JP2948459B2 - Angle measuring device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION The present invention relates to an antenna, a sonar,
The present invention relates to measurement of an arrival angle of an incoming wave using an array signal processing such as an optical detector, and measurement of a signal spectrum and a time delay.

[0002]

2. Description of the Related Art The resolution limit of an antenna has been defined as wavelength / antenna aperture diameter. In recent years, an algorithm has been proposed to obtain an angular resolution exceeding the resolution limit by using a plurality of antennas and signal processing. I have. As such an algorithm, for example, ROSchmidt: ■ Multiple E
mitter Location and Signal Parameter Estimation ■,
There is a MUSIC algorithm as shown in IEEE Trans., AP-34, 3, pp. 276-280 (1986). First, MUSI
A conventional angle measuring device using the C algorithm (hereinafter, referred to as a conventional angle measuring device) will be described with reference to FIGS. FIG. 1 is an overall configuration diagram of a conventional and the angle measuring apparatus of the present invention, wherein 1 is a receiving antenna, 2 is a receiver, 3 is an analog / digital converter, 4 is a signal processing device based on the MUSIC algorithm, and 5 is It is an angle display device. In the figure, x
_m indicates a reception signal, and m is a subscript indicating a channel number. Figure 2 is an internal configuration diagram of a conventional signal processing device 4 of FIG. 1, a buffer memory for the received signal x _m is 201, correlation matrix R calculation means 202, the eigenvalues and eigenvectors calculating means R is 203, 204 Direction evaluation function F
(Θ) calculation means 205 is a peak detection means.

The operation will be described below. The received signal vector X is defined by the following equation. X (i) = [x ₁ (i), x ₂ (i), ‥, x _M (i)] ^T where i is a time index, T is the transpose of a vector matrix, and M is the number of channels, ie, The number of antennas in the array.
The buffer memory 201 stores received signal vectors X (1) to X (I) from time 1 to I. Correlation matrix calculation means 20
2 calculates the correlation matrix R as follows. Here, H represents the complex conjugate transpose of the matrix vector. Therefore, R is an M × M square matrix. 203 is the eigenvalue λ of R
_{1 to} λ _M and eigenvectors e _{1 to} e _M are calculated. The azimuth evaluation function F calculating means 204 calculates the azimuth evaluation function F using the eigenvector e _N corresponding to the minimum eigenvalue among the eigenvalues λ _{1 to} λ _M as in the following equation. F (θ) = 1 / | a (θ) H e N | (2) where, indicates that (theta) is a function of the angle theta, F
(Θ) is calculated for all θ. a (θ) is called a steering vector, and a radio wave s incident at an angle of arrival θ
Is a coefficient vector such that the received signal vector X when one wave exists is given by the following equation.
It is determined by the directional characteristics of the direction. X = a (θ) s + n (3) where n is a noise vector, and noise n _m generated by the receiver 2 and A / D converter 3 of each channel or leaked from the antenna 1 as shown in the following equation. Is an element. The peak detecting means 405 searches for an angle θ that gives K peaks having the largest value of F (θ), and outputs this value as an estimated angle of arrival. The value of K is determined from the distribution of eigenvalues λ _{1 to} λ _M.

[0004] The principle of angle measurement of the MUSIC algorithm will be described below. It is assumed that the number K of incident waves is smaller than the total number M of channels (M> K). When a plurality of radio waves are incident on the array antenna, the received signal vector X is changed to Expression (3) and is given by the following expression. Here, k is a suffix indicating the number of the incident signal, and A is an M × K matrix composed of the steering vector a as shown in the following equation. S is a K × 1 vector given by the following equation with the incident signal as an element. s = [s ₁ , s ₂ , ‥, s _K ] ^T (7) Using the equation (5), the correlation matrix R defined in the equation (1) can be expanded as the following equation. R = ASA ^H + σ ² I (8) where S is a K × K square matrix configured as in the following equation. Here, σ ² is noise power, and I is an M × M unit matrix. Each noise n _m (m = 1, 2, ‥, M) was assumed to be uncorrelated and equal in power. Since the correlation matrix R is a Hermitian matrix, its eigenvalues λ _{1 to} λ _M are all positive real numbers, and it can be seen from Equation (8) that the smallest M−K eigenvalues are equal to the noise power σ ² . The eigenvector corresponding to such a minimum eigenvalue λ = σ ² is e
_When represented by _N , the following equation holds from the relationship between the eigenvalue eigenvectors. (R−σ ² I) e _N = 0 (10) If the left side of the above equation is transformed using equation (8), the following equation is obtained. _{^{ASA H e N = 0 (11}} ) where the matrix S is full rank, i.e. if the cross-correlation coefficient of the incident signal is less than 100%, (11) can be transformed to the following equation. _{^{A H e N = 0 (12}} ) which is expressed using the steering vector MUSIC
The following equation, which is important in principle of the algorithm, is obtained. On the other hand, when the azimuth evaluation function F (θ) is searched for the variable θ, if the variable θ matches any one of the radio wave incident angles θk, the denominator of the expression (2) becomes 0 from the expression (13), and The value is a very large value. FIG. 3 shows an example of the azimuth evaluation function F (θ) at this time. Therefore, the incidence angle can be estimated by the operation of the peak detection means 205.

However, in the angle measuring apparatus, when (element spacing d of the array antenna) / (wavelength λ of incident wave) is close to 1 or exceeds 1, the true arrival angle is included in the azimuth evaluation function F (θ). Spurs occur at an angle completely different from that of FIG. For example, considering a two-element linear array with an element spacing d, the steering vector a (θ) is given by a (θ) = [1, exp {j2πsin (θ) (d / λ)}] ^T , / Λ = 2, a (θ +
ξ that satisfies ξ) = a (θ) exists. Assuming now that a radio wave is incident from the θ _k direction, the azimuth evaluation function F (θ) has a peak at θ = θ _k and a peak also occurs at θ = θ _k + ξ. The latter peak is spurious, and the angle measurement result has an ambiguity or error of ξ. Similarly, even in the case of 0.5 << (d / λ) <1, there is a case where spurious is generated due to the existence of ξ which becomes a (θ + ξ) to a (θ) depending on the antenna arrangement. Further, since the number I of received signal data samples (see equation (1)) when calculating the correlation matrix R is finite, the actual noise contribution of R is shown in the second term on the right side of equation (8). It is not a perfect diagonal matrix like σ ² I. Then, an error occurs in the eigenvector e _N. This also promotes the generation of spurious, and the angle measurement error increases. Further, F (θ) calculated by the azimuth evaluation function calculating means 204 needs to be calculated for all θ in principle, but it is difficult to change θ continuously. The calculation is performed for each of the values θ. At this time, if the step width Δθ of θ is large (compared to the rate of change of the angle of the steering vector a (θ)), generally, the azimuth evaluation function F
Since the peak of (θ) is sharp, the peak detection unit 205 may not be able to detect this peak. On the other hand, as the step size of θ decreases, the change of θ approaches a continuous change, but the azimuth evaluation function calculating means 204 for calculating the azimuth evaluation function F (θ) for all θ, The amount of calculation in the peak detection means 205 that refers to the value of F (θ) increases significantly.

