JP2000111630A - Radio wave incoming direction inferring method and its device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To accurately infer the incoming directions of incoming signals at all times by eliminating an offset in time delay of the incoming signals by a specific method. SOLUTION: Signal radio waves caught by N antennas 11,... not less than 3 disposed close to each other, are converted into N pieces of normalized signals 21,..., N auto- correlation matrices 31,... and M cross-correlation matrices 6 not less than 2 are produced from the N normalized signals, and an equalized auto-correlation matrix 4 is produced by equalizing the N auto-correlation matrices 31,.... K incoming signals and K incoming-signal time delays 7 are produced by signal-separation processing of the equalized auto-correlation matrix 4 with high time resolution 5, K incoming signals are produced by M kinds by using the M cross-correlation matrices, the number of incoming signals, and K corrected incoming signal time delays. The incoming directions of the incoming K signals are inferred based on phases of the K incoming signals produced in M kinds. The K corrected incoming signal time delays are obtained by evaluating errors based on the amplitude of the K incoming signals produced in M kinds and by correcting the K incoming-signal time delays so as to minimize the errors.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and apparatus for estimating the direction of arrival of radio waves, and more particularly to N (where N is 3).
Radio signal arrival direction estimation that can stably and accurately estimate the direction of arrival of each of multiple signal components arriving close in time in signal radio waves captured by the above (integer) antennas A method and an apparatus therefor.

[0002]

2. Description of the Related Art Conventionally, there has been known a moving object tracking method using radio waves in order to track the current position of a moving object that appropriately moves within a predetermined area. In this mobile tracking method, a mobile carries a transmitter for transmitting signal radio waves, and a base station having a plurality of antennas for receiving signal radio waves radiated from the transmitter is arranged near a required area. It is. The base station extracts, as the main signal component, the signal wave component arriving earliest among the plurality of signal wave arriving close in time, that is, the signal component having the shortest delay time among the arriving signal wave components. By determining the arrival direction of the main signal component, the position of the source of the signal radio wave can be specified.
If the source of the signal radio wave is a transmitter carried by the mobile, the current position of the mobile can be tracked.

In this case, in such a mobile object tracking system, various signal modulation systems can be used as a signal modulation system transmitted and received between a transmitter carried by the mobile object and a base station. Is one of the signal modulation schemes
There is a spread modulation method using an N (pseudo random noise) code.

FIG. 9 is a block diagram showing an example of a configuration of a known radio wave direction-of-arrival estimation device used in the mobile object tracking system, and shows an example of a configuration in which four antennas are used. is there.

[0005] As shown in FIG. 9, the known DOA estimation apparatus includes four antennas 1 ₁ to 1 _4, and four normalized signal generating unit 2 ₁ to 2 _4, four self a correlation matrix generating section 3 ₁ to 3 _4, averaged autocorrelation matrix generator 4
And a high-resolution signal processing unit 5 and a cross-correlation matrix generation unit 6
, An arrival signal generation unit 7, an arrival angle estimation unit 8, and a signal output terminal 9.

[0006] Then, the normalized signal generator 2 ₁ has an input connected to the antenna 1 _1, the output is the autocorrelation matrix generator 3
₁ input and a corresponding input of the cross-correlation matrix generator 6. The input of the normalized signal generator 2 ₂ is the antenna 1 ₂
Is connected to the output is connected to a corresponding input of the input autocorrelation matrix generation unit 3 ₂ and the cross-correlation matrix generation unit 6. Normalized signal generating unit 2 ₃ has an input connected to the antenna 1 _3,
Output is connected to the corresponding input of the input autocorrelation matrix generator 3 ₃ and the cross-correlation matrix generation unit 6. Normalized signal generating unit 2 ₄ has an input connected to the antenna 1 _4, the output is connected to the corresponding input of the input autocorrelation matrix generator 3 ₄ and the cross-correlation matrix generation unit 6. Averaging autocorrelation matrix generator 4
It has four inputs connected to respective outputs of the autocorrelation matrix generation unit 3 ₁ to 3 _4, output high resolution signal processing unit 5
Connected to the input of The arriving signal generator 7 has six inputs connected to corresponding outputs of the cross-correlation matrix generator 6 and two inputs having two corresponding outputs (the number of arriving signals output, Signal delay time output), and the six outputs are respectively connected to the corresponding six inputs of the angle-of-arrival estimation unit 8. Arrival angle estimation unit 8
Has an output connected to the signal output terminal 9. In this case, the high-resolution signal processing unit 5 is, for example, a high-time-resolution signal processing device using a multiple signal classification (MUSIC) method.

[0007] FIG. 2 is a block diagram showing an example of the illustrated configuration of the normalized signal generating unit 2 ₁ 2 ₄ 9, used in the radio wave arrival direction estimating apparatus of the present invention described later It has the same configuration as the configuration of each normalized signal generation unit.

As shown in FIG. 2, each normalized signal generator 2
_{1-2 4,} of the same configuration, a receiving portion 10, the first
Fourier transform unit 11, first filter means 12, reference signal generation unit 13, second Fourier transform unit 14, second filter means 15, division unit 16, noise removal processing unit 17
, A received signal input terminal 18 and a normalized signal output terminal 19
It consists of In this case, the receiving unit 10 includes a baseband (hereinafter, referred to as BB) signal generating unit 20, an analog-digital (hereinafter, referred to as A / D) converting unit 21, and a memory 22, and the noise removing processing unit 17 includes: , Frequency window multiplying unit 23, Fourier transform unit 24, noise removing unit 25
And an inverse Fourier transform unit 26 and a frequency window dividing unit 27.

The receiving unit 10 has an input connected to the received signal input terminal 18 and an output connected to the first Fourier transform unit 11.
Connected to the input of The input of the first filter unit 12 is connected to the output of the first Fourier transform unit 11, and the output is connected to the first input of the division unit 16. Reference signal generator 13
Has an output connected to the input of the second Fourier transform unit 14. The second filter means 15 has an input connected to the output of the second Fourier transform unit 14 and an output connected to the second input of the divider 16. The input of the noise removal processing unit 17 is
6 and the output is connected to the normalized signal output terminal 19.

In the receiving unit 10, the input of the BB signal generating unit 20 is connected to the input of the receiving unit 10, and the output is connected to the input of the A / D converting unit 21. The memory 22
The input is connected to the output of the A / D converter 21, and the output is connected to the output of the receiver 10. In the noise removal processing unit 17, the input of the frequency window multiplication unit 23 is
, And the output is connected to the input of the Fourier transform unit 24. The input of the noise removing unit 25 is connected to the output of the Fourier transform unit 24, and the output is connected to the input of the inverse Fourier transform unit 26. The frequency window divider 27 has an input connected to the output of the inverse Fourier transformer 26 and an output connected to the output of the noise removal processor 17.

[0011] followed, FIG. 10 is a block diagram showing an example of the four incoming state of the configuration state and signal wave antenna 1 ₁ to 1 ₄ shown in FIG.

