JP2019203720A - Arrival direction estimating device and arrival direction estimating method - Google Patents

Abstract

To provide an arrival direction estimation technique with which it is possible to suppress the side lobe of an angle calculation result while improving angle separation performance in a two-dimensionally or a three-dimensionally arranged antenna group with a few number of antennas.SOLUTION: The correlation matrix of a received signal is divided into a block matrix. In a block matrix including a matrix element not associated with a virtual receive antenna, a matrix element interpolation process of interpolating matrix elements associated with the virtual receive antenna by matrix elements not associated with the virtual antenna with equal antenna intervals is performed and a first extended correlation matrix is generated. A block matrix interpolation process is applied to the first extended correlation matrix and a second extended correlation matrix is generated. The block matrix interpolation process involves performing an interpolation similar to the matrix element interpolation process on the first extended correlation matrix assuming the block matrix as a matrix element. The arrival direction of a radio wave is estimated on the basis of the second extended correlation matrix.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a technique for estimating the arrival direction of radio waves.

  The radar apparatus estimates the arrival direction of a radio wave (a reflected wave from an object) by ESPRIT (Estimation of Signal Parameters via Rotational Invariance Techniques) method or the like. The arrival direction estimation method is a method of calculating an arrival direction (angle) from signals received by a plurality of antennas. The angle calculation performance varies greatly depending on the antenna interval. Further, if all the antenna intervals are wider than the half wavelength of the incoming radio wave, phase folding occurs.

  Increasing the angle separation performance by narrowing the half width MW of the main lobe of the angle spectrum can generally be dealt with by increasing the distance L1 between the antennas at both ends.

  For example, when the one-dimensionally arranged antenna group shown in FIG. 10A is used, the angle calculation result (angle spectrum) shown in FIG. 10B is obtained, whereas the one-dimensional shown in FIG. When the arrangement antenna group is used, an angle calculation result (angle spectrum) shown in FIG. 11B is obtained. The angle calculation result shown in FIG. 10B and the angle calculation result shown in FIG. 11B are both angle calculation results of radio waves coming from 0 deg.

  In the one-dimensionally arranged antenna group shown in FIG. 10A and the one-dimensionally arranged antenna group shown in FIG. 11A, the antennas are equally spaced L0. The one-dimensionally arranged antenna group shown in FIG. 11A has a larger number of antennas than the one-dimensionally arranged antenna group shown in FIG. Therefore, in the angle calculation result shown in FIG. 11B, the half width MW of the main lobe is narrower than the angle calculation result shown in FIG.

JP 2017-058359 A

  However, increasing the number of antennas to increase the distance L1 between the antennas at both ends leads to an increase in cost. This tendency becomes more prominent in a two-dimensionally arranged antenna group that enables angle estimation with two axes (see, for example, Patent Document 1) and a three-dimensionally arranged antenna group that allows angle estimation with three axes.

  In order to suppress the increase in the number of antennas and increase the distance L1 between the antennas at both ends while preventing the occurrence of phase folding, the distance between the antennas is unequal as in the one-dimensionally arranged antenna group shown in FIG. An interval (interval L0 ≠ interval L0 ′ ≠ interval L0 ″) may be set.

  However, in the angle calculation result shown in FIG. 12B obtained when the one-dimensionally arranged antenna group shown in FIG. 12A is used, the sidelobe SH is compared with the angle calculation result shown in FIG. There is a problem that the height becomes high and causes ghost detection.

  In view of the above problems, the present invention provides a direction-of-arrival estimation technique capable of suppressing side lobes of angle calculation results while improving angle separation performance with a two-dimensionally arranged antenna group or a three-dimensionally arranged antenna group with a small number of antennas. For the purpose.

  An arrival direction estimation device according to the present invention, an acquisition unit that acquires a signal for each reception antenna based on a reception signal received by an antenna group in which a real reception antenna and a non-existing virtual reception antenna are two-dimensionally arranged; A generation unit that generates a correlation matrix based on the signal for each of the reception antennas acquired by the acquisition unit, and a first interpolation unit that performs a first interpolation process on the correlation matrix to generate a first extended correlation matrix A second interpolation unit that generates a second extended correlation matrix by performing a second interpolation process on the first extended correlation matrix, and estimates a radio wave arrival direction based on the second extended correlation matrix An antenna that is adjacent to each other along the first axis direction of the antenna group, and the antenna that is adjacent to the second axis direction of the antenna group is the constant antenna. At intervals The first axis direction and the second axis direction are different from each other, and the first interpolation processing is performed by the number of antennas arranged along the first axis direction × the first axis. A block matrix composed of the number of antennas arranged along the axial direction of 1 unit, a partition process for partitioning the correlation matrix into a plurality of the block matrices, and a matrix element not related to the virtual receiving antenna A matrix element interpolation process for interpolating a matrix element related to the virtual receiving antenna in a block matrix with a matrix element not related to the virtual receiving antenna having the same antenna interval, and the second interpolation process includes: 1 extended correlation matrix having a block matrix interpolation process (first configuration) in which the block matrix is regarded as a matrix element and interpolation similar to the matrix element interpolation process is performed. The

  The arrival direction estimation apparatus having the first configuration may be configured (second configuration) in which the reception antennas are arranged with minimum redundancy.