[0006]

The conventional angle measuring apparatus is configured as described above. When (element spacing of array antenna) / (wavelength of incident wave) is close to 1 or exceeds 1, There was a problem that spurious was generated in the azimuth evaluation function F (θ) and the incident angle of the incoming wave could not be accurately estimated. In addition, there is a problem that the amount of calculation required for angle measurement increases. SUMMARY OF THE INVENTION The present invention has been made to solve the above problems, and has an azimuth evaluation function F (θ).
It is an object of the present invention to obtain an angle measuring device capable of accurately estimating an incident angle of an incoming wave by suppressing spurious components generated in the signal. It is another object of the present invention to obtain an angle measuring device that requires a small amount of calculation for angle measurement.

[0007]

In order to achieve the above object, an angle measuring apparatus according to the first aspect of the present invention comprises: M receivers;
Correlation matrix calculating means for calculating the correlation matrices of some of the received signals of the M-channel received signals output from the receiver; estimating the eigenvalues and eigenvectors of the output matrix of the correlation matrix calculating means; An eigenvalue and eigenvector calculating means for extracting an eigenvector corresponding to an eigenvalue close to, and a vector having a phase delay or a directional characteristic in an angle θ direction of a receiver corresponding to the mth input of the correlation matrix calculating means as an mth element as a _{j (θ),} the orientation evaluation for outputting the inverse of the inner product calculated inner product of eigenvectors and above the eigenvalue and eigenvector computing means it has extracted every θ a _j (θ) as a direction evaluation function F _j (theta) and a function calculating unit, said a _j by switching the combination of the received signal channel to the input of the correlation matrix calculating means
(Θ) are different from each other, and the azimuth evaluation function F _j (θ) is obtained independently each time the combination of the received signal channels is switched. The function G (θ) obtained by the relational expression of
And a peak detecting means for searching for an peak corresponding to the peak and outputting an angle corresponding to the peak as an arrival angle of an incoming wave.

According to a second aspect of the present invention, the value of G (θ) at the angle θ when calculating the function G (θ) of the angle measuring apparatus according to the first aspect is a minimum value of F _j (θ). It is configured to give in.

According to a third aspect of the present invention, there is provided an angle measuring apparatus in which the value of G (θ) at an angle θ when calculating the function G (θ) of the angle measuring apparatus according to the first aspect is a median of F _j (θ). It is configured to give in.

According to a fourth aspect of the present invention, there is provided an angle measuring apparatus, comprising: M receivers; a correlation matrix calculating means for calculating a correlation matrix of an M-channel received signal which is an output of the receiver; An eigenvalue and an eigenvector calculating means for obtaining an eigenvalue and an eigenvector of an output matrix of the means and extracting an eigenvector corresponding to a minimum or eigenvalue close to the minimum; and an angle θ direction of a receiver corresponding to an m-th input of the correlation matrix calculating means. Assuming that a vector having the phase delay and the directional characteristic as the m-th element is a (θ), an inner product of the eigenvalue and the eigenvector extracted by the eigenvector calculating means and the above a (θ) is obtained for each θ, and the reciprocal of the inner product is evaluated for azimuth. Azimuth evaluation function calculating means for outputting as a function F _j (θ), wherein the correlation matrix calculating means switches the combination of the corresponding times of the received signals used for the correlation calculation. In this way, J independent azimuth evaluation functions F _j (θ) whose function forms are different from each other are obtained, The function H (θ) obtained by the relational expression is obtained, and the function H (θ) is obtained.
And a peak detection means for outputting an angle corresponding to this peak as an incident angle of an incoming wave.

According to a fifth aspect of the present invention, the value of G (θ) at the angle θ when obtaining the function G (θ) of the angle measuring device according to the fourth aspect is the minimum value of F _j (θ). It is configured to give in.

According to a sixth aspect of the present invention, there is provided an angle measuring apparatus in which the value of G (θ) at an angle θ when calculating the function G (θ) of the angle measuring apparatus according to the fourth aspect is a median of F _j (θ). It is configured to give in.

According to a seventh aspect of the present invention, there is provided an angle measuring apparatus comprising: M receivers; correlation matrix calculating means for calculating a correlation matrix of an M-channel received signal which is an output of the receiver; An eigenvalue and an eigenvector calculating means for obtaining an eigenvalue and an eigenvector of an output matrix of the means and extracting an eigenvector corresponding to a minimum or eigenvalue close to the minimum; and an angle θ direction of a receiver corresponding to an m-th input of the correlation matrix calculating means. Assuming that a vector having the phase delay and the directional characteristic as the m-th element is a (θ), an inner product of the eigenvalue and the eigenvector extracted by the eigenvector calculating means and the above a (θ) is obtained for each θ, and the reciprocal of the inner product is evaluated for azimuth. Azimuth evaluation function calculating means for outputting as a function F (θ), and a peak for searching for the peak of F (θ) and outputting an angle corresponding to the peak as an incident angle of an incoming wave Means out, and a crude angle measuring means for coarse angle measuring an angle of arrival of the input arrival wave signals received M-channel is the wave receiver outputs, as the angle measurement results of the coarse angle measuring means theta _est The azimuth evaluation function calculating means calculates
F (θ) is calculated only in the angle range near _est , and the peak detection means calculates F (θ) only in the angle range near θ _est.
It is configured to search for the peak of (θ).