[0012] As shown in FIG. 10, the four antennas 1 ₁ to 1 _4, the X axis extending through the reference point O, a Y-axis extending in a direction orthogonal to the X-axis through the reference point O A radius r ′ about a reference point O on a plane
r ′ is selected to be 0.36λ, where λ is the wavelength of the signal radio wave), and is arranged at equal angular intervals on the circumference of the circle to form a four-element circular array antenna.

In this case, the signal wave is a propagation wave including a signal spread-modulated with a PN (pseudo-random noise) code, and the first wave is parallel to the plane and 5 ° with respect to the X axis.
Arriving from an azimuth direction of 0 °, the second wave is parallel to the plane and arriving from an azimuth direction of 0 ° with respect to the X axis, and the second wave is relative to the first wave. In this example, the delay is 0.25 chip (Tc).

FIG. 11 is a signal waveform diagram showing an example of a band-limited spread modulation signal in which transmission data (period T) is spread-modulated by a PN code on the transmission side and the band is further restricted.

[0015] Figure 12 is one of the four normalized signal generating unit 2 ₁ to 2 _4, for example, normalized signal generating unit 2 ₁ B
FIG. 3 is a signal waveform diagram illustrating an example of a BB spread modulation signal output from a B signal generation unit 20. In the figure, a curve a shown by a solid line is a signal of an in-phase channel, and a curve b shown by a dotted line is a signal of an orthogonal channel.

[0016] Figure 13 is a signal waveform diagram showing an example of a frequency domain reference signal outputted from the normalization signal generator 2 ₁ of the second Fourier transform unit 14. In the figure, the portion of the frequency bandwidth A is a frequency region extracted by the second filter unit 15.

[0017] FIG. 14 is a signal waveform diagram showing an example of the frequency-domain received signal outputted from the first Fourier transform unit 11 of the normalized signal generating unit 2 _1. In the figure, the portion of the frequency bandwidth B is a frequency region extracted by the first filter unit 12.

FIG. 15 shows a normalized signal including a noise component input to the noise removal processing unit 17 and the noise removal processing unit 1.
FIG. 7 is a signal waveform diagram illustrating each example of a normalized signal from which a noise component output from FIG. 7 has been removed. In the figure, a curve a shown by a dotted line is a normalized signal containing a noise component, and a curve b shown by a solid line is a normalized signal from which the noise component has been removed.

FIG. 16 is a signal waveform diagram showing an example of the correlation metric signal output from the high-resolution signal processing section 5.

The operation of the known radio wave direction-of-arrival estimation device having the above configuration will be described with reference to FIG. 2, and FIGS. 9 to 16.

In this case, the signal radio wave transmitted from the transmitting side (not shown) is obtained by converting the transmission data into a primary modulation signal formed by performing modulation such as PSK (phase shift keying).
A code is multiplied to form a spread modulated signal, and the band of the spread modulated signal is further restricted to form a band limited spread modulated signal as shown in FIG. It is converted and transmitted.

Further, the four operation normalized signal generating unit 2 ₁ to 2 ₄ in the description of the operation of the radio wave arrival direction estimating apparatus, respectively, four antennas 1 ₁ to signal 1 ₄ receives The same operation is performed only with a different signal waveform handled according to the radio wave. Therefore, where the description of the operation of the normalization signal generator 2 by _one of the operation was described, other normalized signal generating unit 2 ₂ to 2 ₄ are omitted.

When the spread-modulated signal as shown in FIG. 11 is frequency-converted and transmitted from the transmitting side,
As shown in FIG. 2, the _four antennas 11 to ₁₄ respectively capture signal radio waves including the first wave and the second wave.

The signal wave captured by the four antennas 1 ₁ to 1 ₄ are supplied to the corresponding reception unit 10 of the four normalized signal generating unit 2 ₁ to 2 ₄ as a received signal. At this time, in the normalized signal generating unit 2 _1, it is carried out the operation as described below.

First, the BB signal generator 20 of the receiver 10
Reproduces a BB analog spread modulation signal as shown in FIG. 12, and the A / D converter 21 converts the BB analog spread modulation signal into a BB digital spread modulation signal,
Output to 2.

Next, the first Fourier transformer 11 Fourier-transforms the BB digital spread-spectrum modulated signal into a frequency-domain received signal having a signal waveform as shown in FIG.

Next, the first filter means 12 limits the frequency band of the frequency domain received signal so as to fall within the range of the bandwidth B shown in FIG.

On the other hand, the reference signal generator 13 generates a band-limited PN code as shown in FIG. 11 as a reference signal. This band-limited PN code is obtained by band-limiting the same code as the PN code used when the transmission data is spread-modulated, and is one transmission data period (1T).
Is the same as that of the band-limited spread modulation signal.

Next, the second Fourier transformer 14 Fourier-transforms the reference signal to convert it into a frequency-domain reference signal having a signal waveform as shown in FIG.

Next, the second filter means 15 limits the frequency band of the frequency domain reference signal so as to fall within the range of the bandwidth A shown in FIG.
In this case, the bandwidth A and the bandwidth B are selected so as to have the same bandwidth and both become (1 / Tc).

The division unit 16 calculates a ratio based on the frequency domain reference signal to the frequency domain received signal for each frequency, and obtains a normal signal including a noise component such as a curve a shown by a dotted line in FIG. Generates an activation signal.

Next, the noise processing and removing unit 17 multiplies the normalized signal including the noise component by the frequency window signal at the frequency window multiplying unit 23 and multiplies the normalized signal including the frequency window signal by the Fourier transform unit 24 to obtain the normalized signal. Fourier transform is performed to convert the signal into a time domain signal, a noise removing unit 25 removes a noise component in the time domain signal, and an inverse Fourier transform unit 26 performs a Fourier transform on the time domain signal to re-convert it into a frequency domain normalized signal. The frequency window divider 27 divides the normalized signal by the frequency window signal,
5, a normalized signal X ₁ containing no noise component such as a curve b shown by a solid line is formed and output from the normalized signal output terminal 19.

The above description is with respect to operation of the normalized signal generating unit 2 _1, another similar operation also in the normalized signal generating unit 2 ₂ to 2 ₄ is performed, does not include a noise component, respectively normalized signal X ₂ to X ₄ are formed.

[0034] Subsequently, in FIG. 9, four autocorrelation matrix generation unit 3 ₁ to 3 _4, corresponding four normalized signal generating unit 2 ₁ to 2 ₄ normalized signal X ₁ to X ₄ which occurs And generates a corresponding autocorrelation matrix. As an example, the normalized signal generating unit 2 ₁ generates an autocorrelation matrix by performing signal processing such as calculating the product of the complex conjugate signal of the normalized signal X ₁ and the normalized signal X ₁ for each frequency component . Generating an autocorrelation matrix and the other normalized signal generating unit 2 ₂ to 2 ₄ similarly.

The averaging auto-correlation matrix generator 4 obtains the four auto-correlation matrices generated by the four auto-correlation matrix generators 3 _{1 to} 3 ₄ by averaging for each element of the auto-correlation matrix. The averaged autocorrelation matrix is generated and supplied to the high-resolution signal processing unit 5.

