JP2000209018A - Array antenna control method, device therefor and recording medium recording array antenna control program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide the array antenna control method and device therefor which can obtain a receiving processing signal having an improved S/N compared with the conventional one regardless of the radio wave environment. SOLUTION: An inherent value analysis part 10 calculates a covariance matrix of a space area based on M signals received by the antenna elements of an array antenna, calculates the proper value of the covariance matrix of the space area that is calculated by a prescribed proper value analysis method and then calculates a characteristic vector corresponding to the calculated proper value. A beam forming part 6 multiplies M received signals by the characteristic vector to form the multi-beam signals corresponding to the number of proper values. A radio wave environment deciding part 11 switches the switches SW1 and SW3 to output a receiving process signal so that the received signals include a signal that is equal to a prescribed reference signal, calculates the correlation between the beam and reference signals and outputs the beam signal having the largest correlative coefficient.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and an apparatus for controlling an array antenna, and a recording medium storing a control program for an array antenna.

[0002]

2. Description of the Related Art With the recent development of digital signal processing technology, an adaptive array antenna based on digital signal processing has attracted attention. The adaptive array antenna adapts to the ever-changing electromagnetic environment, directs the peak of the antenna pattern in the direction of arrival of the desired wave, captures the delayed signal due to fading, synthesizes in-phase, and forms a null in the direction of arrival of the interference wave Thus, spatial filtering can be realized. Such flexible adaptive control,
Interest is increasing from the viewpoint of effective use of frequency.

When an array antenna existing under a certain radio wave environment is used and an eigenvector corresponding to the maximum eigenvalue of the spatial domain covariance matrix obtained therefrom is applied to the array antenna as a weighting factor, the array antenna forms a certain antenna pattern. . In particular, when the radio wave environment is a multipath fading environment in which the propagation delay is extremely small, it has been conventionally known that an array antenna using such eigenvectors as a weighting factor realizes maximum ratio combining of multipath signals. ing. This is hereinafter referred to as a first conventional example.

Further, a space F in digital signal processing is used.
A configuration in which a fixed beam is formed in advance in all stages for performing signal processing on an array antenna reception signal by FT, a Butler matrix by analog signal processing, and the like and a space is divided in advance is called a beam space method. A method of reducing the number of weighting factors for adaptive control by comparing each of the beam outputs according to a predetermined criterion such as SN ratio and power and selecting only a part thereof and performing adaptive control on the output is conventionally known. (For example, Prior Art Document 1 "Yukihiro Kamiya et al.," Study on Beam Space RLS Adaptive Array ", IEICE Technical Report, SAT9
7-79, November 1997 ". ). Hereafter,
This is referred to as a second conventional example.

[0005]

However, in the first conventional example, when an interference signal is present, a plurality of eigenvalues of the spatial domain covariance matrix appear, so that the maximum eigenvalue is selected and the corresponding eigenvector is selected. By using as a weighting factor, it was not possible to combine only the multipath signal of the desired signal with the maximum ratio.

In the second conventional example, the space FF
In a multi-beam formed by T and a Butler matrix, since a space is simply divided almost equally, a beam in which signals are completely spatially separated cannot be formed. For this reason, some beams mainly including signal components are selected. However, naturally, beams which are not selected also include some signal components, and the components are lost. When the beam is reduced based on the conventional beam space method, the SN ratio of the final output of the array antenna is inevitably deteriorated.

[0007] An object of the present invention is to solve the above-mentioned problems, and to improve S / S compared to the conventional example regardless of the radio wave environment.
An object of the present invention is to provide an array antenna control method and a control device capable of obtaining a reception processing signal having N, and a recording medium storing an array antenna control program.

[0008]

According to a first aspect of the present invention, there is provided a method for controlling an array antenna comprising a plurality of M antenna elements arranged in close proximity in a predetermined arrangement shape. Control method for performing
A covariance matrix in the spatial domain is calculated based on a plurality of M received signals received by each antenna element of the array antenna, and a covariance matrix in the spatial domain calculated using a predetermined eigenvalue analysis method is calculated. An analysis step of calculating an eigenvalue and calculating and outputting an eigenvector corresponding to the calculated eigenvalue; and calculating the eigenvector calculated for a plurality of M received signals respectively received by the antenna elements of the array antenna. Forming and outputting a multi-beam signal corresponding to the number of eigenvalues by multiplying the received signal, the received signal includes the same signal as a predetermined reference signal, and the output beam signal and the reference Calculating a correlation coefficient between the received signal and the received signal, and outputting a beam signal having the largest correlation coefficient as a reception processing signal. And,

According to a second aspect of the present invention, there is provided a control method of an array antenna for controlling an array antenna including a predetermined plurality of M antenna elements closely arranged in a predetermined arrangement shape. A plurality of M received by each antenna element of the array antenna, respectively.
Based on the received signals, calculate a covariance matrix in the spatial domain, calculate an eigenvalue of the covariance matrix in the spatial domain calculated using a predetermined eigenvalue analysis method, and calculate an eigenvector corresponding to the calculated eigenvalue. An analysis step of calculating and outputting;
A multi-beam signal corresponding to the number of eigenvalues is formed by multiplying the plurality of M received signals received by each antenna element of the array antenna by the calculated eigenvector, and outputting the multibeam signal. A transversal filter having at least the number of the multi-beam beam signals, inputting the output beam signals to each of the transversal filters, and setting a weight coefficient of each of the transversal filters to a predetermined adaptive control algorithm. And a first control step of adding the signals output from the transversal filters and outputting the sum as a reception processing signal.

Further, according to a third aspect of the present invention, there is provided a method for controlling an array antenna, comprising: a plurality of M antenna elements arranged in close proximity in a predetermined arrangement shape. A method for calculating a covariance matrix of a spatial domain based on a plurality of M received signals respectively received by the antenna elements of the array antenna, and calculating a spatial domain covariance matrix using a predetermined eigenvalue analysis method. An estimating step of calculating an eigenvalue of a covariance matrix of, calculating and outputting an eigenvector corresponding to the calculated eigenvalue, and a plurality of M received signals respectively received by each antenna element of the array antenna, A forming step of forming and outputting a multi-beam signal corresponding to the number of eigenvalues by multiplying the calculated eigenvectors, A second control step of judging the radio wave environment of the plurality of M received signals based on the output eigenvalues and selectively switching and executing signal processing on the beam signals based on the output beam signals. It is characterized by the following.

According to a fourth aspect of the present invention, in the method of controlling an array antenna according to the third aspect, the second control step includes the step of calculating the number of the eigenvalues based on the calculated number of the eigenvalues. A first case in which the number is 1 is determined, the received signal is multiplied by an eigenvector corresponding to the eigenvalue, the output one beam signal is output as a reception processing signal, and the output beam signal is output. And determining a second case in which a time delay equal to or more than a predetermined time delay amount has occurred, including a transversal filter having at least the number of the multi-beam beam signals, and It is input to a transversal filter, and the weighting factor of each transversal filter is controlled using a predetermined adaptive control algorithm. A signal output from the sversal filter is added and output as a reception processing signal. In cases other than the first and second cases, the signal processing apparatus includes a multiplier for at least the number of the multi-beam beam signals. Input to each multiplier, a weighting coefficient which is a multiplication coefficient of each of the multipliers is controlled using a predetermined adaptive control algorithm, and a signal output from each of the multipliers is added to generate a reception processing signal. Is output.

Further, in the method of controlling an array antenna according to a fifth aspect of the present invention, in the method of controlling an array antenna according to the fourth aspect, the second control step may be performed in a case other than the first and second cases. The received signal includes the same signal as a predetermined reference signal, calculates a correlation coefficient between the output beam signal and the reference signal, and compares the calculated correlation function with a predetermined threshold. When there is a beam signal having a correlation function equal to or greater than the threshold value, a beam signal having a maximum correlation coefficient is output as a reception processing signal.