According to an eighth aspect of the present invention, there is provided an angle measuring apparatus comprising: M receivers; a correlation matrix calculating means for calculating a correlation matrix of an M-channel received signal which is an output of the receiver; An eigenvalue and an eigenvector calculating means for obtaining an eigenvalue and an eigenvector of an output matrix of the means and extracting an eigenvector corresponding to a minimum or eigenvalue close to the minimum; and an angle θ direction of a receiver corresponding to an m-th input of the correlation matrix calculating means. Assuming that a vector having the phase delay and the directional characteristic as the m-th element is a (θ), an inner product of the eigenvalue and the eigenvector extracted by the eigenvector calculating means and the above a (θ) is obtained for each θ, and the reciprocal of the inner product is evaluated for azimuth. Azimuth evaluation function calculating means for outputting as a function F (θ), peak detecting means for searching for the peak of F (θ) and outputting an angle corresponding to the peak, and the receiver And a coarse angle measuring means for coarse angle measuring an angle of arrival of the input arrival wave signals received M channels is a force, as the angle measurement results theta _est of the crude angle measuring means, and outputs the above peak detecting means An angle closest to _θest among a plurality of angles corresponding to a plurality of peaks of F (θ) is output as an incident angle of an incoming wave.

[0015]

In the angle measuring apparatus according to the first, second, and third aspects of the present invention, the a _j (θ) is different from each other by switching the combination of the received signal channels input to the correlation matrix calculating means. Place the above receiver so that
An angle function given by the average, minimum, or median of the azimuth evaluation functions F _j (θ) independently obtained each time the combination of received signal channels is switched, and the angle corresponding to the peak of this angle function is determined by the incidence of the incoming wave. By setting the angle, the spurious of the azimuth evaluation function can be suppressed, and the incident angle of the incoming wave can be more accurately estimated.

Further, in the angle measuring apparatus according to the fourth, fifth and sixth aspects of the present invention, the azimuth evaluation function F _j () obtained independently by the correlation matrix calculation means every time the combination of the corresponding times of the received signals used for the correlation calculation is switched. θ), the angle function given by the average, the minimum value, or the median value is obtained, and the angle corresponding to the peak of this angle function is used as the incident angle of the arriving wave. The incident angle of the arriving wave can be estimated.

Further, in the angle measuring device of the invention of claim 7, the angle measurement results of the coarse angle measuring means as theta _est, orientation evaluation function calculating unit is limited to an angle range in the vicinity of the theta _est orientation evaluation function F ( θ) is calculated, and the peak detection means is configured to search for the peak of F (θ) only in the angle range near θ _est , thereby reducing the amount of calculation required for angle measurement. Also, suppress the spurious of the azimuth evaluation function,
It is possible to more accurately estimate the incident angle of the incoming wave.

Further, in the angle measuring device of the invention of claim 8, the angle measurement results of the coarse angle measuring means as theta _est, orientation evaluation function F
Among the plurality of angles corresponding to the plurality of peaks of (θ), the angle closest to θ _est is configured to be output as the incident angle of the arriving wave. The incident angle of the arriving wave can be estimated.

[0019]

【Example】

Embodiment 1 FIG. First Embodiment A first embodiment of the present invention will be described with reference to FIGS. FIG. 1 is an overall configuration diagram of the first embodiment, which is the same as the conventional example except for the signal processing device 4, and a description thereof will be omitted. However, the antenna elements 1 are arranged so as not to form a linear array. FIG. 4 is a diagram illustrating an internal configuration of the signal processing device 4 according to the first embodiment.
Buffer memory 40 for _L received signals x _{1 to} x _L
1b denotes received signals x _{L + 1} to antennas # (L + 1) to #M.
buffer for x _M memory, 202a, 202b are correlation matrix R calculation means, 203a, 203b are eigenvalues and eigenvectors calculating means, 204a, 204b azimuth evaluation function calculating unit, the fusion orientation evaluation function calculating unit 406, 20
5 is a peak calculating means. Each buffer memory 401
The numbers of received signal channels input to the a and 401b are L and (ML), respectively. The number of channels is set so that L> K and (ML)> K with respect to the number of incident radio waves K. Set. Hereinafter, the operation will be described. In FIG. 4, 202a to 204a,
Reference numerals 202b to 204b denote the 2nd of the conventional example shown in FIG.
The description is omitted because it is the same as 02 to 204. Assuming that the azimuth evaluation functions calculated by the two azimuth evaluation function calculation means 204a and 204b are F ₁ (θ) and F ₂ (θ),
F ₁ (θ) is an antenna # 1 to # received signal x ₁ ~x _L of L
Azimuth evaluation function determined from the data of antenna #
(L + 1) is the azimuthal evaluation function determined given from the received signal x _{L + 1} ~x _M data ~ # M. The fusion azimuth evaluation function 406 calculates the fusion azimuth evaluation function G (θ) as follows for all angles θ. The peak detecting means 205 searches for the K peaks having the largest fusion azimuth evaluation function G (θ), and outputs an estimated angle of arrival. Here, K is the number of incident radio waves and is determined from the distribution of the eigenvalues and the eigenvalues calculated by the eigenvector calculation means 203. The incident wave number K ₁ determined by the eigenvalues and eigenvectors calculating means 203a, the incident wave number K ₂ are determined by the eigenvalues and eigenvectors calculating unit 203b is a principle same value. However, when the values of K ₁ and K ₂ are actually different, the smaller value of the two is set as the number K of incident radio waves.