The high-resolution signal processing unit 5 analyzes the averaged autocorrelation matrix with high time resolution using the MUSIC method,
A correlation metric signal as shown in FIG. 16 is formed, and the number of incoming signals is determined by counting the number of peak metric values having a correlation metric value equal to or greater than a predetermined value based on the obtained correlation metric signal. In addition, the delay time at the time of taking the peak measurement value is measured, each arrival signal delay time is determined, and the number of arrival signals and each arrival signal delay time are supplied to the arrival signal generation unit 7. In the case of the example shown in FIG. 16, the number of arriving signals (the number of peak measurement values) is 2, and the arriving signal delay time is 16.05 Tc for the first wave,
The second wave is 16.36 Tc, and the relative signal delay time of the second wave with respect to the first wave is 0.31 Tc.

Further, the cross-correlation matrix generation unit 6 includes four normalized signal generating unit 2 ₁ to normalized signal X ₁ 2 ₄ occurs
To undergo X _4, 4 pieces of relative six combinations of different first and second normalized signals in the normalized signal X ₁ to X _4, first normalized signal and the second normal 6 obtained by calculating the product of the normalized signal and the complex conjugate signal for each frequency component
Generate two cross-correlation matrices. Here, since the number of any two combinations of the _four normalized signals X _{1 to} X ₄ is 6, six cross-correlation matrices are generated.

The arriving signal generator 7 calculates the six cross-correlation matrices generated by the cross-correlation matrix generator 6 and the number (K) of arriving signals generated by the high-resolution signal processor 5 and the K arriving signal delay times. Then, the number of arriving signals corresponding to the number of arriving signals K is calculated for each cross-correlation matrix, six types of arriving signals are generated for each of the K arriving signals, and supplied to the angle-of-arrival estimating unit 8. In this case, the six types of arriving signals include information on the amplitude of the arriving signal and information on the zenith angle and the azimuth of the arriving signal for each of the K signals. In the example shown in FIG. 16, K = 2.

The arrival angle estimating unit 8 is configured such that the arrival signal
Receiving the six incoming signals generated for each of the three incoming signals, comparing the phases of the six incoming signals with each other,
The arrival direction (the arrival zenith angle and the arrival azimuth angle) of each arrival signal is calculated, and the calculation result is output from the signal output terminal 9. In this case, as in the example shown in FIG. 16, when it is estimated that the number of arriving signals is two (K = 2), if the arrival angles of these two arriving signals are estimated, the first The arrival angle of the wave is 50.7 °, and the arrival angle of the second wave is 6.2.
°.

As described above, the known radio wave arrival direction estimating apparatus extracts the arrival signals such as the first wave and the second wave, and
The direction of arrival of these incoming signals can be determined.

[0041]

The known radio wave direction-of-arrival estimating apparatus can not only extract a signal component arriving close in time but also know the direction of arrival relatively accurately. but since the normalized signal generating unit 2 ₁ to 2 ₄ is contained noise components normalized signal X ₁ to X ₄ occurs, the high resolution signal based on such a normalized signal X ₁ to X ₄ When the arriving signal delay time is obtained by the processing unit 5, an offset component is added to the arriving signal delay time under the influence of noise components in the normalized signals X _{1 to} X _4. The incoming signal cannot be generated.

As a result, the known radio wave direction-of-arrival estimation device may not be able to correctly determine the direction of arrival of the main signal component, or may correctly determine the direction of arrival of signal components other than the main signal component. There is a problem that it may not be possible.

[0043] Most, the known DOA estimation apparatus, a noise removing unit 17 is arranged to normalize the signal generating unit 2 ₁ a 2 _4, the noise of the noise component of the normalized signal X ₁ through in X ₄ The noise is removed by the removal processing unit 17, but even if such a noise removal processing unit 17 is arranged, the MU
There is an essential problem that the signal delay time obtained by high-resolution signal processing using a technique such as the SIC method includes an offset.

Therefore, the incoming signal generator 7 cannot always generate an accurate incoming signal.

As a result, the known radio wave direction-of-arrival estimating apparatus still has the noise removal processing unit 17 disposed therein.
There is a problem that a case where the arrival direction of the main signal component cannot be correctly obtained or a case where the direction of arrival of other signal components other than the main signal component cannot be correctly obtained may occur.

For example, in the case of the above-described specific example, the arrival azimuth of the second wave is originally 0 °,
The azimuth of arrival of the second wave estimated by the known radio wave arrival direction estimating device becomes 6.2 °, and a large azimuth error is introduced.

An object of the present invention is to solve such a problem. An object of the present invention is to eliminate an offset component added to an arrival signal delay time, to always generate an accurate arrival signal in an arrival angle estimation unit, It is an object of the present invention to provide a radio wave arrival direction estimating method and an apparatus thereof that enable accurate estimation of the radio wave direction.

[0048]

In order to achieve the above object, a radio wave direction-of-arrival estimation method according to the present invention comprises a method of estimating N autocorrelation matrices and M
Generating (an integer of 2 or more) cross-correlation matrices, generating an averaged auto-correlation matrix by averaging N auto-correlation matrices,
The averaged autocorrelation matrix is subjected to signal separation processing with high time resolution to generate the number of arriving signals (K) and K arriving signal delay times, and to obtain M cross-correlation matrices, arriving signal numbers and K arriving arrivals. The K delay signals are generated using the signal delay time and the M arrival signals are generated, and the arrival directions of the K arrival signals are estimated based on the phases of the K arrival signals generated in each of the M delay signals. The corrected arrival signal delay times are obtained by performing error evaluation based on the amplitudes of the K arrival signals generated in each of the M types and correcting the K arrival signal delay times so that the error is minimized. 1 means is provided.

In order to achieve the above object, a radio wave direction of arrival estimating apparatus according to the present invention comprises an N normalized signal generator for generating N normalized signals, and an N normalized signal generator. N autocorrelation matrix generators for generating an autocorrelation matrix, an averaged autocorrelation matrix generator for generating an averaged autocorrelation matrix of the N autocorrelation matrices, and an averaged autocorrelation matrix with high time resolution A high-resolution signal processing unit that generates the number of arriving signals (K) and K arriving signal delays by performing signal separation processing, and generates a cross-correlation matrix that generates M cross-correlation matrices from N normalized signals , An incoming signal generator for generating K incoming signals each using the M cross-correlation matrices, the number of incoming signals, and the K corrected incoming signal delay times, and an output from the high-resolution signal processing unit. K incoming signal delay times are An error is evaluated based on the amplitudes of the generated K incoming signals, and K corrected incoming signal delay times obtained by correcting the K incoming signal delay times so as to minimize the error are supplied to the incoming signal generating unit. A second means comprising a delay time correction unit and an arrival angle estimation unit for estimating the arrival directions of the K arrival signals based on the phases of the K arrival signals generated in each of the M cases.