According to a sixth aspect of the present invention, there is provided an array antenna control device for controlling an array antenna including a predetermined plurality of M antenna elements closely arranged in a predetermined arrangement shape. Then, a covariance matrix in the spatial domain is calculated based on a plurality of M received signals respectively received by the antenna elements of the array antenna, and the covariance matrix in the spatial domain calculated using a predetermined eigenvalue analysis method is calculated. Analysis means for calculating eigenvalues of the variance matrix, calculating and outputting eigenvectors corresponding to the calculated eigenvalues, and a plurality of M received by each antenna element of the array antenna
Forming means for forming and outputting a multi-beam signal corresponding to the number of eigenvalues by multiplying the received signals by the eigenvector calculated by the analyzing means, and the received signal is a predetermined reference signal Selecting means for calculating a correlation coefficient between the beam signal output from the forming means and the reference signal, and outputting a beam signal having a maximum correlation coefficient as a reception processing signal. It is characterized by having.

According to a seventh aspect of the present invention, there is provided an array antenna control apparatus for controlling an array antenna including a predetermined plurality of M antenna elements closely arranged in a predetermined arrangement shape. A plurality of M received by each antenna element of the array antenna, respectively.
Based on the received signals, calculate a covariance matrix in the spatial domain, calculate an eigenvalue of the covariance matrix in the spatial domain calculated using a predetermined eigenvalue analysis method, and calculate an eigenvector corresponding to the calculated eigenvalue. Analysis means for calculating and outputting, and multiplying the plurality of M received signals respectively received by the respective antenna elements of the array antenna by the eigenvectors calculated by the analysis means correspond to the number of eigenvalues. Forming means for forming and outputting a multi-beam signal, and transversal filters of at least the number of the multi-beam signals are provided, and the beam signals output from the forming means are input to the transversal filters. Controlling the weighting factor of each transversal filter using a predetermined adaptive control algorithm, Characterized by comprising a first control means for outputting a signal outputted from the over transversal filter as a reception processing signal by adding.

Further, according to an eighth aspect of the present invention, there is provided an array antenna control device for controlling an array antenna comprising a predetermined plurality of M antenna elements closely arranged in a predetermined arrangement shape. An apparatus for calculating a covariance matrix of a spatial domain based on a plurality of M received signals respectively received by the antenna elements of the array antenna, and calculating a spatial domain covariance matrix using a predetermined eigenvalue analysis method. Analysis means for calculating the eigenvalue of the covariance matrix of, calculating and outputting an eigenvector corresponding to the calculated eigenvalue, and for a plurality of M received signals respectively received by each antenna element of the array antenna, Forming means for forming and outputting a multi-beam signal corresponding to the number of eigenvalues by multiplying the eigenvectors calculated by the analyzing means, Based on the eigenvalue calculated by the analyzing means and the beam signal output from the forming means, the radio wave environment of the plurality of M received signals is determined, and signal processing for the beam signal is selectively switched and executed. And a second control means.

According to a ninth aspect of the present invention, in the array antenna control device according to the eighth aspect of the present invention, the second control means is configured to execute the control based on the number of eigenvalues calculated by the analysis means. Determining a first case in which the number of the eigenvalues is 1, multiplying the received signal by an eigenvector corresponding to the eigenvalue and outputting one beam signal output from the forming unit as a reception processing signal; A second case in which a time delay equal to or more than a predetermined time delay has occurred is determined based on the beam signal output from the forming unit, and a transversal filter having at least the number of the multi-beam beam signals is provided, The beam signal output from the forming means is input to each of the transversal filters, and the weight coefficient of each of the transversal filters is set to a predetermined value. Controlling using an adaptive control algorithm, adding the signals output from the transversal filters and outputting the sum as a reception processing signal. At times other than the first and second cases, at least the multi-beam beam signal Number of multipliers, the beam signal output from the forming means is input to each multiplier, and a weighting coefficient, which is a multiplication coefficient of each multiplier, is controlled using a predetermined adaptive control algorithm. It is characterized in that signals output from the multiplier are added and output as a reception processing signal.

In a tenth aspect of the present invention, in the array antenna control device according to the ninth aspect, the second control means includes the first and the second control means.
In other cases, the received signal includes the same signal as the predetermined reference signal, and calculates a correlation coefficient between the beam signal output from the forming means and the reference signal, and calculates the calculated correlation. When a beam signal having a correlation function equal to or greater than the threshold value exists by comparing the function with a predetermined threshold value, a beam signal having a maximum correlation coefficient is output as a reception processing signal. .

According to the present invention, there is provided a recording medium storing a control program for an array antenna according to the present invention.
An array antenna control program including the array antenna control method according to any one of the first to fifth aspects is recorded.

[0019]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram showing the configuration of an array antenna control device according to an embodiment of the present invention. The control device for an array antenna according to this embodiment is a control device for controlling an array antenna 1 including a predetermined plurality of M antenna elements 1-1 to 1-M closely arranged in a predetermined arrangement shape. And the beam forming unit 6
An eigenvalue analysis unit 10 and a radio wave environment determination unit 11 are provided. The radio wave environment determination unit 11 determines a radio wave propagation environment or a radio wave environment in a received signal received by the array antenna 100 as follows.
It is characterized in that optimal signal processing is executed by categorizing and determining the cases and switching the switches SW1, SW2, SW3 according to each case. (A) Case 1: When there is no interference wave and there are a plurality of multipath waves of the desired wave, but the spread of the delay is very small. (B) Case 2-1: In addition to the multipath wave of the desired wave,
When there is an interference wave, and the order of the power of the interference wave and the power of the desired wave is substantially the same. (C) Case 2-2: In addition to the multipath wave of the desired wave,
When an interference wave exists and the order of the power of the interference wave and that of the desired wave is different. (D) Case 3: When there are a plurality of multipath waves of the interference wave and the desired wave, but the spread of the delay is large.

In this embodiment, the received signal is
For example, it is a digital data signal that is digitally modulated with an audio signal, a video signal, or a data signal, and that includes a synchronization pattern signal including a reference signal generated by the reference signal generator 8.

In the present embodiment, a case is considered in which the M array of antenna elements 1-1 to 1-M are arranged side by side on a straight line at equal intervals of, for example, half a wavelength. In FIG. 1, receivers 2-1 to 2-M
Include a frequency converter and a demodulator, and are configured similarly. The complex signal generators 3-1 to 3-M convert an input signal into a complex signal having two signal components orthogonal to each other and output the complex signal. The A / D converters 4-1 to 4-M convert the received analog reception signal into a digital reception signal and have the same configuration. Are divided into two and output, and are configured similarly. In the circuits after the complex signal generators 3-1 to 3-M, processing is performed by complex operations on complex signals.

Here, the reception signal received by the antenna element 1-1 is transmitted to the receiver 2-1 and the complex signal generator 3-
1 and the digital reception signal x via the A / D converter 4-1.
The received signal input to the beam forming unit 6 and the eigenvalue analyzing unit 10 as 1 and received by the antenna element 1-2 is converted into a receiver 2-2, a complex signal generator 3-2, and an A / D converter 4-2. is input bets to the beam forming unit 6 and the eigenvalue analysis unit 10 as a digital reception signal x ₂ via a further similarly, the received signal received by the antenna element 1-M is, the receiver 2-M and the complex signal generator vessel 3-M and via the a / D converter 4-M are input to the beam forming unit 6 and the eigenvalue analysis unit 10 as a digital reception signal x _M.

FIG. 4 is a flowchart showing the eigenvalue analysis processing executed by the eigenvalue analysis unit 10 of FIG. As shown in FIG. 4, first, from the distributor 5-1 to 5-M in step S1, the received signal _{X = [x 1, x 2} ,
.., X _M ] ^T is input, and in step S2, the received signal X
And the spatial domain covariance matrix R _{x using}
Is calculated.

R _x = E [X ^* X ^T ] where * is complex conjugate, T is transpose, and E [•] is ensemble average.

[0025] Then, the eigenvalues lambda ₁ in the spatial domain covariance matrix R _x by using a known eigenvalue analysis in step S3,
lambda _2, ..., and calculates the lambda _M, eigenvalue lambda ₁ calculated in step S4, lambda _2, ..., a predetermined noise threshold and lambda _M respectively (the noise threshold, whether the noise signal , ^Which are set to, for example, 10 ⁻⁹ ), and search for effective eigenvalues λ ₁ , λ _{2 ,.} , And removes the eigenvalues serving as noise), and outputs the number Ne to the radio wave environment determination unit 11. Further, the effective eigenvalues lambda ₁ in step S5, lambda _2, ..., eigenvector corresponding to λ _Ne V _1, V _2,
.., V _Ne with a certain eigenvalue λm and its eigenvector V _n =
[V ₁ , v ₂ ,..., V _M ] ^T (n = 1, 2,..., Ne)
Output to

[Number 2] After the _{R x V n = λ m V} n step S5, in order to execute the processing of the received signal X for the next sample, the flow returns to step S1.