When (array distance d of array antenna) / (wavelength λ of incident wave) is close to 1 or exceeds 1, as shown in FIG. 5A, the azimuth evaluation function F ₁ (θ)
Spurious is generated in a completely different angle than the true of the arrival angle θ _t in. The angle at which one spurious component of F ₁ (θ) occurs is θ _s . The reason why spurious is generated at this angle is that the spurious generated angle θ _s
This is because the steering vector a ₁ (θ _s ) becomes the same value as or close to the steering vector a ₁ (θ _t ) at the true arrival angle θ _t . That is, a ₁ (θ _s ) = a ₁ (θ _t ) (102). Note that the steering vector a
₁ (θ) is a function of θ determined by the positions and directivities of the antennas # 1 to #L. Expressed the eigenvector corresponding to the minimum eigenvalue of the eigenvalue and eigenvector calculation unit 203a calculates at e _N1, orientation evaluation function F ₁ (θ) _{is, F 1 (θ) = 1} / | a 1 (θ) H e N1 | (103), and F ₁ (θ _s ) =
F ₁ (θ _t ). a ₁ (θ _t ) is orthogonal to e _N1 and F
_As the value of ₁ (θ _t ) increases, the value of F ₁ (θ _s ) also increases, causing a spur in θ _s . As shown in FIG. 5B, spurs also occur in the azimuth evaluation function F ₂ (θ). In general, however, spurious generation angle of F ₂ is different from the spurious generation angle theta _s of F _1. To explain the reason, the spurious generation angle θ _{s of} F ₁
And F ₂ (θ _s ) is considered. Direction evaluation function F
₂ (theta) _{is, F 2 (θ) = 1} / | a 2 (θ) H e N2 | given by _{(104), a 2 (θ} t) is perpendicular to the e _N2 F
_{Even if 2} (θ _t ) becomes a large value, generally, a ₂ (θ _s ) ≠ a ₂ (θ _t ) (105), so that F ₂ (θ _s ) does not become a large value and does not become spurious. . Equation (105) even if equation (102) holds
The reason why the equation does not hold is that the steering vector a ₁
Array of antennas # 1 to #L for determining (θ) and a
_Since the arrangement of the antennas # (L + 1) to #M for determining ₂ (θ) is different, the function forms of a ₁ (θ) and a ₂ (θ) are different from each other. Fusion direction evaluation function G
In (θ), the fact that the spurious is suppressed as shown in FIG. 5C will be described below. Equation (101) that gives G (θ) is modified as follows using Equations (103) and (104). In the true arrival angle _{θ t, a 1 (θ)} H e N1, a 2 (θ) H
Since e _N2 both a small value, the fusion orientation evaluation function at θ _t G (θ _t) is the peak takes a large value. On the other hand, at the spurious occurrence angle θ _s of F ₁ , a ₁ (θ) ^H
Even if e _N1 is small, as described above, a
_{2 (θ)} ^H e _N2 does not take a small value. Therefore, equation (1)
06) is not so small, so G
(Θ _s ) does not take a large value, and the spurious at θ _s is suppressed. Spurious components at angles other than θ _s are similarly suppressed, and the fusion azimuth evaluation function G (θ _t ) forms a peak only at the true arrival angle θ _t as shown in FIG.

Azimuth evaluation function F in the conventional angle measuring device
Since (θ) has a functional form as shown in FIG. 5A or FIG. 5B, the radio wave arrival angle is estimated from the spurious occurrence angle,
It often happens that the displayed angle of arrival estimates differ significantly from the true angle of arrival. On the other hand, the angle measuring apparatus of the present invention estimates the arrival angle from the peak of the fusion azimuth evaluation function G (θ) in which spurious is suppressed. The display as angles is extremely reduced, and the angle of arrival can be estimated more accurately than in the conventional example. In the above description of the embodiment, the case where the number of arriving radio waves is 1 has been described, but there is no problem even if the number of arriving radio waves is plural. Although the case where the antenna element 1 is arranged so as not to form a linear array has been described, even when the antenna element 1 is arranged in a linear array, the function forms of the steering vectors a ₁ (θ) and a ₂ (θ) are different from each other. There is no problem if an antenna is arranged or the buffer memory 401 is connected to the antenna reception signal.

In the above embodiment, the buffer memory 401a
Are connected to the reception signals of antennas # 1 to #L, and the reception signals of antennas # (L + 1) to #M are connected to buffer memory 401b. For example, antennas # (L-10) to ## are connected to buffer memory 401b. Similar effects can be obtained even when the reception signal channels are partially shared between the two buffer memories and connected as in the case of connecting the reception signals of M. In the above embodiment, the buffer memory 401,
Correlation matrix calculation means 202, eigenvalue and eigenvector calculation means 203, and azimuth evaluation function F (θ) calculation means 204
A set is prepared, two azimuth evaluation functions F ₁ and F ₂ are calculated, and a fusion azimuth evaluation function G (θ) is calculated from equation (101).
2. J sets of eigenvalue and eigenvector calculation means 203 and azimuth evaluation function F (θ) calculation means 204 are prepared (J>
3) Even if the J azimuth evaluation functions F _{1 to} F _J are calculated and the arrival angle is estimated using the fusion azimuth evaluation function G (θ) shown in the following equation, the same effect can be obtained. Further, in the above embodiment, the angle measuring device for measuring the angle of arrival of a radio wave has been described as an example, but the present invention can be applied to the case of measuring the angle of arrival of a sound wave or a light wave.

Embodiment 2 FIG. Second Embodiment A second embodiment of the present invention will be described with reference to FIGS. FIG. 1 shows the second embodiment.
3 is the same as the conventional example except for the signal processing device 4, and the description is omitted. FIG. 4 is a diagram showing the internal configuration of the signal processing device 4 according to the second embodiment.
06 sets the fusion orientation evaluation function G (θ) to all angles θ.
Is calculated as follows. Here, min [*] means the minimum value of the element *. F ₁ (θ) and F ₂ (θ) are antennas # 1 to # 1, respectively.
Received signal x ₁ ~x _L orientation evaluation function determined from the data of the #L, antenna # (L + 1) ~ # M received signal x
_{L + 1} is the azimuthal evaluation function determined given from ~x _M data. The other units in FIG. 4 are the same as those in the first embodiment, and thus the description will be omitted.

The azimuth evaluation functions F ₁ (θ) and F ₂ (θ) are
As shown in FIGS. 5A and 5B, spurs may occur. As described in the first embodiment, F ₁ ,
Since the spurious generation angle on F ₂ are different from each other, spurious is suppressed by the calculation of equation (201) in the fusion orientation evaluation function calculating unit 406. On the other hand, the true angle of arrival theta _t, since the F _1, F ₂ are both peaks, G
(Θ _t ) also peaks. Therefore, the fusion orientation evaluation function G
(Theta) is the result of forming the peak only at the true arrival angle theta _t as shown in FIG. 5 (C). As described above, also in the angle measuring apparatus of the present invention, the fusion azimuth evaluation function G in which spurious is suppressed
Since the arrival angle is estimated from the peak of (θ), the occurrence angle of the spurious of the azimuth evaluation function is erroneously displayed as the arrival angle unlike the conventional example, and the arrival angle is estimated more accurately than the conventional example. it can.

Embodiment 3 FIG. A third embodiment according to claim 3 will be described with reference to FIGS. FIG. 1 is an overall configuration diagram of a third embodiment, which is the same as the conventional example except for the signal processing device 4, and a description thereof will be omitted. FIG. 6 is a diagram showing the internal configuration of the signal processing device 4 according to the third embodiment.
Calculates the fusion azimuth evaluation function G (θ) for all angles θ as in the following equation. Here, med [*] means the median value in the element *. F ₁ (θ), F ₂ (θ), ‥, and F _J (θ) are the received signals of antennas # 1 to # L1 and the antenna # (L1
+1) to # L2, antenna # (LJ + 1) to
This is an azimuth evaluation function determined from the received signal data of #M. The other units in FIG. 6 are the same as those in the first embodiment, and thus description thereof is omitted.