According to the first means and the second means,
Supplying the number of K arriving signals and the K arriving signal delay times output from the high resolution signal processing unit to the arriving signal generation unit;
When the K arriving signals are generated by the arriving signal generator, an error evaluation is performed based on the amplitudes of the M arriving K signals, and the K arriving signal delays are set so that the error is minimized. The time is corrected and converted to each corrected arrival signal delay time, and then supplied to the arrival signal generation unit. The arrival signal generation unit uses the corrected arrival signal delay time to generate K signals for each cross-correlation matrix unit. M incoming signals are generated. For this reason, the offset of the arrival signal delay time due to the noise component in the N normalized signals output from the normalized signal generation unit and the offset of the K arrival signal delay times output from the high resolution signal processing unit Is substantially eliminated by correcting the K arriving signal delay times to the K corrected arriving signal delay times.
An error component is not included in the K arriving signals generated as described above, and the arriving angle estimation unit can always accurately estimate the arriving directions of the K arriving signals.

[0051]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS In a first embodiment of the present invention, a radio wave arrival direction estimating method is described in which a signal radio wave captured by three (3) or more integer (N) antennas arranged in close proximity is converted into N normalized signal radio waves. To generate N autocorrelation matrices and 2 or more integer (M) cross-correlation matrices from the N normalized signals,
An averaged autocorrelation matrix is generated by averaging the N autocorrelation matrices, and the averaged autocorrelation matrix is subjected to signal separation processing with high time resolution to determine the number of arriving signals (K) and the K arriving signal delay times. Generated, using the M cross-correlation matrices, the number of arriving signals, and the K corrected arriving signal delay times, to generate K arriving signals for each of the M arriving signals. The direction of arrival of the K arriving signals is estimated based on the phase. The K corrected arriving signal delay times are evaluated based on the amplitude of each of the M arriving K incoming signals, Is corrected so that K is minimized.

In one example of the first embodiment of the present invention, the radio wave arrival direction estimating method comprises the steps of converting a normalized signal into a baseband signal generated by frequency-converting each signal radio wave and a reference signal serving as a reference. It was obtained by determining the ratio.

In another example of the first embodiment of the present invention, the radio wave arrival direction estimating method is a method in which a normalized signal is obtained by obtaining a ratio between a baseband signal and a reference signal in a frequency domain. is there.

Still another one of the first embodiment of the present invention
In one example, the radio wave direction-of-arrival estimation method includes a baseband signal including a signal spread-modulated with a predetermined code, and a reference signal having a high correlation with the predetermined code.

Still another one of the first embodiment of the present invention.
In one example, the direction-of-arrival estimation method is:

[0056]

(Equation 1)

This is obtained by:

Still another one of the first embodiment of the present invention.
In one example, the radio wave direction-of-arrival estimation method uses a maximum of {N (N-1) / 2} cross-correlation matrices that can generate M cross-correlation matrices by combining N normalized signals. , And M cross-correlation matrices are selected, and the cross-correlation matrices corresponding to the normalized signals of two antennas that are arranged relatively far apart from each other among the N antennas It consists of a matrix.

In the second embodiment of the present invention, the radio wave direction-of-arrival estimating apparatus is an integer (N) of 3 or more closely arranged.
Antennas, N normalized signal generators that generate N normalized signals by receiving signal radio waves captured by the N antennas, and generate autocorrelation matrices from the N normalized signals, respectively. N autocorrelation matrix generators, an averaged autocorrelation matrix generator for generating an averaged autocorrelation matrix of the N autocorrelation matrices, and signal separation processing of the averaged autocorrelation matrix with high time resolution. A high-resolution signal processing unit for generating the number of signals (K) and K arriving signal delay times, and a cross-correlation matrix generating two or more integer (M) cross-correlation matrices from N normalized signals Part, M cross-correlation matrices, the number of arriving signals, and K corrected arriving signal delay times,
And a K-ary signal delay time output from the high-resolution signal processing unit, wherein an error is generated based on the amplitude of each of the M types of K-ary signals. A delay time correction unit that performs evaluation and corrects the K arrival signal delay times to minimize the error so as to minimize the error, and supplies the K arrival signal delay times to the arrival signal generation unit; And an arrival angle estimator for estimating the arrival directions of the K arrival signals based on the phases of the arrival signals.

In one example of the second embodiment of the present invention, in the radio wave direction-of-arrival estimation apparatus, N normalized signal generators perform frequency conversion on the reception signal of each signal radio wave captured by an antenna. A baseband signal generating means for generating a baseband signal, a reference signal generating means for generating a reference signal serving as a reference, and a dividing means for generating a normalized signal by taking a ratio of the baseband signal to the reference signal. It is provided with.

In another example of the second embodiment of the present invention, in the radio wave direction-of-arrival estimation apparatus, N normalized signal generators each include a dividing means for dividing a baseband signal and a reference signal by a dividing means. The ratio is taken in the frequency domain to generate a normalized signal.

Still another one of the second embodiment of the present invention
In one example, the radio wave direction-of-arrival estimation apparatus includes a baseband signal generation unit that generates a baseband signal including a signal that is spread-modulated with a predetermined code, and a reference signal generation unit that has a high correlation with the predetermined code. Is to occur.

Still another one of the second embodiment of the present invention.
In one example, the radio wave direction-of-arrival estimation device includes a delay time correction unit,

[0064]

(Equation 1)

Thus, the delay time of the incoming signal is corrected, and the delay time of each incoming signal is corrected.

Still another one of the second embodiment of the present invention.
In one example, the radio wave direction-of-arrival estimation apparatus generates a cross-correlation matrix generator from a maximum of {N (N-1) / 2} cross-correlation matrices that can be generated by a combination of N normalized signals. It is obtained by selecting M cross-correlation matrices, and the selected M cross-correlation matrices correspond to normalized signals of two antennas which are arranged relatively far apart from N antennas It was done.

According to these embodiments of the present invention, the number of K arriving signals and the K arriving signal delay times output from the high-resolution signal processing section are supplied to the arriving signal generating section, and the arriving signal generating section is provided. In the section, K for each cross-correlation matrix unit
When M kinds of arriving signals are generated, the K arriving signal delay times are sequentially corrected in accordance with predetermined steps based on the amplitudes of the K arriving signals generated in the M ways so that their errors are minimized. After being converted to the corresponding corrected arrival signal delay times, the signals are supplied to the arrival signal generator, and the arrival signal generator uses the corrected arrival signal delay time to generate M kinds of K arrival signals. Therefore, the offset of the arrival signal delay time caused by the noise component in the N normalized signals output from the N normalized signal generation units and the delay of each arrival signal output from the high-resolution signal processing unit The time offset gives K incoming signal delay times to K
Correction to the number of corrected arrival signal delay times can be substantially eliminated. As a result, there are no error components included in the K arriving signal components generated from the M arriving signal generators, and the arrival angle estimator always accurately estimates the directions of arrival of the K arriving signals. Can be.

[0068]

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

FIG. 1 is a block diagram showing the configuration of the first embodiment of the radio wave direction of arrival estimating apparatus according to the present invention.