FIG. 2 is a block diagram showing the configuration of the beam forming unit 6 of FIG. In FIG. 2, the beam forming unit 6
Are delay circuits 41-1 through 41-M for respectively delaying the reception signal X by the calculation time of the eigenvalue analysis unit 10, and M
× M multipliers 42-1-1 to 42-MM and M ′
(1 ≦ M ′ ≦ M−1) adders 43-1 to 43-
M ′, and the delay circuits 41-1 through 41-
For the delayed received signal X output from M, the eigenvector V _n (n = 1,
, Ne), and by adding each component signal (element signal) of the received signal X, M ′ beam signals B _{1 to} B _{M ′} are calculated, and the four-way splitter 7 Output to

In FIG. 1, the four-way splitter 7 converts M ′ beam signals output from the beam forming unit 6 into first to fourth beam signals.
4 distributed to the distribution signal, a first beam signal B1 of the first distributed signal and outputs through the contact a of the switch SW3 as the reception processing signal P ₀ of the case 1, also, the second distributed signal the M ': 1 as a reception processing signal P ₀ of the case 2-2 via the switch SW1 is selecting the changeover switch via the contact b of the switch SW3 and outputs, furthermore, the contacts of the third distribution signal of the switch SW2 Through a, M '
A reception processing signal P _{0 of} case 3 via a TDL (Tapped Delay Line) circuit 20 which is a time delay circuit having a plurality of transversal filters and a contact c of a switch SW3.
And the third distributed signal is output as the reception processing signal P ₀ of the case 2-1 via the contact point b of the switch SW2, the weight coefficient multiplier 30 and the contact point d of the switch SW3, and Further, it outputs the fourth distribution signal to the radio wave environment determining unit 11. Here, the switch SW
The switching between the SW1, SW2, and SW3 is controlled by the radio wave environment determination unit 11 as described later in detail.

FIG. 3 is a block diagram showing a configuration of TDL circuit 20 of FIG. As shown in FIG. 3, it is configured to include M ′ transversal filters 51-1 to 51-M ′ and an adder 52. Here, each of the transversal filters 51-1 to 51-M ′ is connected to N stages of cascaded delay circuits 53-1 to 53-N and its input / output terminals, and is calculated by the weight coefficient calculator 21. (N + 1) multipliers 54-0 for multiplying the obtained weight coefficient vector W
To 54-N and an adder 55 for adding all the output signals of the multipliers 54-0 to 54-N. The adder 52 includes all the transversal filters 5
The signals from 1-1 to 51-M ′ are added and output as a reception processing signal P ₀ to the weight coefficient calculator 21.
The signal is output via the contact c of the switch SW3.

[0029] In the transversal filter 51-1 TDL circuit 20 of FIG. 3, as shown in FIG. 3, the beam signals B ₁ as an input signal b _{0, 1,} b ₁ output signal of the delay circuit _{53-1, 1,} and similarly in the following, the delay circuit 53-N
Is an output signal of b _{N, 1} . At this time, the signals b _{0,1 to} b
_{Let the} weighting factors for _{N, 1 be} w _0,1 through w _{N, 1} . Less than,
Similarly, in the transversal filter 51-M ′, the beam signal B _{M ′} is set as the input signal b _{0, M ′} , the output signal of the delay circuit 53-1 is set as b _{1, M ′,} and so on. The output signal of the circuit 53-N is denoted by _{bN, M '} . At this time, the weighting factor for the signals b _{0,1 to} b _{N, M ′} is w
_{0,1 to} w _{N, M '} .

The weighting coefficient calculator 21 is based on the reception processing signal P ₀ output from the TDL circuit 20 and the reference signal R generated by the reference signal generator 8, for example, a known equation represented by the following equation: RLS (Recursive Least Squares)
The weight coefficient vector W is calculated using the algorithm and T
The signal is output to the DL circuit 20 and the radio wave environment determination unit 11.

[0031]

W (n + 1) = W (n) + k (n + 1) [R−
W ^T (n) B (n + 1)] where

K (n + 1) = P ₀ (n) B (n + 1) [1+
B ^T (n + 1) P ₀ (n) B (n + 1)] ⁻¹

P ₀ (n + 1) = [I−k (n + 1) B ^T (n +
1)] P ₀ (n)

Here, the beam signal vector B and the weight coefficient vector W are represented by the following equations.

[0033]

[Mathematical formula-see original document] B = [ _b0,1 , _b1,1 ,..., _BN , ₁ , _b0,2 , b
_1,2 , ..., bN _{, 2} , ..., b0 _{, M '} , b1 _{, M '} , ...,
b _{N, M '} ] ^T

W = [w _0,1 , w _1,1 ,..., W _{N, 1} , w _0,2 , w
_1,2 , ..., wN _{, 2} , ..., w0 _{, M '} , w1 _{, M '} , ...,
w _{N, M '} ] ^T

Further, n is a repetition step repeated for each sample of the received signal X, and I is a unit matrix. The weight coefficient calculator 21 performs a 2M ′ repetition operation until a predetermined convergence condition such as a difference between the weight coefficient vectors W before and after one step becomes equal to or less than a predetermined threshold is satisfied (in the RLS algorithm). It is known that the above converges.) Equation 3 above is repeatedly calculated, and after convergence, the weight coefficient vector W is output to the TDL circuit 20 and the radio wave environment determining unit 11. At this time, the TDL circuit 20
The reception processing signal P ₀ (scalar value) output from is represented by the following equation.

[0035]

## EQU8 ## P ₀ = W ^T B

The weight coefficient multiplier 30 receives the input M ′ beam signals, and applies a heavy coefficient vector generated by the weight coefficient calculator 31 to the input M ′ beam signals. W ′ = [w ₁ , w ₂ ,..., W _{M ′} ] are multiplied and output, and a plurality of M ′ multipliers are added, and all the output signals of the M ′ multipliers are added, and the addition result signal is obtained. The signal is output to the weighting coefficient calculator 31 as the reception processing signal P ₀ of the case 2-2, and is output via the contact d of the switch SW3.

The weighting factor calculator 31 includes a weighting factor multiplier 3
0, based on the received processed signal P ₀ and the reference signal R generated by the reference signal generator 8, for example, a known RLS (Recursive Least Square) represented by the following equation:
s) The weight coefficient vector W ′ =
[W ₁ , w ₂ ,..., W _{M ′} ] are calculated and the weight coefficient multiplier 3
Output to 0.

[0038]

W ′ (n + 1) = W ′ (n) + k (n + 1) [R−
W ' ^T (n) B' (n + 1)] where

K (n + 1) = P ₀ (n) B ′ (n + 1) [1+
B ′ ^T (n + 1) P ₀ (n) B ′ (n + 1)] ⁻¹

## EQU11 ## P ₀ (n + 1) = [I−k (n + 1) B ′
^T (n + 1)] P ₀ (n)

Here, the beam signal vector B 'is expressed by the following equation.

[0040]

B ′ = [B ₁ , B ₂ ,..., B _{M '} ] ^T

Further, n is a repetition step repeated for each sample of the received signal X, and I is a unit matrix. The weighting factor calculator 31 performs a 2M ′ repetition operation until a predetermined convergence condition such as a difference between the weighting factor vector W ′ before and after one step becomes equal to or less than a predetermined threshold value is satisfied (in the RLS algorithm). It is known that convergence is achieved by performing the above.)
After the convergence, the weight coefficient vector W is output to the weight coefficient multiplier 30. At this time, the reception processing signal P ₀ (scalar value) output from the weight coefficient multiplier 30 is represented by the following equation.

[0042]

P ₀ = W ' ^T B'

FIGS. 5 and 6 show the radio wave environment judgment unit 1 shown in FIG.
3 is a flowchart showing a radio wave environment determination process executed by the first embodiment. In FIG. 5, first, in step S11, the eigenvalue analysis unit 10 outputs the effective eigenvalues λ ₁ , λ ₂ ,.
The number Ne of λ _Ne is input, and Ne = Ne in step S12.
1 is determined, and if YES, it is determined as case 1 in step S13, and the switch SW3
Is switched to the contact point a, and the radio wave environment determination processing ends. At this time, the beam signals B ₁ is a desired wave is output as the received processed signal P ₀ via the contact a of the beam forming unit 6 4 distributor 7 and the switch SW3.