The azimuth evaluation functions F ₁ (θ), F ₂ (θ),
‥, F _J (θ) may cause spurious as shown in FIGS. 5 (A) and 5 (B). As described in the first embodiment, since the spurious occurrence angles on F ₁ and F ₂ are different from each other, the spurious is suppressed by the calculation of Expression (301) in the fusion azimuth evaluation function calculation means 606. Now, let J = 5 and let _one of the spurious occurrence angles of F _{1 be} θs. For example, when J = 5, the values of the azimuth evaluation functions F _{1 to} F ₅ at this angle are: F ₁ (θ _s ) = 60 dB F ₂ (θ _s ) = 5 dB F ₃ (θ _s ) = 2 dB F ₄ ( θ _s ) = 6 dB It is assumed that F ₅ (θ _s ) = 5 dB is calculated. At this time, since the fusion azimuth evaluation function G (θ _s ) at this angle θ _s is the median of F _{1 to} F ₅ , G (θ _s ) = 5 dB is calculated and does not become a large value. On the other hand, the true angle of arrival theta _t, since the F ₁ to F ₅ together high value remains median is high value. That is, G (θ _t ) also has a peak. Therefore, the fusion azimuth evaluation function G (θ) becomes the true arrival angle θ _{t as} shown in FIG.
Only to form a peak. As described above, even in the angle measuring apparatus of the present invention, since the arrival angle is estimated from the peak of the fusion azimuth evaluation function G (θ) in which the spurious is suppressed, the spurious occurrence angle of the azimuth evaluation function erroneously arrives as in the conventional example. The display as angles is extremely reduced, and the angle of arrival can be estimated more accurately than in the conventional example.

Embodiment 4 FIG. Fourth Embodiment A fourth embodiment of the present invention will be described with reference to FIGS. FIG. 1 shows the fourth embodiment.
Is the same as the conventional one except for the signal processing device 4, and the description is omitted. FIG. 7 shows a signal processing device 4 according to the fourth embodiment.
In a diagram showing the internal structure, 701 denotes an antenna #. 1 to # M of the received signal x ₁ ~x _M a connected buffer memory, 70
2a to 702z are correlation matrix R calculating means;
3z is an eigenvalue and eigenvector calculating means of R, 204a
204z is an azimuth evaluation function calculation means, 706 is a fusion azimuth evaluation function calculation means, and 205 is a peak calculation means. R
_{1 to} R _J are correlation matrix calculation means 702a to 702, respectively.
2z indicates a correlation matrix calculated, F _{1 to} F _J indicate azimuth evaluation functions calculated by the respective azimuth evaluation function calculation means 204a to 204z, and H indicates a fusion azimuth evaluation function calculation means 7
6 shows a fusion azimuth evaluation function calculated in step S06. In addition, i is a subscript indicating discrete time, θ indicates an angle, and θ _{1 to} θ _K are symbols indicating the angles of arrival of radio waves. The number L _{j of} reception signal channels input to the buffer memory 701 is set so as to satisfy L _j > K (j = 1, ‥, J) with respect to the number K of incident radio waves.

The operation will be described below. The buffer memory 701 stores a vector X (i) having received signals of all input channels as elements from time 1 to I as in the conventional example.
(1) to X (I) are stored. Correlation matrix calculation means 702a
Are reception signals X (1) to X (I / I) from time 1 to I / J.
Using J), the correlation matrix R ₁ is calculated from the following equation. The correlation matrix calculation means 702b calculates the received signal X (I / J + 1)
ＸX (2I / J) is used to calculate the correlation matrix R ₂ from the following equation. Correlation matrix calculating unit 702z also has only different times of the received signals, calculates a correlation matrix R _J by similar calculation. FIG. 8A shows the correspondence between the received signal data and the correlation matrix R of the conventional example, and FIG. 8B shows the correspondence between the received signal data and the correlation matrices R _{1 to} R _J in the present invention. It is a thing. In the figure, a black point indicates received data, and a black point interval Δt
Means the sampling period of the analog / digital converter 3. The horizontal axis corresponds to the time axis. 203a-2
03z and 204a to 204z are 2 of the conventional example shown in FIG.
The description is omitted because it is the same as 03,204. The azimuth evaluation functions F ₁ (θ) to F _J (θ) calculated by the plurality of azimuth evaluation function calculation means 204 a to 204 z are azimuth evaluation functions determined from the correlation matrices R _{1 to} R _J , respectively.
The fusion azimuth evaluation function calculating means 706 calculates the fusion azimuth evaluation function H
(Θ) is calculated for all angles θ as follows: The peak detecting means 205 searches for the K peaks having the largest fusion azimuth evaluation function G (θ), and outputs an estimated angle of arrival. Here, the number K of incident radio waves is determined from the distribution of the eigenvalues and the eigenvalues calculated by the eigenvector calculation means 203a to 203z. Each eigenvalue and eigenvector calculation means 20
3a to 203z, the number of incident radio waves K _j (j = 1,
If the values of (‥, J) are different, the minimum K _{j is set} to the number K of incident radio waves.

As described in the conventional example, it is a condition for the MUSIC algorithm to be established that the correlation matrix R is expanded as shown in equation (8). However, since the observation time is not actually infinite and the number I of input signal data that can be used for averaging the correlation operation is finite, each noise sequence _nm (m = 1,
2, ‥, M) are not completely uncorrelated with each other and are not completely equal power. That is, the actually calculated correlation matrix R is given by the following equation. R = R ₀ + R _d (704) R ₀ is a true value when the observation time is infinite, and R _d is an M × M matrix indicating an error of R. Then, the eigenvector e _N corresponding to the minimum eigenvalue also includes an error, and thus the azimuth evaluation function F
(Θ) also causes an error, and the peak angle of F (θ) deviates from the true arrival angle of the radio wave, or as shown in FIG. 5 (A), the peak (spurious) at an angle completely different from the true arrival angle. May occur. The angle measuring apparatus of the present invention divides I pieces of observation data into J pieces as described above to obtain J pieces of correlation matrices R _j (j = 1, 2, ‥, J). If R _j is expressed as in equation (704), then R _j = R ₀ + R _dj (705). Each azimuth evaluation function F calculated by the azimuth evaluation function calculation means 204a to 204z under the influence of the error matrix R _dj
j (θ) may cause spurious as shown in FIG. 5A as in the conventional example. However, the spurious occurrence angle differs for each F _j (θ). That is, the noise sequence in the received signal is different for each j, and therefore the equation (70)
This is because the error matrix _{Rdj in} 5) differs for each j. Therefore, the fusion azimuth evaluation function calculation means 706 calculates the expression (703)
Example 1 by the calculation of the fusion evaluation function H (θ) shown in FIG.
As described above, the spurious and the like are suppressed. Since the peak detection means 205 estimates the angle of arrival from the peak of the fusion evaluation function H (θ) in which the spurious is suppressed, the spurious occurrence angle of the azimuth evaluation function is erroneously displayed as the angle of arrival as in the conventional example.