As shown in FIG. 1, when the radio wave direction-of-arrival estimation device of the first embodiment is compared with the known radio wave direction-of-arrival estimation device shown in FIG. Only the configuration is different in that a delay time correction unit 26 not used in the first embodiment is used. In addition, between the radio wave arrival direction estimating device of the first embodiment and the known radio wave arrival direction estimating device. There is no difference in configuration.

In this case, the delay time correction unit 26 used in the radio wave direction-of-arrival estimation apparatus of this embodiment has two inputs, six inputs, and one output, and has two outputs. Inputs are respectively connected to the number-of-arrival-signals output of the high-resolution signal processing unit 5 and each signal-delay-time output. Six inputs are respectively connected to the six outputs of the arriving-signal generation unit 7, and one output is received. It is connected to the delay time input of the signal generator 7.

In FIG. 1, the same components as those shown in FIG.
Since the configuration of the radio wave direction-of-arrival estimation apparatus according to the present embodiment shown in (1) is the same as that described above, further detailed description will be omitted.

[0073] Also, FIG. 2 is a block diagram showing an example of each normalized signal generating unit 2 ₁ 2 ₄ configuration illustrated in FIG. 1, a known radio wave arrival direction estimation illustrated in FIG. 9 it is of the same structure as the normalization signal generating unit 2 ₁ to 2 ₄ in the device.

[0074] for the constitution of the normalized signal generating unit 2 ₁ to 2 ₄ in the present embodiment shown in FIG. 2, the overlap with the previous description, is omitted no more detailed description.

Next, FIG. 3 is a diagram corresponding to each change in the arrival signal delay time of the first wave and the second wave of the arrival wave, which is executed by the delay time correction unit 26 of the arrival direction estimating apparatus of this embodiment. FIG. 7 is a contour diagram showing an example of the change of the accumulated normalized squared error, in which two axes in the plane direction indicate the arrival signal delay time of the first wave and the second wave in time unit of one chip (Tc) of the PN code. The vertical axis indicates the cumulative normalized squared error.

The operation of the radio wave direction-of-arrival estimation apparatus according to the first embodiment having the above configuration will be described.

However, in this description of the operation, the part where the same operation as the operation of the known radio wave arrival direction estimating apparatus is performed will not be described again, because the description is duplicated.

In the radio wave arrival direction estimating apparatus of the present embodiment, the signal radio wave transmitted from the transmitting side is the same as the signal radio wave used in the known radio wave arrival direction estimating apparatus.

[0079] The arrangement of four antennas 1 ₁ to 1 ₄ in a radio wave arrival direction estimating apparatus of the present embodiment, as shown in FIG. 10, four antennas in the known radio wave arrival direction estimating apparatus 1 ₁ or is the same as the arrangement of the 1 _4. In the radio wave direction-of-arrival estimation device of this embodiment, the first
The wave is parallel to the plane of placement of the _four antennas 11 to ₁₄ ,
The second wave arrives from an azimuth direction of 50 ° with respect to the X axis, and the second wave is parallel to the arrangement surface of the _four antennas 11 to ₁₄ and has an azimuth angle of 0 ° with respect to the X axis. Coming from a direction,
It is assumed that the power of the first wave and the power of the second wave are equal, and the power of the second wave is relatively delayed by 0.25 chip (Tc) with respect to the first wave.

[0080] In addition, four operation normalized signal generating unit 2 ₁ to 2 _4, four operations of the autocorrelation matrix generation unit 3 ₁ to 3 _4, the operation of the averaging autocorrelation matrix generation unit 4, a high resolution The operation of the signal processing unit 5, the operation of the cross-correlation matrix generation unit 6, the operation of the arrival signal generation unit 7, and the operation of the arrival angle estimation unit 8
Both the four operations of normalized signal generating unit 2 ₁ to 2 ₄ in the known DOA estimation apparatus, four autocorrelation matrix generation unit 3 operation of the ₁ to 3 _4, averaged autocorrelation matrix generation unit 4, the operation of the high-resolution signal processing unit 5, the operation of the cross-correlation matrix generation unit 6, the operation of the arrival signal generation unit 7, and the operation of the arrival angle estimation unit 8.

The high-resolution signal processing section 5 has the M
The signal processing is performed using the USIC method, and the number of incoming signals and each signal delay time are output.
Output six cross-correlation matrices. The arrival signal generator 7 generates six kinds of arrival signals, and the arrival angle estimator 8
The phases of the six arriving signals are compared with each other, the first wave having the shortest delay time and the second wave following the first wave are extracted, and the arrival directions (the zenith angle and the zenith angle of the extracted first wave and the second wave) are extracted. ), And outputs the calculation result from the signal output terminal 9.

In the radio wave direction-of-arrival estimation apparatus of the first embodiment, since _four antennas 11 to ₁₄ are used, the combinations of the cross-correlation matrices output from the cross-correlation matrix generator 6 are all in all. There are six types. At this time, the delay time correction unit 26 corrects each incoming signal delay time according to the following equation (1) so that the accumulated normalized squared error is minimized, and generates a corrected incoming signal delay time. It is supplied to the generator 7.

[0083]

(Equation 2)

[0084] The antenna 1 ₁ to 1 ₄ are four (N =
4), the combination of (i, j) is (2,
1), (3, 1), (4, 1), (3, 2), (4,
6) (M = 6) of (2) and (4, 3).

The radio wave arrival direction estimating apparatus of the first embodiment has a configuration in which a delay time correction unit 26 is added to the known radio wave arrival direction estimating apparatus shown in FIG. Since the arriving signal delay time is corrected to each corrected arriving signal delay time, the estimation accuracy of the arriving direction of each arriving signal can be improved as compared with a known radio wave arriving direction estimating apparatus.

The state of obtaining the corrected arrival signal delay time by correcting the arrival signal delay time will now be described with reference to the contour diagram shown in FIG.

In FIG. 3, one side of each lattice is a time step for adjusting the delay time. In this embodiment, the time step is selected to be 0.0125 chip (Tc).

First, the arrival signal delay time of the first wave output from the high resolution signal processing unit 5 is 16.05 chips (T
c) Since the arrival signal delay time of the second wave is 16.36 chips (Tc), the cumulative normalized squared error is at the position of the black circle in FIG. At this time, the cumulative normalized squared error at that position is not minimized.

Therefore, the delay time correction unit 26 changes the arrival signal delay time of the first wave and the second wave in the time step units, and adjusts the arrival signal delay time obtained for each change to the corrected arrival signal delay time. And supplies it to the incoming signal generator 7. At this time, the delay time correction unit 26 uses the arriving signals output from the arriving signal generation unit 7 for each of the six types of arriving signals, and again performs the error evaluation according to the equation (1).
That is, the signal delay times of the first wave and the second wave are changed in units of the time steps in a direction in which the accumulated normalized squared error is minimized.

