JPH0138270B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION (Technical Field) The present invention receives a narrowband signal containing noise with an array consisting of a plurality of spatially arranged elements, and detects the phase difference between the output signals of two elements of the array. The present invention relates to a method for estimating the angle of incidence of a signal by estimating the angle of incidence of the signal onto an array.

(Background Art) A signal incidence angle measurement device that uses an array consisting of two elements and estimates the signal incidence angle with respect to a reference axis on the array by estimating the phase difference between the output signals of the two elements of the array. is widely used in sonar, acoustic positioning equipment or radar. As is well known, when the received signal is a narrowband signal, in order to avoid ambiguity due to growing lobes in estimating the angle of incidence of the signal, the arrangement spacing of the elements of the array should be adjusted to match the narrowband signal. There is a restriction that it must be less than 1/2 of the wavelength. On the other hand, since noise is generally added to the received signal, an estimation error occurs in the estimated value of the signal incidence angle. “Positioning Method”, Transactions of the Institute of Electronics and Communication Engineers, Vol.J65−
As shown in "A, No. 3", it is inversely proportional to the length of the arrangement interval of the elements. For these reasons, conventionally, taking both of the above into consideration, the spacing between the two elements of the array is set to 1/2 of the signal wavelength, or each element of the array is made to have directivity. A method has been used in which ambiguity due to side lobes is avoided by limiting the viewing angle, thereby increasing the arrangement interval of the elements to 1/2 or more.

Therefore, in such conventional methods, in order to widen the viewing angle of the array for estimating the signal incidence angle, the array spacing of the elements must be set to 1/2 or less of the signal wavelength. As the signal-to-noise ratio decreases, sufficient accuracy cannot be obtained, and when the array elements are arranged at intervals of 1/2 or more of the signal wavelength, the viewing angle of the array is restricted. There are drawbacks.

(Problems to be solved by the invention) In order to eliminate these drawbacks, the present invention provides an array in which two elements arranged at an interval of 1/2 or less of the signal wavelength are added with another element arranged at an interval longer than the interval. The purpose of this method is to simultaneously achieve a wide viewing angle and high estimation accuracy when estimating the signal incidence angle, and will be described in detail below.

(Structure and operation of the invention) FIG. 1 is an explanatory diagram of an array according to a first embodiment of the invention, in which 1 ₁ is the first element of the array, 1 ₂ is the second element of the array, and 1 ₃ is the second element of the array. is the third element of the array. In this example, the array is a linear array,
Each element 1 ₁ , 1 ₂ , 1 ₃ is arranged on the X axis, the spacing between elements 1 ₁ and 1 ₂ is d ₁ , and the spacing between elements 1 ₁ and 1 ₃ is d ₂ , and the narrowband signal source is sufficient. It's far away. The incident signal input from this narrow band signal source to the elements 1 ₁ , 1 ₂ , 1 ₃ has a center frequency f ₀ , a wavelength λ ₀ , and a period T ₀ (=
1/f ₀ ), which is a signal having an incident angle θ with respect to the X-axis. Letting the wavelength of the center frequency of the signal be λ ₀ ,
The spacing d ₁ and d ₂ between each element is d ₁ ≦λ ₀ /2, d ₂ >λ ₀ /
Selected as number 2.

FIG. 2 is a block diagram showing a first embodiment of the signal incidence angle estimation method of the present invention. 2 ₁ , 2 ₂ , 2 ₃
are amplifiers to which the output signals of elements 1 ₁ , 1 ₂ , and 1 ₃ are input, and 3 ₁ , 3 ₂ , and 3 ₃ are bandpass filters (BPF) to which the output signals of amplifiers 2 ₁ , 2 ₂ , and 2 ₃ are input. )
It is. 4 ₁ is a phase difference estimator (PHD) which inputs the output signals of band filters 3 ₁ and 3 ₂ and outputs an output value φ^ ₁ . 4 ₂ is a phase difference estimator (PHD) which inputs the signals of band filters 3 ₁ and 3 ₃ and outputs an output value φ^ ₂ . 5 ₁ is a multiplier (x) that multiplies the output value φ^ ₁ of the phase difference estimator 4 ₁ by a constant −(d ₂ /d ₁ 2πf ₀ ) to obtain a sparse estimated value τ〓 ₂ . 5 ₂ is a multiplier (x) that multiplies the output value φ^ ₂ of the phase difference estimator 4 ₂ by a constant -(1/2πf ₀ ) to obtain a time delay difference μ^ ₂ .