In the above embodiment, the correlation matrix calculating means 702
a is the reception signals X (1) to X (I) from time 1 to I / J.
/ J) is used to calculate the correlation matrix R ₁ , and the correlation matrix calculation means 702b receives the received signals X (I / J + 1) to X (2I /
Although the correlation matrix R2 is calculated using J), the received signal used by the correlation matrix calculation means 702b is represented by X (I / J-1).
0) to X (2I / J), a plurality of correlation matrix calculating means partially share received signal data,
Similar effects can be obtained even if the correlation matrix is calculated with the number of received signal data slightly different from each other. Further, in the above embodiment, the correlation matrix calculating means 702 a
Until the received signal X (1) ~X (I / J) as to calculate the correlation matrix R ₁ with, but using the received signal data arranged in the order, the order data may even discrete Absent.

Embodiment 5 FIG. A fifth embodiment of the present invention will be described with reference to FIGS. FIG. 1 shows the fifth embodiment.
3 is the same as the conventional example except for the signal processing device 4, and the description is omitted. FIG. 7 is a diagram showing the internal configuration of the signal processing device 4 according to the fifth embodiment.
06 sets the fusion azimuth evaluation function H (θ) to all angles θ
Is calculated as follows. Here, min [*] means the minimum value of the element *. F _j (θ) is a bearing evaluation function determined from the correlation matrix R _j . R _j is the received signal X ((j−1) I / J +
1) is a correlation matrix determined from X (jI / J).
The other units in FIG. 7 are the same as those in the fourth embodiment, and thus the description is omitted. Since the spurious occurrence angles on F _j are different from each other, the spurious is suppressed by the calculation of Expression (501) in the fusion azimuth evaluation function calculating means 706. As described above, even in the angle measuring apparatus of the present invention, since the arrival angle is estimated from the peak of the fusion azimuth evaluation function H (θ) in which spurious is suppressed, the spurious occurrence angle of the azimuth evaluation function is erroneously arrived as in the conventional example. The display as angles is extremely reduced, and the angle of arrival can be estimated more accurately than in the conventional example.

Embodiment 6 FIG. A sixth embodiment of the invention according to claim 6 will be described with reference to FIGS. FIG. 1 shows the sixth embodiment.
3 is the same as the conventional example except for the signal processing device 4, and the description is omitted. FIG. 7 is a diagram showing the internal configuration of the signal processing device 4 according to the sixth embodiment.
06 sets the fusion azimuth evaluation function H (θ) to all angles θ
Is calculated as follows. Here, med [*] means the median value in the element *. The other units in FIG. 7 are the same as those in the fourth embodiment, and thus the description is omitted. As described in the fourth embodiment, F ₁ ,
Since the spurious generation angle on F ₂ are different from each other, spurious is suppressed by the calculation of equation (601) in the fusion orientation evaluation function calculating unit 706. A specific example of the calculation of Expression (601) has been described in the description of the third embodiment. As described above, even in the angle measuring apparatus of the present invention, since the arrival angle is estimated from the peak of the fusion azimuth evaluation function H (θ) in which spurious is suppressed, the spurious occurrence angle of the azimuth evaluation function is erroneously arrived as in the conventional example. The display as angles is extremely reduced, and the angle of arrival can be estimated more accurately than in the conventional example.

Embodiment 7 FIG. A seventh embodiment of the present invention according to claim 7 will be described with reference to FIGS. FIG. 1 shows the seventh embodiment.
3 is the same as the conventional example except for the signal processing device 4, and the description is omitted. FIG. 9 is a diagram showing the internal configuration of the signal processing device 4 according to the seventh embodiment. Reference numeral 910 denotes a monopulse angle measurement processing unit for inputting a reception signal x _m (m = 1, ‥, M) of each antenna, and reference numeral 911 denotes a search. It is a range determining means. θ ₁ indicates a radio wave arrival angle true value, θ _est indicates an estimated angle of the monopulse angle measuring unit 910, and θ _min and θ _max indicate lower and upper limits of the search angle range determined by the search range determining unit, respectively. It is.

The operation will be described below. In FIG. 9, reference numerals 201 to 205 are the same as in the conventional example. The monopulse angle measurement processing means 910 multiplies the received signal x _m (m = 1, ‥, M) by an appropriate load to generate a sum beam and a difference beam and performs monopulse angle measurement, thereby obtaining a coarse angle measurement result θ
Output _est . The search range determining means 911 determines θ _min and θ _max as follows, for example. The azimuth evaluation function calculating means 204 calculates the azimuth evaluation function F (θ) from the equation (2), as in the conventional example, but refers to the output of the search range determining means 911 and determines only the angle θ range given by the following equation. F (θ) is calculated. θ _min <θ <θ _max (702) The peak detection means 205 is the same as the conventional example, but the peak search range of the azimuth evaluation function F (θ) is, as shown in FIG. Squeeze. In FIG.
The dotted line is the azimuth evaluation function F (θ) obtained by the azimuth evaluation function calculation means 204 of the conventional example, and the solid line is F (θ) obtained by the azimuth evaluation function calculation means 204 of the present invention. As described above, in the angle measuring apparatus of the present invention, the azimuth evaluation function F (θ), the calculation of the steering vector a (θ), and the search angle range of F (θ) are limited to the vicinity of the true arrival angle. The amount of calculation is significantly reduced as compared with. Also, since the probability of searching for the spurious occurrence angle of the azimuth evaluation function is reduced, the display of the spurious occurrence angle as the angle of arrival by mistake is reduced,
The angle of arrival can be estimated more accurately than in the conventional example. In the above embodiment, the coarse angle measurement is performed by the monopulse angle measurement processing means 910, but other coarse angle measurement means may be used, and similar results can be obtained by using the analog coarse angle measurement means. Can be. Further, the monopulse angle measurement processing means 910 need not be connected to all reception channels. Also, the azimuth angle θ
In addition, when searching for the elevation angle φ, the same effect can be obtained by calculating φ _min and φ _max in the same manner as described above.