Such a series of operations is continued until the cumulative normalized square error relating to the amplitude of the arriving signal is minimized. When the cumulative normalized square error finally reaches the position indicated by a white circle in FIG. , It is determined that the cumulative normalized squared error relating to the amplitude of the incoming signal has been minimized. At this time, it is estimated that the signal delay time of the first wave is 16.125 chips (Tc) and the signal delay time of the second wave is 16.375 chips (Tc). The estimation error of each signal delay time of the two waves is both zero. The relative signal delay time of the second wave with respect to the first wave is 0.25
Chip (Tc), which coincides with the initially set time. Further, each incoming signal delay time is calculated by the delay time correction unit 2.
When the corrected arrival signal delay time corrected in step 6 is used, the signal amplitude error is +0.15 dB in the first wave and −0.05 dB in the second wave, and thus has almost no error. .

As described above, according to the radio wave direction-of-arrival estimation apparatus of the first embodiment, the arrival signal delay time output from the high-resolution signal processing unit 5 is corrected by the delay time correction unit 26, and the arrival signal generation unit 7 In the above, six kinds of arriving signals are formed for each arriving signal using the corrected arriving signal delay time, and the arriving angle estimating unit 8 uses the six arriving signals per one arriving signal to arrive at the shortest delay time When the signal, that is, the arrival angle of the first wave was estimated, 49.5 ° was obtained as the estimated arrival angle with respect to the original arrival angle of 50 ° of the first wave, as shown in FIG. The estimated azimuth error is 0.
5 °, and with respect to the original angle of arrival of the second wave of 0 °,
0.3 ° is obtained as the estimated arrival angle, and the estimated azimuth error is 0.3 °, so that the estimated azimuth error can be extremely reduced as compared with the known radio wave arrival direction estimating apparatus.

FIG. 4 is a block diagram showing the configuration of a second embodiment of the radio wave direction-of-arrival estimation apparatus according to the present invention.

[0093] DOA estimating apparatus of the second embodiment shown in FIG. 4, two antennas on the largest antenna open position, for example, antenna 1 ₁ shown in FIG. 10, 1 ₃ and the antenna 1 _2, It illustrates an example of selecting two correlation matrix corresponding to the two combinations of 1 _4.

The radio wave direction-of-arrival estimating apparatus according to the second embodiment comprises a correlation matrix generator 6, an arrival signal generator 7, an arrival angle estimating section 8, and a delay used in the radio direction-of-arrival estimating apparatus according to the first embodiment. Instead of the time correction unit 26, a correlation matrix generation unit 6 ', an arrival signal generation unit 7', an arrival angle estimation unit 8 ', and a delay time correction unit 26' are used, respectively. In this case, the correlation correlation matrix generator 6 and the correlation correlation matrix generator 6 ', the arrival signal generator 7 and the arrival signal generator 7', the arrival angle estimator 8 and the arrival angle estimator 8 ', and the delay time corrector 26
And the delay time correction unit 26 'are connected to the same position in the same manner and have the same function.

In FIG. 4, the same components as those shown in FIG.
The configuration of the radio wave direction-of-arrival estimation device of the second embodiment shown in FIG. 7 is the same as the configuration of the radio wave direction-of-arrival estimation device of the first embodiment described above, and the description thereof is repeated. Description is omitted.

[0096] DOA estimating apparatus of the second embodiment, a combination of the two antennas at the maximum open position, the antenna 1 ₁ and the antenna 1 ₃ and two types of antenna 1 ₂ and the antenna 1 ₄ (M = 2 ). At this time, the delay time correction unit 26 'corrects each incoming signal delay time according to the following equation (3) so that the accumulated normalized squared error is minimized, and generates a corrected incoming signal delay time. The signal is supplied to the signal generator 7 '.

[0097]

(Equation 3)

In the expression (3), when the number of cross-correlation matrices in the expression (1) is changed from 6 to 2, the expression (3) has the same form as the expression (1).

FIG. 5 is a contour diagram showing the state of obtaining the corrected arrival signal delay time by correcting the arrival signal delay time in the second embodiment.

The state of obtaining the corrected arrival signal delay time by correcting the arrival signal delay time will be described with reference to the contour diagram shown in FIG.

In FIG. 5, one side of each lattice is
This is a time step when adjusting the delay time. In this embodiment, the time step is 0.0125 chip (Tc).
I have chosen.

First, the arrival signal delay time of the first wave output from the high-resolution signal processing unit 5 is 16.05 chips (T
c) Since the arrival signal delay time of the second wave is 16.36 chips (Tc), the accumulated normalized squared error is at the position of the black circle in FIG. At this time, the cumulative normalized squared error at that position is not minimized.

The delay time correction section 26 'is provided with the first wave and the second wave.
The arriving signal delay time of the wave is changed in units of the time steps, and the arriving signal delay time obtained at each change is supplied to the arriving signal generator 7 'as a corrected arriving signal delay time.
At this time, the delay time correction unit 26 'performs the error evaluation again according to the equation (3) using the two types of arriving signals per one arriving signal from the arriving signal generating unit 7'. That is, the signal delay times of the first wave and the second wave are changed in units of the time steps in a direction in which the accumulated normalized squared error is minimized.

Such a series of operations is continued until the cumulative normalized square error relating to the amplitude of the arriving signal is minimized. When the cumulative normalized square error finally reaches the position indicated by the white circle in FIG. , It is determined that the cumulative normalized squared error relating to the amplitude of the incoming signal has been minimized. At this time, it is estimated that the signal delay time of the first wave is 16.125 chips (Tc) and the signal delay time of the second wave is 16.375 chips (Tc). The estimation error of each signal delay time of the two waves is both zero. The relative signal delay time of the second wave with respect to the first wave is 0.25
Chip (Tc), which coincides with the initially set time. Further, each incoming signal delay time is calculated by the delay time correction unit 2.
When the corrected arrival signal delay time corrected in 6 ′ is used, the signal amplitude error is +0.10 dB in the first wave,
Since it is -0.10 dB in the second wave, there is almost no error.

FIG. 6 shows an example of a change state of the estimation error of the signal delay time obtained by the delay time correction section 26 'in the second embodiment and the high-resolution signal processing section 4 in the known radio wave direction-of-arrival estimation apparatus. FIG. 4 is a characteristic diagram showing a case where the arrival azimuth of the second wave is fixed to 0 ° and the arrival azimuth of the first wave is 0 °.
It shows the characteristics obtained when the angle was changed from 180 to 180 °.

FIG. 7 is a characteristic diagram showing an example of a change state of the signal amplitude error obtained by the arriving signal generation units 7 'and 7 in the second embodiment and the known radio wave arrival direction estimating apparatus. It shows the characteristics obtained when the arrival azimuth of the two waves is fixed at 0 ° and the arrival azimuth of the first wave is changed from 0 ° to 180 °.

FIG. 8 is a characteristic diagram showing an example of a change state of the estimation error of the arrival azimuth angle obtained by the arrival angle estimation units 8 'and 8 in the present embodiment and the known radio wave arrival direction estimation device. The azimuth of arrival of the second wave is fixed at 0 °,
This shows the characteristics obtained when the wave arrival azimuth is changed from 0 ° to 180 °.