If NO in step S12, it is determined that the case is 2 or 3, and in step S14, the switch SW2 is switched to the contact point a, and the weight coefficient calculator 2
1 is operated to calculate the weight coefficient vector W and input to the radio wave environment determining unit 11. Next, in step S15, each element of the weight coefficient vector W of each tap of the TDL circuit 20 is compared with a predetermined coefficient threshold value (for example, set to 0.8) for each beam signal, and the coefficient is calculated. The number Nb of weight coefficients larger than the threshold value is obtained. Further, at step S16, Nb ≧ Nb with at least one beam signal.
It is determined whether the number is 2 or not. If YES, the desired wave and the interference wave are temporally dispersed, and the desired wave and the interference wave have a time delay equal to or more than a predetermined amount of time delay. Cannot be performed, and in step S17, case 3
Then, the switch SW2 is switched to the contact point b and the switch SW3 is switched to the contact point d, thereby ending the radio wave environment determination processing. Thus, by using the TDL circuit 20 having a transversal filter, to compensate for the time delay to obtain reception processing signal P ₀ having improved S / N.

If NO in the above step S16,
Goes to step S18, and is output from the four distributor 7;
Each beam signal B _i (t) on the time axis (i = 1, 2,...,
M ′) and the correlation coefficient C _i between the corresponding reference signal R (t) on the time axis are calculated for each beam signal B _i using the following equation (14). Here, as described above, since the received signal X includes a signal having the same pattern as the reference signal R as a preamble signal, a reference signal also in each beam signal B _i R
And the reference signal R
Calculate the correlation coefficient with

[0046]

C _i = {(1 / N) B _i ^* (t) R (t)} /
[√ {(1 / N) B _i ^* (t) B _i (t) · (1 / N) R ^* (t) R
(t)}] Here, each beam signal B _i (t) on the time axis and the corresponding reference signal R (t) on the time axis are represented by time t1, t2,
, TN 'is expressed by the following equation.

[0047]

## EQU15 ## B _i (t) = [b _i (t ₁ ), b _i (t ₂ ),
_{..., b i (t N '} )] T

R (t) = [r (t ₁ ), r (t ₂ ),..., R
( _{TN '} )] ^T

[0048] Then, the correlation coefficient C _i is a predetermined correlation coefficient threshold (e.g., 0.95) in step S19 it is determined whether the beam signal exceeding the exists, YES
In step S20, it is determined in step S20 that the power of the desired wave is different from the power of the interference wave, that is, the case 2-2 is determined, and the switch SW1 is set so that a beam signal having the maximum correlation coefficient is output. And the switch SW3 is switched to the contact "b" in step S21 to end the radio wave environment determination processing. Therefore, Case 2
-2, a plurality M ′ output from the beam forming unit 6
Switch the largest beam signal among the beam signals to switch SW
And it outputs the selected one as the reception processing signal P _0.

If NO in step S19,
Proceeding to step S22, it is determined that the power of the desired wave is substantially in the same order as the power of the interference wave, that is, the case 2-1 is determined, and the switch SW2 is set to the contact point a. The switch SW3 is switched to the contact point C, and the radio wave environment determination processing ends. Therefore,
In Case 2-1, using the weight coefficient multiplier 30,
For example, the maximum ratio combining is adaptively performed using the RLS algorithm to obtain the reception processing signal P ₀ .

As described above, in case 1, there is no interference wave and there are a plurality of multipath waves of the desired wave. When the delay spread is very small, the corresponding eigenvector V ₁ already desired signal to one of the beam signals B _1, which is formed is output is spatially separated from the interference signal. At this time, no adaptive control algorithm is required. Therefore, the beam signal B ₁ is output as it is as the reception processing signal P ₀ .

In case 2-1, when an interference wave exists in addition to the multipath wave of the desired wave and the order of the power of the interference wave and the power of the desired wave is substantially the same, the weight coefficient multiplication is performed. The maximum ratio combining is adaptively performed by using the RLS algorithm, for example, by using the RLS algorithm to obtain the reception processing signal P _0.
Get.

Further, in case 2-2, when there is an interference wave in addition to the multipath wave of the desired wave, and when the order of the power of the interference wave and the power of the desired wave is different, the reference signal and the maximum for taking out a signal to be received a desired signal having a correlation, and outputs the switch SW1 as a reception processing signal P ₀ is switched so that the beam signals are output with the highest correlation coefficient.

Further, in case 3, there are a plurality of multipath waves of the interference wave and the desired wave, and when the spread of the delay is large, for example, the RDL using the TDL circuit 20 having the transversal filter is used. obtain received processed signal P ₀ adaptively performing maximum ratio combining using the algorithm.

<Effects of Embodiment> Case 2 of the present embodiment
With the configuration in -2, it is possible to select a beam that spatially extracts only a desired wave from a plurality of beam signals formed by eigenvectors. This makes it possible to control the array antenna by feedforward without using a weighting coefficient control algorithm having a feedback loop such as LMS, RLS, CMA, or the like. Further, since this method is based on eigenvalue analysis, it is possible to easily execute an array antenna control process based on radio wave environment determination, which selectively switches its own signal processing while determining a radio wave environment by eigenvalue analysis.

Also, with the configuration of case 3 of the present embodiment, all signal components are included in the beam outputs of a number smaller than the number of antenna elements by using a multi-beam with eigenvectors as in case 2-1. Therefore, the calculation load of the weighting factor calculation unit 31 can be reduced as compared with the case where signal processing is directly performed on the output from the antenna element. In Case 3, if the propagation delay in a multipath fading environment cannot be ignored, the effects of fading cannot be avoided unless signal processing in the time domain is performed.
In this case 3, the signal processing in the time domain is performed using the TDL circuit 20. However, since weight coefficients exist not only in the spatial domain but also in the time domain, the weight coefficients to be controlled are very large. In such a configuration, the reduction of the number of input signals to the weighting factor calculation unit 21 by the configuration using multi-beams with eigenvectors can greatly reduce the number of weighting factors.

Further, by providing the radio wave environment determining section 11 to determine the radio wave environment of the cases 1 to 3 and selectively switch the signal processing for the received signal, the following specific effects can be obtained. Comparing the calculation amounts of Case 1, Cases 2-1 and 2-2, and Case 3, the calculation amounts increase in the order of Case 1, Case 2-2, Case 2-1, and Case 3. For example, like a base station of a mobile communication system,
If the system has to deal with signals from a plurality of slave stations, and if case 3 is to be dealt with for all terminals, the resources of hardware such as DSP to realize this are enormous. Become. However, the propagation characteristics of signals from each terminal are different, and it is not necessary to deal with the thickest case 3 for all signals. In this case, it is assumed that the environment of the signal of the terminal to be handled by the base station is stochastically dispersed, and if the signal processing of the simpler cases 1, 2-1, and 2-2 can be performed, the environment If the signal processing can be switched based on the judgment of, the overall scale of the signal processing is stochastically reduced,
The effect of stochastically reducing the size of hardware resources can be expected.

[0057]

Hereinafter, an experiment and a result thereof simulated by using the apparatus of the present embodiment will be described.

<Experiments in Cases 1 and 2> FIGS. 7 to 18
2 shows a comparison between an antenna pattern based on an eigenvector and a pattern obtained by the RLS method when two incoherent signals are incident. Experimental specifications in the experiments of FIGS. 7 to 18 are shown in the following tables.