Embodiment 8 FIG. An eighth embodiment of the present invention will be described with reference to FIGS. FIG. 1 is an overall configuration diagram of the eighth embodiment. The configuration other than the signal processing device 4 is the same as that of the conventional example, and the description is omitted. FIG. 11 is a diagram showing the internal configuration of the signal processing device 4 according to the eighth embodiment. Reference numeral 206 denotes an angle matching unit. θ _{1 to} θ _K are the output results of the peak detection means 205 and indicate the angle estimation values obtained by the K MUSIC algorithms, θ _est indicates the estimated angle of the monopulse angle measurement processing means 910, and θ _out is determined by the angle collation means. Is a symbol indicating the estimated angle measurement value to be measured.

The operation will be described below. 11, reference numerals 201 to 205 are the same as those in the conventional example. The monopulse angle measurement processing unit 910 performs the same operation as in the seventh embodiment,
Output _θest as a result of the coarse angle measurement. Angle collating means 20
6 is θ ₁ to θ _{K which} is closest to θ _est.
Output as _out . For example, suppose that K = 4 and the azimuth evaluation function F (θ) and its peaks θ _{1 to} θ ₄ and θ _est are given as shown in FIG. θ ₁ , θ ₂ , and θ ₄ are spurious generation angles. The angle matching means 206 outputs θ ₃ closest to θ _est to θ _out
Output as As described above, in the angle measuring device of the present invention, the most likely angle estimation value is selected and displayed from the angle estimation values obtained by the MUSIC algorithm with reference to the coarse angle measurement result. Thus, the display of the spurious occurrence angle as the angle of arrival is extremely reduced, and the angle of arrival can be estimated more accurately than in the conventional example. In the above embodiment, the coarse angle measurement is performed by the monopulse angle measurement processing unit 910.
Similar results can be obtained by using other coarse angle measuring means, or by using analog coarse angle measuring means. Further, the monopulse angle measurement processing means 910 need not be connected to all reception channels. Further, even when searching not only for the azimuth angle θ but also for the elevation angle φ, the pair of the closest peak angles (θ _K , φ _K ) is output to (θ _est , φ _est ) output from the angle collation means 206. Thus, a similar effect can be obtained.

[0037]

As described above, according to the first, second, and third aspects of the present invention, the steering vector a _j (θ) differs from each other by switching the combination of the reception signal channels input to the correlation matrix calculation means. The receiver is arranged as described above, and each time the combination of the received signal channels is switched, an angle function given by either the average, the minimum, or the median of the azimuth evaluation functions F _j (θ) obtained independently is obtained.
By setting the angle corresponding to the peak of the angle function as the incident angle of the incoming wave, it is possible to suppress the spurious of the azimuth evaluation function and obtain an angle measuring device capable of more accurately estimating the incident angle of the incoming wave than before. .

According to the present invention, the azimuth evaluation function F _j (θ) obtained independently by the correlation matrix calculating means every time the combination of the corresponding times of the received signals used for the correlation calculation is switched. ) Is determined as an average, minimum, or median value, and the angle corresponding to the peak of the angle function is defined as the incident angle of the incoming wave, thereby suppressing the spurious of the azimuth evaluation function. It is possible to obtain an angle measuring device that can more accurately estimate the incident angle of an incoming wave than before.

Further, according to the invention according to claim 7, the angle measurement results of the coarse angle measuring means as theta _est, orientation evaluation function calculating unit is limited to an angle range in the vicinity of the theta _est orientation evaluation function F (theta ) Is calculated, and the peak detecting means is configured to search for the peak of F (θ) only in the angle range near θ _est , thereby obtaining an angle measuring device with a reduced amount of calculation required for angle measurement. it can. Further, it is possible to obtain an angle measuring device capable of suppressing the spurious of the azimuth evaluation function and more accurately estimating the incident angle of the arriving wave than before.

Further, according to the invention according to claim 8, the angle measurement results of the coarse angle measuring means as theta _est, orientation evaluation function F
Among the plurality of angles corresponding to the plurality of peaks of (θ), the angle closest to θ _est is configured to be output as the incident angle of the arriving wave, so that the spurious of the azimuth evaluation function is suppressed. An angle measuring device capable of accurately estimating the incident angle of an incoming wave can be obtained.

[Brief description of the drawings]

FIG. 1 is a basic configuration diagram of a conventional and the angle measurement device of the present invention.

FIG. 2 is an internal configuration diagram of a conventional signal processing device of the angle measuring device of FIG. 1;

FIG. 3 is a diagram for explaining the operation of the azimuth evaluation function calculation means in FIG. 2;

FIG. 4 is an internal configuration diagram of the signal processing device of the angle measuring device of FIG. 1 showing the first and second embodiments according to the first and second aspects of the present invention;

FIG. 5 is a diagram for explaining the operation of the azimuth evaluation function calculation means and the fusion azimuth evaluation function calculation means of FIG. 4;

FIG. 6 is an internal configuration diagram of a signal processing device of the angle measuring device of FIG. 1 showing a third embodiment of the invention according to claim 3;

FIG. 7 is an internal configuration diagram of the signal processing device of the angle measuring device of FIG. 1 showing the fourth, fifth, and sixth embodiments of the present invention;

FIG. 8 is a diagram for explaining the operation of the correlation matrix calculating means and the time of the received signal data in FIG. 7;

FIG. 9 is an internal configuration diagram of a signal processing device of the angle measuring device of FIG. 1 showing a seventh embodiment of the invention according to claim 7;

FIG. 10 is a diagram for explaining operations of an azimuth evaluation function calculating unit, a coarse angle measuring unit, and a search range determining unit of FIG. 9;

FIG. 11 shows an eighth embodiment of the invention according to claim 8;
FIG. 3 is an internal configuration diagram of a signal processing device of the angle measuring device.