6 to 8, curves a connecting white diamond points are characteristics in the radio wave direction-of-arrival estimation apparatus of the second embodiment, and curves b connecting black diamond points are known radio arrival directions. This is a characteristic of the estimating device.

Various characteristics of the radio wave direction-of-arrival estimation apparatus of the second embodiment and a known direction-of-arrival direction estimation apparatus will be compared with each other with reference to FIGS.

First, as shown in FIG. 6, the arrival angle of the second wave is fixed at 0 °, and the arrival angle of the first wave is changed from 0 ° to 18 °.
The signal delay time error obtained when the angle is changed to 0 ° is the maximum error of the known radio wave direction-of-arrival estimation device of 0.19.
In contrast to Tc (see curve b), the maximum error of the radio wave direction-of-arrival estimation device of the second embodiment is 0.05 Tc (see curve a), and the signal delay time of the radio wave direction-of-arrival estimation device of this embodiment is The error has been greatly reduced.

Next, as shown in FIG. 7, the arrival angle of the second wave is fixed at 0 °, and the arrival angle of the first wave is changed from 0 ° to 18 °.
The signal amplitude error obtained when the angle is changed to 0 ° is that the maximum error of the known radio wave arrival direction estimating apparatus is 18.2 dB.
(See curve b), the maximum error of the radio wave direction-of-arrival estimation device of the second embodiment is 4.4 dB (see curve a), and the signal amplitude error of the radio wave direction-of-arrival estimation device of this embodiment is It has been greatly reduced.

Next, as shown in FIG. 8, the arrival angle of the second wave is fixed at 0 °, and the arrival angle of the first wave is changed from 0 ° to 18 °.
When the angle of arrival is changed to 0 °, the maximum error of the known radio wave direction-of-arrival estimation device is 153.
In contrast to 0 ° (see curve b), the maximum error of the radio wave direction-of-arrival estimation apparatus of the second embodiment is 3.2 ° (see curve a).
Thus, the error of the azimuth of arrival of the radio wave direction-of-arrival estimation device of the present embodiment is also greatly reduced.

As described above, according to the radio wave direction-of-arrival estimation apparatus of the second embodiment, the arrival signal delay time output from the high-resolution signal processing unit 5 is corrected to the corrected arrival signal delay time by the delay time correction unit 26 '. Then, the arrival signal generator 7 'uses the corrected arrival signal delay time to form two arrival signals by combining two antennas located at the maximum aperture position for each arrival signal in units of a cross-correlation matrix, and The estimating unit 8 'estimates the arrival signal having the shortest delay time, that is, the arrival angle of the first wave, using the two arrival signals.
As shown in FIG. 10, the original angle of arrival 50 of the first wave
, The estimated azimuth error is 0.5 °, and the estimated azimuth error of the second wave is 0 °, while the estimated azimuth error is 0.3 °. Is obtained, the estimated azimuth error becomes 0.3 °, and the estimated azimuth error similar to that in the first embodiment can be extremely reduced.

Further, according to the radio wave direction-of-arrival estimation device of the second embodiment, the direction of arrival of each signal radio wave can be estimated with the same estimated azimuth error as that of the radio wave direction-of-arrival estimation device of the first embodiment. The number of cross-correlation matrices can be reduced as compared with the radio wave direction-of-arrival estimation device of the first embodiment, and the calculation amount of each unit can be reduced accordingly.

In each of the embodiments described above, an example has been described in which two waves, the first wave and the second wave, arrive as signal radio waves. However, the radio wave arrival direction estimating apparatus of the present invention estimates the angle of arrival. The number of signal radio waves that can be processed is not limited to two, and the same operation can be performed even when three or more signal radio waves arrive.

[0116] Further, in the above each embodiment, the four antennas 1 ₁ to using 1 _4, four normalized signal generator 2 _{'1-2' 4} and four autocorrelation matrix associated with it If has been described by way of example using the generator 3 ₁ to 3 _4, the radio wave arrival direction estimating apparatus antenna used in, normalized signal generation unit of the present invention, the number of the autocorrelation matrix generation unit is four The present invention is not limited to this, and three or more numbers may be used as long as the number is plural.

Further, in each of the above embodiments, an example has been described in which the _four antennas 11 to ₁₄ are arranged in a circle, but the antenna arrangement in the radio wave direction-of-arrival estimation device of the present invention is circular. The arrangement is not limited to a certain case, but may be a lattice-like arrangement or an arrangement of other shapes.

[0118] In addition, in the above each embodiment, each normalized signal generating unit 2 ₁ to 2 ₄ are described using an example of using a configuration illustrated in FIG. 2, the radio wave of the present invention Each of the normalized signal generators used in the DOA estimator is shown in FIG.
It is not limited to the configuration as shown in FIG.
As long as it has a configuration similar to that configuration and performs a similar operation, another configuration may be used.

Further, in each of the above embodiments, an example in which the MUSIC method is used for the signal processing performed by the high resolution signal processing unit 5 has been described. The signal processing unit 5 is not limited to the one using the MUSIC method, but may use another method as long as it has a high temporal signal resolution.

[0120]

As described above, according to the present invention, the number (K) of arriving signals output from the high-resolution signal processing section and K
Are supplied to the arriving signal generator, and when the arriving signal generator generates each of the M arriving K signals, a predetermined number of arriving signals are generated based on the amplitudes of the K arriving signals generated in each of the M arriving signals. According to the steps, their cumulative normalization 2
The K incoming signal delay times are sequentially corrected so that the multiplication error is minimized, converted into K corrected incoming signal delay times, and supplied to the incoming signal generating unit. In the incoming signal generating unit,
Since the K arriving signals are generated M each using the corrected arriving signal delay time, the K arriving signals are caused by noise components in the N normalized signals output from the N normalized signal generators. The offset of the incoming signal delay time and the offset of the K incoming signal delay times output from the high-resolution signal processing unit are substantially eliminated by correcting each incoming signal delay time to each corrected incoming signal delay time. As a result, the error components included in the K arriving signals generated from the M arriving signal generators are removed, and
The arrival angle estimation unit has an effect that the arrival directions of the K arrival signals can always be accurately estimated.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a first embodiment of a radio wave direction of arrival estimation apparatus according to the present invention.

FIG. 2 is a block diagram illustrating an example of a configuration of a normalized signal generator in the first and second embodiments.

FIG. 3 is a contour diagram showing an example of a cumulative normalized squared error with respect to the signal delay time of the first wave and the second wave signal waves arriving at the radio wave direction-of-arrival estimation apparatus of the first embodiment.

FIG. 4 is a block diagram showing a configuration of a second embodiment of a radio wave direction of arrival estimation apparatus according to the present invention.

FIG. 5 is a contour diagram showing an example of a cumulative normalized squared error with respect to a signal delay time of a first wave and a second wave of signal waves arriving at the radio wave direction-of-arrival estimation apparatus of the second embodiment.

FIG. 6 is a characteristic diagram showing an example of a change state of a delay time estimation error with respect to an azimuth of arrival of a first wave in a second embodiment and a known radio wave direction-of-arrival estimation device.