6 ₁ and 6 ₂ are registers (REG), which input constants 1/2πf ₀ and −(d ₂ /d ₁ 2πf ₀ ) to multipliers 5 ₁ and 5 ₂ , respectively. 7 is a variable delay time difference generator, which will be explained in detail below with reference to FIG.
Input the prediction error e(n) from the square calculator 10,
By varying the integer n, the delay time difference nT ₀ corresponding to the prediction error e(n) is successively determined and input to the adder 8, and at the same time, it is checked whether the prediction error e(n) shows the minimum value and this is the minimum value. When indicating a value, the delay time difference nT ₀ at that time is input to the adder 8 as the delay time difference n^T ₀ . 8 is an adder (Σ), and multiplier 5 ₂
, the time delay difference μ ₂ and the delay time difference nT ₀ (where n=
−[2d ₂ /λ]…, −1, 0, ±1,…, [2d ₂ /λ],
[2d ₂ /λ] is the largest positive integer not exceeding 2d ₂ /λ. ), and outputs a predicted value of delay time difference τ ₂ (n) which is the result of their addition. 9
is an adder (Σ), and the predicted value τ ₂ from adder 8 is
(n), the sparse estimated value τ ₂ is input from the multiplier 5 ₁ .
10 is a square calculator (SQR) which inputs the output signal of the adder 9 and outputs the prediction error e(n).