  An arrival direction estimation method according to the present invention obtains a signal for each receiving antenna based on a received signal received by an antenna group in which a real receiving antenna and a non-existing virtual receiving antenna are two-dimensionally arranged; A generation step of generating a correlation matrix based on the signal for each reception antenna acquired by the acquisition step, and a first interpolation step of generating a first extended correlation matrix by performing a first interpolation process on the correlation matrix A second interpolation step of generating a second extended correlation matrix by performing a second interpolation process on the first extended correlation matrix, and estimating a radio wave arrival direction based on the second extended correlation matrix An estimation step, wherein antennas adjacent along the first axis direction of the antenna group are arranged at regular intervals, and antennas adjacent along the second axis direction of the antenna group are The first interpolation direction is different from the second axial direction, and the first interpolation processing is performed by the number of antennas arranged along the first axial direction. A block process composed of the number of antennas arranged along the first axis direction as one unit, a partitioning process for dividing the correlation matrix into a plurality of the block matrices, and a matrix not related to the virtual receiving antenna A matrix element interpolation process for interpolating a matrix element related to the virtual receive antenna with a matrix element not related to the virtual receive antenna having an equal antenna interval in the block matrix including elements, and the second interpolation process Is configured to include a block matrix interpolation process in which the block matrix is regarded as a matrix element and interpolation similar to the matrix element interpolation process is performed in the first extended correlation matrix (third Configuration).

  According to the arrival direction estimation technique according to the present invention, it is possible to suppress the side lobe of the angle calculation result while improving the angle separation performance with the two-dimensionally arranged antenna group or the three-dimensionally arranged antenna group having a small number of antennas.

The figure which shows the structural example of a radar apparatus The figure which shows an example of the antenna arrangement | positioning in a two-dimensional arrangement | positioning antenna group The figure which shows the other example of the antenna arrangement | positioning in a two-dimensional arrangement | positioning antenna group Flow chart showing operation of signal processing apparatus Conceptual diagram showing the correlation matrix of the received signal Conceptual diagram showing the first extended correlation matrix Conceptual diagram showing a second extended correlation matrix Diagram showing angle spectrum The figure which shows the angle spectrum of the comparative example Diagram showing one-dimensionally arranged antenna group and angle spectrum Diagram showing one-dimensionally arranged antenna group and angle spectrum Diagram showing one-dimensionally arranged antenna group and angle spectrum

  Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the drawings.

<1. Configuration of radar device>
FIG. 1 is a diagram illustrating a configuration of a radar apparatus 1 according to the present embodiment. The radar apparatus 1 is mounted on a vehicle such as an automobile. When the radar apparatus 1 is mounted at the front end of the host vehicle, the radar apparatus 1 acquires target data related to a target existing in front of the host vehicle using a transmission wave. The target data includes a distance to the target, a relative speed of the target with respect to the radar device 1, and the like. However, in order to describe the radar apparatus 1 according to the present embodiment as an example of the arrival direction estimation apparatus, only the part related to the arrival direction estimation will be described in the following description.

  As shown in FIG. 1, the radar apparatus 1 mainly includes a transmission unit 2, a reception unit 3, and a signal processing device 4.

  The transmission unit 2 includes a signal generation unit 21 and a transmitter 22. The transmitter 22 modulates the signal generated by the signal generator 21 to generate a transmission signal. The transmission antenna 23 converts the transmission signal into a transmission wave TW and outputs it.

  The receiving unit 3 includes a plurality of receiving antennas 31 that form a two-dimensionally arranged antenna group together with a plurality of virtual receiving antennas, and a plurality of individual receiving units 32 connected to the plurality of receiving antennas 31. The virtual reception antenna is an antenna that does not exist, and the reception antenna 31 is an actual antenna. In the present embodiment, the reception unit 3 includes, for example, 12 reception antennas 31 and 12 individual reception units 32. The twelve individual receiving units 32 correspond to the twelve receiving antennas 31 respectively. Each receiving antenna 31 receives a reflected wave RW from an object and acquires a received signal, and each individual receiving unit 32 processes the received signal obtained by the corresponding receiving antenna 31.

  FIG. 2 is a diagram illustrating an example of antenna arrangement in the two-dimensional arrangement antenna group. In the arrangement example shown in FIG. 2, 4 × 3 reception antennas (ch1, ch2, ch5, ch7, ch8, ch9, 4 × 3) are formed on a lattice point of a square lattice whose one side is a half wavelength (λ / 2) of the transmission wave TW. The receiving antennas of ch12, ch14, ch22, ch23, ch26, and ch28) are arranged with the least redundancy. And a virtual receiving antenna is arrange | positioned in the lattice point where the receiving antenna is not arrange | positioned.