FIG. 7 is a characteristic diagram showing an example of a change state of a signal amplitude error with respect to an azimuth of arrival of a first wave in the second embodiment and a known radio wave arrival direction estimation device.

FIG. 8 is a characteristic diagram showing an example of a change state of an estimation error of the arrival azimuth with respect to the arrival azimuth of the first wave in the second embodiment and the known radio wave arrival direction estimation device.

FIG. 9 is a block diagram illustrating an example of a configuration of a known radio wave arrival direction estimation device.

FIG. 10 is a configuration diagram illustrating an example of an arrangement state of four antennas and an arrival state of a signal radio wave.

FIG. 11 is a signal waveform diagram illustrating an example of a band-limited spread modulated signal that is spread-modulated by a PN code on the transmission side and further band-limited, and a reference signal generated by a reference signal generator.

FIG. 12 is a signal waveform diagram illustrating an example of a baseband (BB) spread modulation signal.

FIG. 13 is a signal waveform diagram illustrating an example of a frequency domain reference signal.

FIG. 14 is a signal waveform diagram illustrating an example of a frequency domain reception signal.

FIG. 15 is a signal waveform diagram showing an example of a normalized signal including a noise component and a normalized signal from which the noise component has been removed.

FIG. 16 is a signal waveform diagram illustrating an example of a correlation metric signal including a delay time output from a high-resolution signal processing unit.

[Explanation of symbols]

1 ₁ to 1 ₄ antennas 2 _{1 to} 2 ₄ Normalized signal generator 3 _{1 to} 3 ₄ Autocorrelation matrix generator 4 Averaged autocorrelation matrix generator 5 High-resolution signal processor 6, 6 'Cross-correlation matrix generator 7 , 7 ′ Arrival signal generator 8, 8 ′ Arrival angle estimator 9 Signal output terminal 10 Receiver 11 First Fourier transformer 12 First filter 13 Reference signal generator 14 Second Fourier transformer 15 Second filter 16 Division unit 17 Noise removal processing unit 18 Baseband (BB) signal generation unit 19 Analog-digital (A / D) conversion unit 20 Memory 21 Fourier transformation unit 22 Noise removal unit 23 Inverse Fourier transformation unit 24 Received signal input terminal 25 Normalization Signal output terminals 26, 26 'Delay time correction unit 27 Frequency window multiplication unit 28 Frequency window division unit

Claims (12)

[Claims] 1. A signal radio wave captured by an integer (N) of three or more antennas arranged close to each other is converted into N normalized signals,
Generating N autocorrelation matrices and an integer (M) of two or more cross-correlation matrices from the N normalized signals; and generating an averaged autocorrelation matrix by averaging the N autocorrelation matrices. , The averaged autocorrelation matrix is subjected to signal separation processing with high time resolution to generate the number of arriving signals (K) and K arriving signal delay times, and the M cross-correlation matrices, the arriving signal number and K K corrected arrival signal delay times are used to generate K arrival signals, and the arrival directions of the K arrival signals are determined based on the phases of the K arrival signals generated in each of the M cases. The K corrected arrival signal delay times are estimated based on the error evaluation based on the amplitudes of the K arrival signals generated in each of the M cases, and the K arrival signals are minimized so that the error is minimized. A radio wave arrival direction estimator characterized in that the delay time is corrected. Method. 2. The N number of normalized signals are obtained by calculating a ratio between a baseband signal generated by frequency-converting each signal radio wave and a reference signal serving as a reference. The radio wave arrival direction estimating method according to claim 1. 3. The radio wave arrival direction according to claim 2, wherein the N normalized signals are obtained by calculating a ratio between the baseband signal and the reference signal in a frequency domain. Estimation method. 4. The baseband signal includes a signal spread-modulated with a predetermined code, and the reference signal is a signal having a high correlation with the predetermined code. Radio wave arrival direction estimation method described in 1. (5) 2. The method for estimating the direction of arrival of a radio wave according to claim 1, wherein 6. The M cross-correlation matrices are M out of a maximum of {N (N-1) / 2} cross-correlation matrices that can be generated by a combination of the N normalized signals. The selected M cross-correlation matrices are cross-correlation matrices corresponding to normalized signals of two antennas that are arranged relatively far apart from each other among the N antennas. The radio wave arrival direction estimating method according to claim 1, characterized in that: 7. An N-number of normalized signal generators for generating N-number of normalized signals by receiving an integer (N) of three or more antennas arranged in close proximity and receiving signal radio waves captured by the N-number of antennas. An N autocorrelation matrix generator for generating an autocorrelation matrix from each of the N normalized signals; and an averaged autocorrelation matrix generator for generating an averaged autocorrelation matrix of the N autocorrelation matrices And the number (K) of incoming signals and K
A high-resolution signal processing unit for generating a plurality of incoming signal delay times; a cross-correlation matrix generating unit for generating an integer (M) of two or more cross-correlation matrices from the N normalized signals;
Using the number of cross-correlation matrices, the number of arriving signals, and the K corrected arriving signal delay times, an arriving signal generator that generates M arriving K signals, and an output from the high-resolution signal processor. The K incoming signal delay times are evaluated for errors based on the amplitudes of the K incoming signals generated in each of the M ways, and the K incoming signal delay times are corrected so that the error is minimized. A delay time correction unit that supplies the K corrected arrival signal delay times to the arrival signal generation unit, and a direction of arrival of the K arrival signals based on the phases of the M occurrences of the K arrival signals. A radio wave direction-of-arrival estimating device comprising: an angle-of-arrival estimating unit. 8. The baseband signal generating means for frequency-converting a reception signal of each signal radio wave captured by the antenna to generate a baseband signal, and the N normalized signal generation units serve as a reference. The radio wave according to claim 7, further comprising: a reference signal generating unit that generates a reference signal; and a dividing unit that generates a normalized signal by taking a ratio between the baseband signal and the reference signal. Direction of arrival estimation device. 9. The N number of normalized signal generators each generate a normalized signal by taking the ratio of the baseband signal and the reference signal in the frequency domain by the divider. The radio wave direction-of-arrival estimation apparatus according to claim 8, wherein: 10. The baseband signal generating means generates a baseband signal including a signal spread-modulated with a predetermined code, and the reference signal generating means generates a reference signal having a high correlation with the predetermined code. The radio wave arrival direction estimating device according to claim 8 or 9, wherein the radio wave arrival direction is generated. 11. The delay time correction section comprises: 8. The radio wave arrival direction estimating apparatus according to claim 7, wherein the signal arrival time is corrected by calculating the delay time. 12. The cross-correlation matrix generation unit calculates M from a maximum of {N (N-1) / 2} cross-correlation matrices that can be generated by a combination of the N normalized signals.
Of the cross-correlation matrices, wherein the selected M
8. The radio wave arrival direction according to claim 7, wherein the number of cross-correlation matrices correspond to normalized signals of two antennas which are arranged relatively apart from each other among the N antennas. Estimation device.