[0059]

Table 1 S / I values in FIGS. 7 to 18, correlation coefficient between received processing signal P ₀ by RLS method and reference signal, and correlation coefficient between beam signal and reference signal of embodiment ―――――――――――――――――――――――――――――――― Figure 7: S / I = 10dB RLS method: 0.9998, beam Signal B ₁ : 0.9998, Beam signal B ₂ : 0.0036 ―――――――――――――――――――――――――――――――――― FIG. 8: S / I = 0 dB RLS method: 0.9871, beam signal B ₁ : 0.7548, beam signal B2: 0.6360 ――――――――――――――――――――― --------------- Figure 9: S / I = -10dB RLS method: 0.9871, beam signals B _1: 0.0104, beam signals B _2: 0 9869 ―――――――――――――――――――――――――――――――― Figure 10: S / I = 10dB RLS method: 0.9998, Beam signal B ₁ : 0.9990, beam signal B ₂ : 0.0376 ――――――――――――――――――――――――――――――――― ― Figure 11: S / I = 0dB RLS method: 0.9851, beam signal B ₁ : 0.6919, beam signal B ₂ : 0.7005 ――――――――――――――――― ----------------- Figure 12: S / I = -10dB RLS method: 0.9850, beam signals B _1: 0.0316, beam signals B _2: .9845 ―――――――――――――――――――――――――――――――― Figure 13: S / I = 10dB RLS method: 0.9996, beam No. B _1: 0.9964, beam signals B _2: 0.0785 ---------------------------------- FIG. 14: S / I = 0 dB RLS method: 0.9658, beam signal B ₁ : 0.6962, beam signal B ₂ : 0.6684 ―――――――――――――――――― ---------------- Figure 15: S / I = -10dB RLS method: 0.9634, beam signals B _1: .0728, beam signals B _2: .9601 - ――――――――――――――――――――――――――――――― Figure 16: S / I = 10dB RLS method: 0.9989, beam signal B ₁ : 0.9952, beam signal B ₂ : 0.0850 ―――――――――――――――――――――――――――――――― 17: S / I = 0 dB RLS method: 0.9093, beam signal B ₁ : 0.6996, beam signal B ₂ : 0.5788 ―――――――――――――――――――――――― ―――――――――― Figure 18: S / I = -10dB RLS method: 0.8851, beam signal B ₁ : 0.0949, beam signal B ₂ : 0.8790 ――――――― ―――――――――――――――――――――――――――

That is, in FIG. 7 to FIG. 18, when the desired signal arrives from the zenith (0 degree), the pattern when the interference signal arrives from 30 degrees, 20 degrees, 10 degrees, and 5 degrees. 3 shows a comparison. Also, in these FIGS. 7 to 18, the desired wave to interference wave power ratio (S / I)
Indicates a comparison in each case of 10 dB, 0 dB, and -10 dB.

As is apparent from FIGS. 7 to 18, when the powers of the two incident waves are equal (FIGS. 8, 11, 14 and 17), the beam based on the eigenvector cannot separate the two signals. You can see that both are taken.
Further, the correlation coefficients of the beam outputs based on the two eigenvectors are substantially equal. On the other hand, when S / I = −10 dB (FIGS. 9, 12, 15, and 18), one of the beams based on the eigenvectors has already separated the interference signal to acquire the desired signal, and the output Although the correlation coefficient with the reference signal is slightly inferior to the output obtained by the RLS method, it shows almost the same value. Also, the antenna pattern is very close to the antenna pattern obtained by the RLS method. When S / I = 10 dB (FIGS. 7, 10,
13 and 16), since the interference signal has considerably lower power than the desired signal, a clear null is not formed in the direction of the interference signal as the antenna pattern. However, when the output signal is compared with the reference signal using the correlation coefficient, it can be seen that the desired signal is captured with fairly close accuracy, although the antenna pattern based on the eigenvector is inferior to that based on RLS.

From the above results, it can be said that when the incident power is of the same order, the pattern based on the eigenvector tends to take in the interference wave. Based on such observation, in the present embodiment, case 2 is classified into cases 2-1 and 2-2. That is, the case where the powers of the interference signal and the desired signal are in the same order is classified as Case 2-1. As shown in FIGS. 8, 11, 14, and 17, in this case, the desired signal and the interference signal cannot be separated depending on the beam generated by the eigenvector corresponding to each eigenvalue.

FIG. 19 shows respective antenna patterns of all eigenvectors when one wave of each of the desired signal and the interference signal enters at S / I = 0 dB. A first eigenvalue and a second
The antenna patterns generated by the eigenvectors V ₁ and V ₂ corresponding to the eigenvalues of 希望 希望 希望 希望 希望 希望 希望 希望 希望 希望 希望 希望 希望 干 渉 干 渉 干 渉 干 渉 、 、 、. Other beams form nulls in the signal arrival direction.

In such a case, a beam space configuration based on a multi-beam using two beam signals based on the first and second eigenvectors V ₁ and V ₂ including all the signal components is considered. Only RLS method or L
By performing adaptive filtering such as the MS method,
Adaptive control is performed. That is, in case 2-1 of the present embodiment, a multi-beam is formed using an eigenvector whose eigenvalue is greater than the noise power, and only the output of the multibeam is determined by an algorithm such as the RLS method. The output is obtained by multiplying by an adaptively controlled weighting factor. In a beam space based on a normal spatial FFT or the like, since a signal component is included in a beam output truncated when a beam is selected based on power or the like, when the beam is reduced, the S / N ratio of the output decreases. Is inevitable. However, in a beam space based on a multi-beam configuration using eigenvectors, the dimension of the adaptive control algorithm can be reduced theoretically without deterioration.

FIG. 20 shows the RLS method in the element space.
Antenna pattern obtained by using
U. ) And two eigenvalues corresponding to the first and second eigenvalues.
Kuturu V ₁, V_TwoRLS method based on beam space by using
The antenna patterns obtained by the above are compared. Both
Are almost coincident with each other, and
Comparing the correlation coefficients, the former is 0.9872 and the latter is
0.9869, which is almost the same.

In this calculation example, the case where the power of the desired wave and the power of the interference wave are substantially equal and both are only one wave is described as an example. If the total power is of the same order, it is recognized as case 2-1 and the above-described measures are effective.

Next, the experiment of Case 2-2 and the results thereof will be described. The case where the order of the power of the interference signal does not substantially match the order of the power of the desired signal is classified as Case 2-2. In an actual radio wave propagation environment, since the desired wave and the interference wave are superimposed on multipath waves having different phases, the total power including the antenna characteristics of the two is rarely the same. Case 2-2 is expected. At this time, an antenna pattern is obtained in which the interference signal is separated to some extent only by the antenna pattern generated by the eigenvector and only the desired signal is spatially extracted.

FIG. 21 shows an antenna pattern obtained by converging by adaptive control processing by the RLS method in an environment where each of an interference wave and a desired wave has a spatial spread due to the presence of a multipath wave, and FIG. shows the comparison of the beam pattern of the beam signals B ₁ corresponding to the eigenvalues. Here, the parameters are statistically given as one snapshot of the fading environment as follows. First, the arrival directions of the desired wave and the interference wave are each 30 degrees on average,-
It is given as a random number according to a normal distribution with a standard deviation of 10 degrees and 20 degrees, and the number of arriving waves is 20 each. Next, it is assumed that the amplitudes of both the desired wave and the interference wave are distributed in a Rayleigh manner.
B. Further, the phase of all incoming signals is [0,
[2π]].
In such a situation, the probability that the total power of the coherent components of the desired wave and the interference wave match will be very small.

The antenna pattern based on the eigenvector is
7, 10, 13, and 16, and FIGS. 9, 12,
In the case where the arriving signal shown in FIGS. 15 and 18 is not only one of each of the interference wave and the desired wave but also has a spatial spread, the interference wave itself has already been separated to some extent. Table 2 shows the values of the eigenvalues obtained under these conditions, and the correlation coefficient between the beam output and the reference signal and the correlation coefficient with the interference signal based on the eigenvector corresponding to each eigenvalue.

[0070]

[Table 2] Correlation coefficient between the eigenvalue and the reference signal of the beam output by the corresponding eigenvector in case 2-2 ―――――――――――――――――――――――― ―――――――――― Absolute value of eigenvalue Correlation coefficient with desired signal Correlation coefficient with interference signal ―――――――――――――――――――――― ―――――――――――― 143.1883 0.9984 0.0491 22.7632 0.0491 0.9966 0.1087 0.0007 0.0039 0.0971 0.0017 0.0004 ―― ――――――――――――――――――――――――――――――――

As is clear from Table 2, it can be seen that the first beam signal B ₁ corresponding to the first eigenvalue separates the interference signal very well. Here, the correlation coefficient of the beam output obtained by the RLS method is 0.9996.
For reasons such as the estimation error of the covariance matrix, the correlation coefficient of the RLS method has not been obtained. Also, when the arrival directions of the two waves are very close, it is difficult to separate only the eigenvectors. In such a case, the performance can be improved by similarly applying the beam space configuration using multiple beams based on the eigenvectors proposed in Case 2-1.