Reference numeral 11 denotes an angle converter, which inputs the predicted value τ ₂ (n^) from the multiplier 8 and outputs the estimated value θ^ of the incident angle θ, which is the angle conversion value, to the outside via an output terminal 12. ing. Each signal received by each element 1 ₁ , 1 ₂ , 1 ₃ of the array is transmitted to each of the amplifiers 2 ₁ , 2 ₂ , 2
₃ to an appropriate level, and after removing unnecessary band components by band filters 3 ₁ , 3 ₂ , 3 ₃ , the signals are input to the phase difference estimators 4 ₁ , 4 ₂ . In the phase difference estimator 4 ₁ , an estimated value φ^ ₁ of the phase difference (phase difference of the second element) φ ₁ between the output signal of the second element and the output signal of the first element (where |φ ^ ₁ |≦π), and the phase difference estimator 4 ₂ calculates the estimated value of the phase difference (phase difference of the third element) between the output signal of the third element and the output signal of the first element. Find φ^ ₂ (where |φ^ ₂ |≦π). Since the arrangement interval d ₁ between the first element 1 ₁ and the second element 1 ₂ is set to 1/2 or less of _the signal wavelength, the second element The estimated value of the time delay difference between the output signals of 1 _{and 2} can be obtained directly from the estimated value of the phase difference of the second element φ^ ₁ by −φ^ ₁ /2π ₀ (where ₀ is the center frequency of the signal). For the first
Since the arrangement interval d ₂ between the element 1 ₁ and the third element 1 ₃ is set to more than 1/2 of the signal wavelength, the output signal of the third element 1 ₃ with respect to the output signal of the first element 1 ₁ is The estimated value of the delay time difference of the output signals cannot be directly obtained from the estimated value φ^ ₂ of the phase difference of the third element.
However, the estimated time delay difference of the second element -
Using φ^ ₁ / 2π ₀ , the time delay difference of the third element is expressed as (d ₂ / d ₁ ) (−φ^ ₁ / 2π ₀ ) = −d ₂ φ^ ₁ / d ₁ 2π ₀ It is possible to estimate The register 6 ₁ stores a constant −d ₂ /d ₁ 2π ₀ , and the output value of the register 6 ₁ and the output value φ^ ₁ of the phase difference estimator 4 ₁ are multiplied by the multiplier 5 ₁ , and the A sparse estimate of the delay time difference of the third element τ〓 ₂ △ = −d ₂ φ^ ₁ /d ₁ 2πf ₀ is obtained. On the other hand, the register 6 ₂ stores a constant -1/2π ₀ , and the output of the register 6 ₂ and the output value φ^ ₂ of the phase difference estimator 4 ₂ are multiplied by the multiplier 5 ₂ to calculate the phase difference. A time delay difference μ^ ₂ △ = −φ^ ₂ /2π ₀ corresponding to φ^ ₂ (where |φ^ ₂ |≦π) is obtained. The variable delay time difference generator 7 is
Period of signal center frequency T ₀ △ = integer multiple of _1/0
nT ₀ ; n=0, ±1,..., ±[2d ₂ /λ ₀ ] (However, [2d ₂ /λ ₀ ] is the largest positive integer that does not exceed 2d ₂ /λ ₀ )
The adder 8 outputs the sum of the output nT ₀ of the variable delay time difference generator 7 and _{the output μ^2 of} the multiplier 52,
That is, the predicted value τ ₂ (n)△ = μ^ ₂ + nT ₀ of the delay time difference of the second element is calculated, and the adder 9 calculates the output τ ₂ (n) of the adder 8 and the multiplier 5 ₁ The difference between the output τ〓 ₂ and the output τ〓 ₂ (n
)−
τ〓 ₂ is calculated, and the square value calculator 10 calculates the square of the calculated value, that is, the prediction error e(n)Δ={τ ₂ (n)−τ〓 ₂ } ² . The variable delay time difference generator 7 is
A delay time difference value n^T ₀ corresponding to the integer n^ that minimizes the prediction error e(n) is output. Angle converter 11
is the predicted value of delay time difference τ ₂ (n^)△ = μ^ ₂ + n^T ₀ corresponding to the time difference value n^T ₀ , and the output of the third element with respect to the output signal of the first element 1 ₁ . As the true estimated value of the signal delay time difference, the estimated value τ ₂ (n^) is converted into the estimated value θ^ of the signal incidence angle θ as θ^=cos ^-1 {c・τ ₂ (n^)/d ₂ } The estimated value θ^ is output to the output terminal 12. FIG. 5 is a block diagram showing details of the variable delay time difference generator 7. In FIG. 5, 13 is a counter, which inputs a count m (=1, 2, . . . m) to the address generator 14.
Address generator 14 generates address Am for reading data from delay time difference memory 15. The delay time difference memory 15 outputs the delay time difference nT ₀ based on the address Am.

16 is the minimum prediction error Mine (m) and prediction error e
(n), 17 is a minimum value memory, and the delay time difference is stored in the delay time difference memory 15.
nT ₀ , the prediction error e(n) from the square multiplier 10 and the comparison result from the comparator 16 are input, and the minimum prediction error is
Output Mine(m) and the corresponding delay time difference n^T ₀ .

18 is an output switch which inputs the delay time difference n^T ₀ from the minimum value storage 17 and the delay time difference nT ₀ from the delay time difference storage 15, and outputs the delay time difference nT ₀ (finally n^T ₀ ). This is output and inputted to the adder 8.

In the operation of the variable delay time generator 7, the address generator 14 generates a predetermined address according to the integer m=1, 2,...m output from the counter 13.
Generate Am. The delay time difference memory 15 stores the delay time difference nT ₀ (where n=-[2d ₂ /λ]..., -1, 0,
±1, ..., [2d ₂ /λ], [2d ₂ /λ] are the largest positive integers not exceeding 2d ₂ /λ. ),
This is read by address Am. The read delay time difference nT ₀ is input to the adder 8 via the output switch 18.

The minimum value storage unit 17 stores the prediction error e(n) as the minimum prediction error Mine(m) and the delay time difference nT ₀ as the delay time difference n^T ₀ when m=1.

When m>2, the prediction error e(n) corresponding to the delay time difference nT ₀ stored at the address Am of the delay time difference storage 15 and the minimum prediction error Mine ( m) by the comparator 16, and if the minimum prediction error Mine (m) > prediction error e (n), this prediction error e (n) is set as the minimum prediction error Mine (m), and the delay time difference
The memory is updated with n^T ₀ as the delay time difference n^T ₀ .

