i, M 71zr,]T's ``From section 11, ΔXi, 1'
The next simultaneous chi flXl with ΔV:,,1yΔz1,1 as an unknown quantity is obtained. 1-2, ... 2M Note that the error due to the approximation of equation (19) is the (sjl position error ΔP1,1 of the array element -) if it is sufficiently small, f'L (r
J: Secondary and higher 1. This error is negligible. From the above formula (1→, the matrix on the right hand side of the substituent is jF law, jl,
(, -, the inverse matrix of the equation (3) exists, and therefore each of the play elements 172 +"'+ 11M device error ΔF, , , , ]=, J. Each component of 1M is It can be written in the z-order formula (・(:
Become.゛In addition, in the example (,), the number of different 1 source positions is assumed to be K = : (5, but K) 3, and this 1 also corresponds to the f~n position of 1. Error ΔPl, 1 Δτ, (
1) , ΔτH(2) + ”' + 3 (from 1(), it can be obtained, for example, by the least squares method. (Effects of the Invention) The present invention-C( ^
direction cosine!j(1(); k=t,
2,..., even if K is unknown, each of the delay compensators 211, 212, 212,...
1M, and after the delay compensation, the delay amount Δδ2/(k) for the reference signal of (M) , , , Δδya′(k
) is detected, the direction cosine (νjII discrete estimated value) of the signal source corresponding to the discrete delay compensation point is detected.
Intermediate. (10 and p n (1 <)) ((By means of Sodoma tongue, the true direction cosine (t'↓) which maximizes the output level of the envelope 11-p detector 23
l′l: ) k) is determined, and the delay amount Δδ2′(lO
2, Δδh+ 'Lk); k= and true estimated value P
(k: y= 1, -, ■<, and array 1.) An array constant S22 which can be obtained by -Lj only from the geometrical shape and the load of each element of the array. -, Syu9. using
The relative value of each element of the array to the reference element is 11γ(4,1
Error-2; , ΔP, , l..., Δ9, 1 is determined, so when determining the position error of the Flusea element, the direction cosine of the signal source position β(k); k=1, It has the advantage of requiring advance information of 7/, I:. Therefore, it is suitable for fields such as sonar, acoustic positioning equipment, and radar, where it is difficult to accurately determine the signal source location. It is effective in improving the measurement accuracy and reducing the cost required for the measurement of the position of the array element used. The theory showing the scientific relationship, the four circles, the first 21 and the threshold are block diagrams showing actual flight examples of the present invention, and FIG. Figure 2 is a block diagram of the (I□Yl) 1j equipment, Figure 5 is a diagram of the block diagram of the position (j'A: in Figure 2). Array, l11, , IJM; Arraying) So-Ding, 2-1,..., 28; Delay compensator, 22
; Adder, 23: Envelope detector, 24; First signal source position estimator, 25; Second signal position determiner, 26
Adder, 272, . , 97M; Delay amount detector, 282, . , 28. : Delay amount corrector, 29
2,・2gM; Delay error calculator, 302°, ,'
l' Oh+: Register, 31: Vector converter, 3
2; Register, 33; Inverse matrix Ao, utility, 342.・. 3” b1+ fq j@H error calculation instrumental gain 2I; register, 43; inner product calculator, 44; adder, 51, 2
．．・, 51i cod-kake genki, 52□7 + ” '+
J 2 , 'b+; register, 53□; accumulator, 6
21; Register, 6311, 63□2.63, 3; Inner product calculator.

Claims (1)

[Claims] Using a signal source whose position in space is unknown, an array consisting of multiple elements (in an array element position error measurement method that measures the deviation of an element from its rated position) a delay compensator that provides discrete time or phase delay compensation corresponding to a discrete value to the output signal of each element of the array; and a delay compensator that applies discrete time or phase delay compensation to the output signal of each element of the array; a delay detector for detecting a time or phase delay difference; an accumulator for accumulating the output signals of each of the delay compensators; an envelope detector for detecting an envelope of the output signal of the accumulator; and an output of the envelope detector. a first signal source position estimator that selects a discrete value for a position corresponding to temporal or phase delay compensation that maximizes , and said envelope detection in the vicinity of the discrete value selected by said first signal source position estimator. a second signal source position estimator that determines the signal source position that maximizes the output of the detector, the estimator having a second signal source position estimator that determines the signal source position that maximizes the output of the detector, the estimator having a second signal source position estimator that A method for measuring a position error of an array element, characterized in that the position error of each element of the array is determined based on an array constant and outputs of the first and second signal source position estimators.