FIG. 21 shows an antenna pattern obtained by the configuration of case 2-2. It can be seen that the pattern based on the eigenvector is closer to the RLS method. The correlation coefficient between the signal obtained by the beam space based on the eigenvector and the reference signal is 0.9995, which is an improvement.

<Experiment of Case 3> In the computer simulation of Case 3, a multipath fading environment is defined statistically as follows. Table 3 shows the statistical properties of the environment and the simulation conditions.

[0074]

[Table 3] Fading environment and simulation conditions ―――――――――――――――――――――――――――――――― Number of incoming waves Poisson distribution (average 10 Wave) Arrival angle Normal distribution N (0 °, 10 °) Delay exponential distribution (average 1 symbol) Amplitude distribution Rayleigh distribution (overall S / N = 10 dB) Phase distribution Uniform distribution [0.2π] Antenna 4-element linear array Number of taps 32 (4 taps / symbol) algorithm RLS method (Recursive Least Squares) ――――――――――――――――――――――――――――――――

In the computer simulation, an 8-element linear array antenna was assumed, and the RLS method was adopted as an adaptive control algorithm. In the present embodiment, a simulation was performed for one snapshot in a fading environment defined here statistically.

FIG. 22 shows one snapshot of fading.
Obtained in the eight eigenvectors generated in the
3 shows three of the antenna patterns. Third beam signal
Issue B _ThreeDoes not spatially capture signal components
As you can see, all subsequent antenna patterns have the same inclination.
Direction. At the same time, the beam signal in FIG.
B ₁, B_TwoSelect the patterns shown as their output
In addition, control was performed by the RLS method using the configuration of case 3 in FIG.
The final antenna pattern obtained in this case is shown in FIG.
Received processing signal P₀As shown. In this FIG.
The graph in the section shows the amount of delay for each incoming signal in sample units.
It is shown. Table 2 shows the obtained eight eigenvalues.
Generated by the corresponding eigenvectors
Signal received by the transmitted beam and the reference signal given to the RLS method
This shows the correlation coefficient with.

[0077]

[Table 4] Correlation coefficient between the eigenvalue and the corresponding beam output and reference signal ――――――――――――――――――――――――――――――― ― Eigenvalue Correlation coefficient between reference signal and corresponding beam output ―――――――――――――――――――――――――――――― B _{1 9.7565 0.5180 B 2 0.1886 0.5914 B 3} 0.1158 0.0209 B 4 0.1035 0.0182 B 5 0.1035 0.0171 B 6 0.0972 0.0421 B 7 0. 0908 0.0023 B ₈ 0.0982 0.0328 ―――――――――――――――――――――――――――――――― B by beam space ₁ and B ₂ 0.9660 ――――――――――――――――――――――――――――――――――

As is apparent from FIG. 22, the first and second
Eigenvector V corresponding to the eigenvalue of ₁, V_TwoBut take the signal
The other eigenvalues are equal to the noise power,
It can be seen that no signal component is included. As a result, the first
And the eigenvector V corresponding to the eigenvalues of₁, V_TwoUsing
When the beam space method is configured by using
The number is 0.9660 shown in Table 4.

On the other hand, FIG. 23 shows an antenna pattern obtained by the above-described array antenna using the RLS method of the beam space based on the eigenvectors and the RL based on the element space.
7 shows a comparison with an antenna pattern obtained by an array antenna of the S method. At this time, RL by element space
The correlation coefficient between the output of the array antenna of the S method and the reference signal is 0.9652. Since the configuration of case 3 in FIG. 1 includes signal processing on the time axis, there is a slight difference in the spatial characteristics of the antenna, that is, the antenna pattern, but mainly in the direction in which the signal arrives. The main lobes pointed are coincident. In addition, the performance of comparison using the correlation coefficient is almost the same in both cases. At this time, in the configuration of the beam space method using the eigenvectors, only two beam outputs are supplied to the TDL circuit 20 by the RLS method, and the order of the adaptive control algorithm of the RLS method is 1 / compared to the case of the element space. It is 4.
Furthermore, the number of multiplications of 3M ² + 3M times required in the RLS method of order M is almost 1/16, and it has been confirmed that the calculation amount can be reduced without deteriorating the performance.

<Modification> In the above embodiment, the antenna elements 1-1 to 1-M are arranged one-dimensionally.
The present invention is not limited to this, and may be two-dimensionally arranged.

In the above embodiment, the weight coefficient calculators 21 and 31 calculate the weight coefficient vector W using the RLS algorithm. However, the present invention is not limited to this, and is represented by the following equation. A known LMS algorithm or CMA algorithm may be used.

(1) LMS Algorithm A calculation formula for calculating the weight coefficient vector W by the LMS algorithm is represented by the following equation. The weight coefficient calculators 21 and 31 are modified in the same manner as in the above-described embodiment.

W (n + 1) = W (n) + μB (n + 1)
[R−W ^T (n) B (n + 1)] Here, μ is a predetermined constant (0 <μ ≦ 1).

(2) CMA Algorithm The calculation formula for calculating the weight coefficient vector W by the CMA algorithm is expressed by the following equation. The weight coefficient calculators 21 and 31 are modified in the same manner as in the above-described embodiment.

## EQU18 ## W (n + 1) = W (n) -4.mu.B (n + 1) W
^{T (n) B (n +} 1) [| W T (n) B (n + 1) | 2 -σ 2] where, mu is a predetermined constant (0 <μ ≦ 1), σ is the amplitude (= 1).

Also, depending on the algorithm used, the reference signal may not be used except for the case 2-2. Further, when the reference signal is not used, the determination may be made without dividing Case 2 and the same signal processing as in Case 2-1 may be performed.

In the above embodiment, FIG.
The circuits on the output side after the complex signal generators 3-1 to 3-M may be configured by one or a plurality of digital computers for each function. Therefore, the processing for controlling the array antenna may be described by a program and executed by a digital computer such as a DSP. The program may be stored in a storage device such as a ROM, a RAM, or a hard disk, or may be stored in a CD-ROM, a CD-R,
After recording on an external recording medium such as D-RW, MO, etc.
After loading into an internal storage device such as M or a hard disk, control processing of the array antenna may be executed.

[0086]

As described in detail above, according to the present invention, there is provided an array antenna control method or apparatus,
A control method or device for controlling an array antenna composed of a predetermined plurality of M antenna elements closely arranged in a predetermined arrangement shape, wherein the plurality of M antenna elements received by each antenna element of the array antenna are provided. Based on the received signals, calculate a covariance matrix in the spatial domain, calculate an eigenvalue of the covariance matrix in the spatial domain calculated using a predetermined eigenvalue analysis method, and calculate an eigenvector corresponding to the calculated eigenvalue. Calculate and output the multi-beam signals corresponding to the number of eigenvalues by multiplying the plurality of M received signals respectively received by the antenna elements of the array antenna by the calculated eigenvectors. The received signal includes the same signal as a predetermined reference signal, and calculates a correlation coefficient between the output beam signal and the reference signal. , And outputs the beam signal having the greatest correlation coefficient as the receiving process signals. Therefore, according to the present invention, it is possible to select a beam that spatially extracts only a desired wave from a plurality of beam signals formed by eigenvectors. Thereby, for example, LMS, RL
An array antenna can be controlled by feedforward without using a weight coefficient control algorithm having a feedback loop such as S and CMA. further,
Since this method is based on eigenvalue analysis, it is possible to easily execute an array antenna control process based on radio wave environment determination, which selectively switches its own signal processing while determining a radio wave environment by eigenvalue analysis.