By repeating this process for all m, the minimum prediction error Mine(m) and delay time difference n^T ₀ for all m will finally remain in the minimum value storage 17, so this minimum value delay time difference
n^T ₀ is output via the output switch 18.

As explained above, in the first embodiment of the present invention, the second
From the estimated value of the delay time _difference of the elements of A method that minimizes the square error between the estimated value of the phase difference φ^ ₂ and the predicted value τ ₂ (n) of the delay time difference of the third element, which can vary discretely in the T ₀ period obtained from the integer value n. Since the predicted value τ ₂ (n^) = −φ^ ₂ /2π ₀ + n^T ₀ corresponding to the integer n^ is the true estimate of the delay time difference of the first element, any value of the signal incidence angle If there is no error in estimating the integer value n^, the accuracy of estimating the angle of incidence of the signal is as follows: This method is equivalent to the estimation using the element, and therefore has the advantage that the disadvantages in terms of viewing angle and accuracy when estimating the signal incidence angle, which existed in the conventional method, can be improved.

FIG. 3 is an explanatory diagram of an array according to a second embodiment of the present invention, and 1 ₄ is the fourth element of the array.
In this example, the first element 1 ₁ and the second element 1 ₂ of the array are arranged on the X-axis with a spacing d ₁ , and the third element 1 ₃ and the fourth element 1 ₄ of the array are Assume that they are arranged at an interval d ₂ on the X' axis parallel to the X axis,
It is assumed that the array interval values d ₁ and d ₂ are taken in the same manner as in the first embodiment.

FIG. 4 is a structural diagram of the second embodiment of the present invention, where 2 ₄ is a fourth amplification diagram and 3 ₄ is a fourth bandpass filter. In the case of this embodiment, the phase difference estimator 4 ₁ is connected to the first element 1 ₁ as in the first embodiment.
The phase difference estimator 4 ₂ estimates the phase difference of the output signal of the second element 1 ₂ with respect to the output signal of the second element 1 2 .
is the fourth element for the output signal of the third element ₁₃
Estimate the phase difference between the output signals of elements 1 _{and 4} . If the estimated values of the respective phase differences are set as φ^ ₁ and φ^ ₂ , the output signal of the fourth element relative to the output signal of the third element can be calculated by using the same method as in the first embodiment. Since the true estimated value τ ₂ (n^) of the delay time difference of and the estimated value θ^ of the signal incidence angle can be determined, the second embodiment has the same effect as the first embodiment. can be obtained.

(Effects of the Invention) The present invention provides two
Since the estimated value of the signal incidence angle is obtained from the combination of the estimated value of the phase difference of the output signal of each element, the side It has the advantage that there is no ambiguity caused by lobes, and that it can obtain estimation accuracy equivalent to that obtained using elements with an arrangement spacing of 1/2 or more of the signal wavelength.Sonar, which requires high estimation accuracy of the signal incidence angle, It can be used in an acoustic measuring device or a radar signal incident angle measuring device.

[Brief explanation of drawings]

Fig. 1 is an explanatory diagram of an array showing a first embodiment of the present invention, Fig. 2 is a configuration diagram showing the first embodiment of the invention, and Fig. 3 shows a second embodiment of the invention. FIG. 4 is an explanatory diagram of the array, FIG. 4 is a block diagram showing a second embodiment of the present invention, and FIG. 5 is a block diagram of the variable time difference generator shown in FIG. 1 ₁ , 1 ₂ , 1 ₃ , 1 ₄ ; array element, 2 ₁ , 2 ₂ ,
2 ₃ , 2 ₄ ; Amplifier, 3 ₁ , 3 ₂ , 3 ₃ , 3 ₄ ; Bandpass filter, 4 ₁ , 4 ₂ : Phase difference estimator, 5 ₁ , 5 ₂ ; Multiplier, 6 ₁ , 6 ₂ ; Register , 7; variable delay time difference generator; 8; adder; 9; adder; 10; square calculator; 11; angle converter; 12; output terminal.