FIG. 12 is a diagram for explaining the operations of an azimuth evaluation function calculating unit, a coarse angle measuring unit, and a peak detecting unit of FIG. 11;

[Explanation of symbols]

Reference Signs List 1 receiving antenna 2 receiver 3 analog / digital converter 4 signal processing device 5 display device 201 buffer memory 202, 202a to 202z correlation matrix calculation means 203, 203a to 203z eigenvalue and eigenvector calculation means 204, 204a to 204z azimuth evaluation function calculation Means 205 Peak calculation means 206 Angle collation means 401a to 401z Buffer memory 406 Fusion direction evaluation function calculation means 606 Fusion direction evaluation function calculation means 701 Buffer memories 702a to 702z Correlation matrix calculation means 706 Fusion direction evaluation function calculation means 910 Monopulse angle measurement processing Means 911 Search range determination means

──────────────────────────────────────────────────続 き Continued on front page (58) Field surveyed (Int.Cl. ⁶ , DB name) G01S 3/28 G01S 3/04 G01S 3/48

Claims (8)

(57) [Claims] 1. M number of receivers, a correlation matrix calculating means for dividing a received signal of the M channel which is an output of the receiver into a plurality of groups, and calculating a correlation matrix for the received signals of each group; Eigenvalue and eigenvector calculating means for obtaining eigenvalues and eigenvectors of the output matrix of the matrix calculating means and extracting eigenvectors corresponding to the minimum or near eigenvalues, and the angle θ of the receiver corresponding to the m-th input of the correlation matrix calculating means The vector having the phase delay in the direction and the directivity as the m-th element is defined as a _j (θ), and the inner product of the eigen vector extracted by the eigen value and eigen vector calculator and the a _j (θ) is obtained for each θ. and a bearing evaluation function calculating means for outputting as a direction evaluation function F _j (theta) the reciprocal union of the received signal channel to the input of the correlation matrix calculating means It said a _j (theta) is differently arranged above receivers each other, the orientation evaluated independently for each switch a combination of the received signal channel function F by switching allowed
_j (θ), and the above-mentioned azimuth evaluation function F _j as G (θ)
An angle measuring device for calculating an average value of a reciprocal of (θ), searching for a peak of the obtained G (θ), and outputting an angle corresponding to the peak as an incident angle of an incoming wave. apparatus. 2. The function G (θ) is defined as F at an angle θ.
_{2. The} angle measuring apparatus according to claim 1, wherein the angle is given by a minimum value of _j (θ). 3. The function G (θ) is defined as F at an angle θ.
_{2. The} angle measuring apparatus according to claim 1, wherein the angle is given by a median value of _j (θ). 4. M number of receivers, a correlation matrix calculating means for reading out a received signal of the M channel which is an output of the receiver from a buffer memory at a predetermined time and calculating a correlation matrix for the received signal; An eigenvalue and an eigenvector calculating means for obtaining an eigenvalue and an eigenvector of the output matrix of the correlation matrix calculating means and extracting an eigenvector corresponding to a minimum or eigenvalue close to the minimum; and a receiver corresponding to an m-th input of the correlation matrix calculating means. the vector phase lag and directivity characteristics for the angle theta direction and m-th element as a _j (θ), determine the inner product of eigenvectors and above the eigenvalue and eigenvector computing means has extracted every θ a _j (θ) this and a bearing evaluation function calculation means for outputting the inverse of the inner product as a direction evaluation function F _j (θ), received the above correlation matrix calculating means uses the correlation calculation Corresponding to switch a combination of time so that the function form differ from each other by the J seek independent orientation evaluation function F _j (theta) determined the issue, the orientation evaluation function as H (θ) F
a peak detecting means for calculating an average value of a reciprocal of _j (θ), searching for a peak of the obtained H (θ), and outputting an angle corresponding to the peak as an incident angle of an incoming wave. Horn device. 5. The function H (θ) is defined as F at an angle θ.
5. The angle measuring apparatus according to claim 4, wherein the angle is given by a minimum value of _j (θ). 6. The function H (θ) is defined as F at an angle θ.
5. The angle measuring apparatus according to claim 4, wherein the angle is given by a median value of _j (θ). 7. M receivers, correlation matrix calculating means for calculating a correlation matrix of an M-channel received signal output from the receiver, and eigenvalues and eigenvectors of an output matrix of the correlation matrix calculating means. Eigenvalue and eigenvector calculating means for extracting eigenvectors corresponding to eigenvalues obtained from the minimum or close to the minimum; and a phase delay or directional characteristic with respect to the angle θ direction of the receiver corresponding to the m-th input of the correlation matrix calculating means. A vector as an element is defined as a (θ), and an inner product of the eigenvalue and the eigenvector extracted by the eigenvector calculating means and the a (θ) is obtained for each θ, and the reciprocal of the inner product is calculated as the azimuth evaluation function F
Azimuth evaluation function calculating means for outputting as (θ), peak detecting means for searching for the peak of F (θ) and outputting the angle corresponding to this peak as the arrival angle of the arriving wave, and the receiver output Coarse angle measuring means for inputting the M-channel received signal and coarsely measuring the angle of incidence of the arriving wave, wherein the angle measuring result of the coarse angle measuring means is θ _est , and the azimuth evaluation function calculating means is θ _est F (θ) is calculated for the angle range limited to the vicinity of θ _est , and the peak detection means calculates F (θ) for the angle range limited to the vicinity of the θ _est.
An angle measuring device, which searches for a peak of an angle. 8. M number of receivers, correlation matrix calculating means for calculating a correlation matrix of an M-channel received signal which is an output of the receiver, and eigenvalues and eigenvectors of an output matrix of the correlation matrix calculating means. Eigenvalue and eigenvector calculating means for extracting eigenvectors corresponding to eigenvalues obtained from the minimum or close to the minimum; and a phase delay or directional characteristic with respect to the angle θ direction of the receiver corresponding to the m-th input of the correlation matrix calculating means. A vector as an element is defined as a (θ), and an inner product of the eigenvalue and the eigenvector extracted by the eigenvector calculating means and the a (θ) is obtained for each θ, and the reciprocal of the inner product is calculated as the azimuth evaluation function F
Azimuth evaluation function calculating means for outputting as (θ), peak detecting means for searching for the peak of F (θ) and outputting an angle corresponding to this peak, and receiving the M-channel received signal output from the receiver. A coarse angle measuring means for coarsely measuring the angle of arrival of the incoming wave, wherein the angle measurement result of the coarse angle measuring means is defined as θ _est , and a plurality of peaks of F (θ) output from the peak detecting means are provided. An angle measuring device for outputting an angle closest to _θest among a plurality of corresponding angles as an incident angle of an incoming wave.