According to a second aspect of the present invention, there is provided an array antenna control method or apparatus for controlling an array antenna composed of a predetermined plurality of M antenna elements closely arranged in a predetermined arrangement shape. Control method or apparatus for performing, based on a plurality of M received signals respectively received by each antenna element of the array antenna, calculates a covariance matrix in the spatial domain, using a predetermined eigenvalue analysis method Eigenvalues of the spatial domain covariance matrix calculated by
The eigenvectors corresponding to the calculated eigenvalues are calculated and output, and the number of eigenvalues is calculated by multiplying the plurality of M received signals respectively received by the antenna elements of the array antenna by the calculated eigenvectors. Forming and outputting a multi-beam signal corresponding to the above, having at least transversal filters of the number of the multi-beam beam signals, inputting the output beam signal to each of the transversal filters, The weight coefficient of the transversal filter is controlled using a predetermined adaptive control algorithm, and the signals output from the transversal filters are added and output as a reception processing signal. Therefore, according to the present invention, by using multi-beams with eigenvectors, since all signal components are included in the beam outputs less than the number of antenna elements, direct signal processing is performed on the output from the antenna elements. The calculation load of the weight coefficient calculation process can be reduced as compared with the case where the calculation is performed. If a time delay occurs, and if the propagation delay in a multipath fading environment cannot be ignored, the effects of fading cannot be avoided unless signal processing in the time domain is performed. When this time delay occurs, signal processing in the time domain is performed using a TDL circuit having a transversal filter. However, since a weight coefficient exists not only in the spatial domain but also in the time domain, the weight coefficient to be controlled is Very much. In such a configuration, the reduction of the number of input signals to the weighting factor calculation processing by the configuration using multi-beams with eigenvectors can greatly reduce the number of weighting factors.

Further, according to the third or eighth aspect of the present invention, there is provided an array antenna control method or apparatus for controlling an array antenna composed of a predetermined plurality of M antenna elements closely arranged in a predetermined arrangement shape. Control method or apparatus for performing, based on a plurality of M received signals respectively received by each antenna element of the array antenna, calculates a covariance matrix in the spatial domain, using a predetermined eigenvalue analysis method The eigenvalues of the calculated covariance matrix in the spatial domain are calculated, the eigenvectors corresponding to the calculated eigenvalues are calculated and output, and a plurality of M received signals received by each antenna element of the array antenna are calculated. On the other hand, a multi-beam signal corresponding to the number of eigenvalues is formed and output by multiplying the calculated eigenvector,
Based on the calculated eigenvalue and the output beam signal, the radio wave environment of the plurality of M received signals is determined, and signal processing for the beam signal is selectively switched and executed. Therefore, according to the present invention, when the calculation amounts of the first case, the second case, and the other third case are compared, the order of the first case, the third case, and the second case is as follows. Increases the computational complexity. For example, in the case of a system that must deal with signals from a plurality of slave stations, such as a base station of a mobile communication system, it is assumed that the second case should be dealt with for all terminals. The resources of hardware such as a DSP for realizing the above are enormous. However, the propagation characteristics of signals from each terminal are different, and it is not necessary to deal with the thickest second case for all signals. In this case, it is assumed that the environment of the signal of the terminal to be handled by the base station is stochastically dispersed, and if the signal can be handled by the first or third signal processing which is simpler, the signal is determined by the determination of the environment. If the processing can be switched, the overall scale of the signal processing is stochastically reduced, and the effect of stochastically reducing the scale of hardware resources can be expected.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of an array antenna control device according to an embodiment of the present invention.

FIG. 2 is a block diagram showing a configuration of a beam forming unit 6 of FIG.

FIG. 3 is a block diagram showing a configuration of a TDL circuit 20 of FIG.

FIG. 4 is a flowchart illustrating an eigenvalue analysis process performed by the eigenvalue analysis unit 10 of FIG. 1;

FIG. 5 is a flowchart illustrating a first part of a radio wave environment determination process performed by a radio wave environment determination unit 11 of FIG. 1;

FIG. 6 is a flowchart illustrating a second part of the radio wave environment determination process performed by the radio wave environment determination unit 11 of FIG. 1;

FIG. 7 is a diagram illustrating an array antenna control apparatus shown in FIG.
Beam patterns of the beam signals B ₁ and B ₂ at 0 dB,
7 is a graph illustrating an antenna pattern according to an RLS method according to a comparative example.

FIG. 8 is a diagram illustrating a control apparatus for an array antenna in FIG.
9 is a graph showing beam patterns of beam signals B ₁ and B ₂ at dB and an antenna pattern according to an RLS method of a comparative example.

FIG. 9 is a diagram illustrating an array antenna control apparatus shown in FIG.
9 is a graph showing beam patterns of beam signals B ₁ and B ₂ at 10 dB and an antenna pattern according to an RLS method of a comparative example.

FIG. 10 is a diagram illustrating an array antenna control apparatus shown in FIG.
Beam signals B _1, B ₂ of beam patterns when the -10 dB, and is a graph showing an antenna pattern by RLS method of the comparative example.

11 is a diagram illustrating an array antenna control apparatus shown in FIG.
Beam patterns of the beam signals B ₁ and B ₂ at 0 dB,
7 is a graph illustrating an antenna pattern according to an RLS method according to a comparative example.

FIG. 12 is a diagram illustrating an array antenna control apparatus in FIG.
Beam signals B _1, B ₂ of beam patterns when the -10 dB, and is a graph showing an antenna pattern by RLS method of the comparative example.

13 is a diagram illustrating an array antenna control apparatus shown in FIG.
9 is a graph showing beam patterns of beam signals B ₁ and B ₂ at 10 dB and an antenna pattern according to an RLS method of a comparative example.

14 is a diagram illustrating an array antenna control apparatus shown in FIG.
Beam patterns of the beam signals B ₁ and B ₂ at 0 dB,
7 is a graph illustrating an antenna pattern according to an RLS method according to a comparative example.

FIG. 15 is a diagram illustrating a control apparatus for an array antenna shown in FIG.
Beam signals B _1, B ₂ of beam patterns when the -10 dB, and is a graph showing an antenna pattern by RLS method of the comparative example.

FIG. 16 is a diagram illustrating a control apparatus for an array antenna in FIG.
Beam patterns of the beam signals B ₁ and B ₂ at 0 dB,
7 is a graph illustrating an antenna pattern according to an RLS method according to a comparative example.

17 is a diagram illustrating a control apparatus for an array antenna shown in FIG.
9 is a graph showing beam patterns of beam signals B ₁ and B ₂ at dB and an antenna pattern according to an RLS method of a comparative example.

FIG. 18 is a diagram illustrating a control apparatus for an array antenna in FIG.
9 is a graph showing beam patterns of beam signals B ₁ and B ₂ at 10 dB and an antenna pattern according to an RLS method of a comparative example.

19 is a graph showing beam patterns of beam signals B ₁ , B ₂ , B ₃ , and B ₄ generated based on all eigenvectors in case 2-1 in the array antenna control device of FIG.

20 is a graph showing an antenna pattern of a reception processing signal P ₀ in case 2-1 and an RLS method antenna pattern of a comparative example in the array antenna control device of FIG. 1;

21 is a graph showing beam patterns of the beam signals B ₁ in the case 2-2 in the control unit of the array antenna of FIG. 1, the antenna pattern of the receiving process signals P _0, and an antenna pattern by RLS method of the comparative example.

FIG. 22 is a graph showing beam patterns of beam signals B ₁ , B ₂ , and B ₃ and an antenna pattern of a reception processing signal P ₀ in Case 3 in the array antenna control device of FIG.

23 is a graph showing an antenna pattern of a reception processing signal P ₀ in Case 3 and an RLS method antenna pattern of a comparative example in the array antenna control device of FIG. 1;

[Explanation of symbols]

 1-1 to 1-M: antenna element; 2-1 to 2-M: receiver; 3-1 to 3-M: complex signal generator; 4-1 to 4-M: A / D converter; -1 to 5-M: distributor, 6: beam forming unit, 7: 4 distributor, 8: reference signal generator, 10: eigenvalue analysis unit, 11: radio wave environment judgment unit, 20: TDL circuit, 21: weight Coefficient calculator, 30: Weight coefficient multiplier, 31: Weight coefficient calculator, 100: Array antenna, Step SW1, SW2, SW3 ... Switch.