Claims (1)

[Claims] 1. A signal incidence angle estimation method for estimating the incidence angle of an incident signal from the phase difference between the output signals of a plurality of elements arranged in space, wherein the first, second, and third elements are The second element is arranged in a straight line, and the second element is arranged at a position that is less than or equal to λ/2 of the wavelength (λ) of the incident signal from the first element, and the third element is arranged in a position that is less than or equal to λ/2 of the wavelength (λ) of the incident signal. is placed at a position where the wavelength is λ/2 or more of the wavelength of the incident signal from ₁ ),
and an estimated value of a second phase difference (φ^ ₂ ) between the output signal of the first element and the output signal of the third element.
the estimated value of the first phase difference (φ^ ₁ ), the arrangement interval (d ₁ ) of the first and second elements, and the first
and a sparse estimation value (τ〓 ₂ ) of the delay time difference between the output signal of the second element and the output signal of the first element based on the arrangement interval (d ₂ ) of the third element. processing, the time delay difference (μ^ ₂ ) corresponding to the estimated value of the second phase difference (φ^ ₂ ), and the period ( _T0 ) of the center frequency of the incident signal as an integer n (where n= −[2d ₂ /λ]
..., -1, 0, ±1, ..., [2d ₂ /λ], [2d ₂ /λ]
is the largest positive integer not exceeding 2d ₂ /λ. ) and the predicted value (τ ₂ (n)) of the delay time difference of the second element by adding _the value (nT ₀ ) multiplied by a fourth process of calculating a prediction error (e(n)) between the predicted value of the delay time difference of the second element; and repeating the third and fourth processes by sequentially changing the integer. a fifth process for calculating the prediction error (e(n)) having the minimum value; and a fifth process for calculating the delay time difference corresponding to the prediction error (e(n)) having the minimum value by the fifth process. A second step that calculates an estimated value (θ^) of the angle of incidence of the incident signal from the predicted value and outputs the estimated value as a true estimated value of the delay time difference of the third element with respect to the output signal of the first element. 6. A method for estimating an angle of incidence of a signal, comprising the steps of 2. In a signal incidence angle estimation method for estimating the incidence angle of an incident signal from the phase difference between the output signals of a plurality of elements arranged in space, the first and second elements are set at 1/1 of the wavelength of the incident signal.
The third and fourth elements are arranged on the first straight line with an interval of 2 or less, and the third and fourth elements are parallel to the arrangement of the first and second elements, and the wavelength of the incident signal is 1/1/2.
an estimated value (φ^ ₁ ) of a first phase difference between the output signal of the first element and the output signal of the second element, arranged on a second straight line with an interval of 2 or more;
and an estimated value (φ^ ₂ ) of a second phase difference between the output signal of the third element and the output signal of the fourth element.
a first process for determining the first phase difference estimate (φ^ ₁ ), an arrangement interval (d ₁ ) between the first and second elements, and the third
and a sparse estimation value (τ〓 ₂ ) of the delay time difference between the output signal of the second element and the output signal of the first element based on the array interval (d ₂ ) of the fourth element. processing, the time delay difference (μ^ ₂ ) corresponding to the estimated value of the second phase difference (φ^ ₂ ), and the period ( _T0 ) of the center frequency of the incident signal as an integer n (where n= −[2d ₂ /λ]
..., -1, 0, ±1, ..., [2d ₂ /λ], [2d ₂ /λ]
is the largest positive integer not exceeding 2d ₂ /λ. ) to obtain a predicted value (τ ₂ ( _n )) of the delay time difference of the second element; and the sparse estimated value (τ〓 ₂ ); and a predicted value of the delay time difference of the second element. a fifth process to obtain the prediction error (e(n)) having the minimum value; and a fifth process to calculate the delay time difference corresponding to the prediction error (e(n)) having the minimum value by the fifth process. An estimated value (θ^) of the angle of incidence of the incident signal is obtained from the predicted value of , and the estimated value is output as a true estimated value of the delay time difference of the fourth element with respect to the output signal of the first element. A signal incidence angle estimation method comprising a sixth process.