 ────────────────────────────────────────────────── ─── Continuing on the front page (72) Inventor Yoshio Karasawa 2-3-2 Nishi-Shinjuku, Shinjuku-ku, Tokyo F-term within K-DDY Co., Ltd. FA31 FA32 GA01 GA06 GA08 HA02 JA07

Claims (11)

[Claims] 1. A control method for controlling an array antenna composed of a predetermined plurality of M antenna elements juxtaposed and juxtaposed in a predetermined arrangement shape, wherein said array antenna is received by each antenna element of said array antenna. Based on the plurality of M received signals, a covariance matrix in the spatial domain is calculated, an eigenvalue of the covariance matrix in the spatial domain calculated using a predetermined eigenvalue analysis method is calculated, and the eigenvalue corresponding to the calculated eigenvalue is calculated. An estimating step of calculating and outputting the eigenvectors to be performed, and multiplying the plurality of M received signals received by the respective antenna elements of the array antenna by the calculated eigenvectors to correspond to the number of eigenvalues. A forming step of forming and outputting a multi-beam signal, wherein the received signal includes the same signal as a predetermined reference signal, and the output beam Selecting a correlation coefficient between the signal and the reference signal and outputting a beam signal having a maximum correlation coefficient as a reception processing signal. 2. A control method for controlling an array antenna composed of a predetermined plurality of M antenna elements closely arranged in a predetermined arrangement shape, wherein the antenna elements are received by respective antenna elements of the array antenna. Based on the plurality of M received signals, a covariance matrix in the spatial domain is calculated, an eigenvalue of the covariance matrix in the spatial domain calculated using a predetermined eigenvalue analysis method is calculated, and the eigenvalue corresponding to the calculated eigenvalue is calculated. An estimating step of calculating and outputting the eigenvectors to be performed, and multiplying the plurality of M received signals received by the respective antenna elements of the array antenna by the calculated eigenvectors to correspond to the number of eigenvalues. A forming step of forming and outputting a multi-beam signal; and a transversal filter having at least the number of the multi-beam signals. Inputting the output beam signal to each of the transversal filters, controlling a weighting factor of each of the transversal filters using a predetermined adaptive control algorithm, and outputting a signal output from each of the transversal filters. And outputting the signal as a reception processing signal. 3. A control method for controlling an array antenna composed of a predetermined plurality of M antenna elements closely arranged in a predetermined arrangement shape, wherein the antenna elements are received by respective antenna elements of the array antenna. Based on the plurality of M received signals, a covariance matrix in the spatial domain is calculated, an eigenvalue of the covariance matrix in the spatial domain calculated using a predetermined eigenvalue analysis method is calculated, and the eigenvalue corresponding to the calculated eigenvalue is calculated. An estimating step of calculating and outputting the eigenvectors to be performed, and multiplying the plurality of M received signals received by the respective antenna elements of the array antenna by the calculated eigenvectors to correspond to the number of eigenvalues. A forming step of forming and outputting a multi-beam beam signal; a plurality of M beam signals based on the calculated eigenvalues and the output beam signal; A second control step of determining the radio wave environment of the received signal and selectively switching and executing the signal processing for the beam signal. 4. The method of controlling an array antenna according to claim 3, wherein the second control step determines a first case in which the number of the eigenvalues is 1 based on the number of the calculated eigenvalues. Multiplying the received signal by an eigenvector corresponding to the eigenvalue and outputting the one output beam signal as a reception processing signal; and a time delay of a predetermined time delay or more based on the output beam signal. Is determined, a transversal filter having at least the number of the multi-beam beam signals is input, and the output beam signals are input to the transversal filters, and the transversal filters are output. Is controlled using a predetermined adaptive control algorithm, and the signals output from the transversal filters are added. In the cases other than the first and second cases, at least a multiplier of the number of the multi-beam beam signals is provided, and the output beam signal is input to each multiplier. A weighting coefficient, which is a multiplication coefficient of each of the multipliers, is controlled using a predetermined adaptive control algorithm, and a signal output from each of the multipliers is added and output as a reception processing signal. Control method. 5. The method of controlling an array antenna according to claim 4, wherein the second control step is performed when the received signal is the same signal as a predetermined reference signal when the signal is other than the first and second cases. Calculating a correlation coefficient between the output beam signal and the reference signal, comparing the calculated correlation function with a predetermined threshold value, and calculating a beam having a correlation function not less than the threshold value. A method for controlling an array antenna, comprising outputting a beam signal having a maximum correlation coefficient as a reception processing signal when a signal is present. 6. A control device for controlling an array antenna composed of a predetermined plurality of M antenna elements which are closely arranged in a predetermined arrangement shape, and each of which is received by each antenna element of the array antenna. Based on the plurality of M received signals, a covariance matrix in the spatial domain is calculated, an eigenvalue of the covariance matrix in the spatial domain calculated using a predetermined eigenvalue analysis method is calculated, and the eigenvalue corresponding to the calculated eigenvalue is calculated. Analyzing means for calculating and outputting the eigenvectors to be processed; and multiplying the plurality of M received signals respectively received by the antenna elements of the array antenna by the eigenvectors calculated by the analyzing means, thereby obtaining the number of eigenvalues. Forming means for forming and outputting a multi-beam signal corresponding to the above, the received signal includes the same signal as a predetermined reference signal, and the forming means Selecting means for calculating a correlation coefficient between the beam signal output from the reference signal and the reference signal, and outputting a beam signal having the maximum correlation coefficient as a reception processing signal. Control device. 7. A control device for controlling an array antenna composed of a predetermined plurality of M antenna elements juxtaposed and juxtaposed in a predetermined arrangement shape, wherein each of the antenna elements is received by each antenna element of the array antenna. Based on the plurality of M received signals, a covariance matrix in the spatial domain is calculated, an eigenvalue of the covariance matrix in the spatial domain calculated using a predetermined eigenvalue analysis method is calculated, and the eigenvalue corresponding to the calculated eigenvalue is calculated. Analyzing means for calculating and outputting the eigenvectors to be processed; and multiplying the plurality of M received signals respectively received by the antenna elements of the array antenna by the eigenvectors calculated by the analyzing means, thereby obtaining the number of eigenvalues. Forming means for forming and outputting a multi-beam signal corresponding to the above, and transversal of at least the number of the multi-beam signal Having a filter, inputting a beam signal output from the forming means to each of the transversal filters, controlling a weighting factor of each of the transversal filters using a predetermined adaptive control algorithm, and A first control means for adding the output signals and outputting the sum as a reception processing signal. 8. A control device for controlling an array antenna composed of a predetermined plurality of M antenna elements closely arranged in a predetermined arrangement shape, wherein the antenna elements are received by respective antenna elements of the array antenna. Based on the plurality of M received signals, a covariance matrix in the spatial domain is calculated, an eigenvalue of the covariance matrix in the spatial domain calculated using a predetermined eigenvalue analysis method is calculated, and the eigenvalue corresponding to the calculated eigenvalue is calculated. Analyzing means for calculating and outputting the eigenvectors to be processed; and multiplying the plurality of M received signals respectively received by the antenna elements of the array antenna by the eigenvectors calculated by the analyzing means, thereby obtaining the number of eigenvalues. Forming means for forming and outputting a multi-beam signal corresponding to the above, eigenvalues calculated by the analyzing means, and output from the forming means And a second control means for judging the radio wave environment of the plurality of M received signals based on the beam signal and selectively switching and executing the signal processing for the beam signal. Antenna control device. 9. The control device for an array antenna according to claim 8, wherein said second control means is configured such that the number of eigenvalues is one based on the number of eigenvalues calculated by said analysis means. And outputs one beam signal output from the forming means as a reception processing signal by multiplying the received signal by an eigenvector corresponding to the eigenvalue, based on the beam signal output from the forming means. A second case in which a time delay equal to or more than a predetermined time delay occurs is determined, and at least the number of transversal filters equal to the number of the multi-beam beam signals is provided. It is input to the transversal filter, and the weight coefficient of each transversal filter is controlled using a predetermined adaptive control algorithm. Adding a signal output from the transversal filter and outputting the sum as a reception processing signal; and in a case other than the first and second cases, a multiplier having at least the number of the beam signals of the multi-beams, The beam signal output from is input to each multiplier,
A weighting coefficient, which is a multiplication coefficient of each of the multipliers, is controlled by using a predetermined adaptive control algorithm, and a signal output from each of the multipliers is added and output as a reception processing signal. Control device. 10. The control device for an array antenna according to claim 9, wherein the second control means outputs the same signal as a predetermined reference signal when the reception signal is other than the first and second cases. Calculating a correlation coefficient between the beam signal output from the forming means and the reference signal, comparing the calculated correlation function with a predetermined threshold, and calculating a correlation function equal to or greater than the threshold. A beam signal having a maximum correlation coefficient is output as a reception processing signal when a beam signal having the following is present. 11. A recording medium having recorded thereon an array antenna control program, wherein the array antenna control program includes the array antenna control method according to any one of claims 1 to 5.