  FIG. 3 is a diagram illustrating an example of antenna arrangement in the two-dimensional arrangement antenna group. In the arrangement example shown in FIG. 3, 4 × 3 receiving antennas (ch1, ch2, ch5, ch7, ch8, ch9, 4 × 3) are arranged on the lattice point of a rhombus lattice whose one side is a half wavelength (λ / 2) of the transmission wave TW. The receiving antennas of ch12, ch14, ch22, ch23, ch26, and ch28) are arranged with the least redundancy. And a virtual receiving antenna is arrange | positioned in the lattice point where the receiving antenna is not arrange | positioned.

  Returning to FIG. 1, each individual reception unit 32 includes a mixer 33 and an A / D converter 34. The received signal obtained by the receiving antenna 31 is amplified by a low noise amplifier (not shown) and then sent to the mixer 33. The transmission signal from the transmitter 22 of the transmission unit 2 is input to the mixer 33, and the transmission signal and the reception signal are mixed in the mixer 33. As a result, a beat signal having a beat frequency that is the difference between the frequency of the transmission signal and the frequency of the reception signal is generated. The beat signal generated by the mixer 33 is converted to a digital signal by the A / D converter 34 and then output to the signal processing device 4.

  When the radar apparatus 1 is an FMCW (Frequency Modulation Continuous Wave) radar apparatus, the frequency difference between the transmission wave TW and the reception wave RW increases or decreases in proportion to the distance between the target and the radar apparatus. Is a fluctuation component of the distance. On the other hand, when the radar apparatus 1 is an FCM (First Chirp Modulation) type radar apparatus, the phase difference between the transmission wave TW and the reception wave RW increases or decreases in proportion to the distance between the target and the radar apparatus. Therefore, the fluctuation component of the beat signal due to this phase difference becomes the fluctuation component of the distance. In addition, since the received wave RW is affected by the speed of the target when reflected by the target, the frequency difference between pulses increases and decreases in proportion to the relative speed (Doppler frequency) between the target and the radar device. The fluctuation component of the beat signal due to the frequency difference between the pulses becomes the velocity fluctuation component. When there are a plurality of targets having different relative speeds and distances, each receiving antenna 31 receives a plurality of reflected waves having different phase shift amounts and Doppler shift amounts, and the beat signal obtained from the mixer 33 includes each target. Various components corresponding to are included.

  The signal processing device 4 includes a microcomputer including a CPU (Central Processing Unit) and a memory 41. The signal processing device 4 stores various data to be calculated in a memory 41 that is a storage device. The memory 41 is, for example, a RAM (Random Access Memory). The signal processing device 4 includes a transmission control unit 42, a Fourier transform unit 43, and a data processing unit 44 as functions implemented by a microcomputer as software. The transmission control unit 42 controls the signal generation unit 21 of the transmission unit 2. The data processing unit 44 includes a peak extraction unit 45, an extension calculation unit 46, and an azimuth calculation unit 47.

  Since the Fourier transform unit 43 is received by the receiving antenna 31 in a state where the reflected waves from a plurality of targets are overlapped, the frequency component based on the reflected waves of each target is generated from the beat signal generated based on the received signal. Is performed (for example, FFT (Fast Fourier Transfer) processing). In the FFT processing, reception level and phase information are calculated for each frequency point (sometimes referred to as frequency bin) set at a predetermined frequency interval.

  The peak extraction unit 45 detects a peak from the result of the FFT process or the like by the Fourier transform unit 43. The extended calculation unit 46 calculates an extended correlation matrix based on the signal of the frequency bin that caused the peak. The azimuth calculation unit 47 estimates the azimuth (angle) at which the target exists based on the extended correlation matrix. The azimuth calculation unit 47 outputs the azimuth (angle) at which the estimated target exists to the memory 41, the vehicle control ECU 51, and the like.

<2. Operation of Signal Processing Device>
Next, the operation of the signal processing device 4 will be described. FIG. 4 is a flowchart showing the operation of the signal processing device 4. The signal processing device 4 periodically repeats the process shown in FIG. 4 at regular time intervals.

  The signal processing device 4 acquires a predetermined number of beat signals (step S10). Next, the Fourier transform unit 43 performs an FFT operation on the beat signal (step S20).

  Next, the peak extraction unit 45 extracts a peak from the result of the FFT calculation (step S30). Then, the extended calculation unit 46 calculates an extended correlation matrix based on the signal of the frequency bin that caused the peak (step S40). Finally, the azimuth calculation unit 47 calculates an angle spectrum based on the extended correlation matrix, and estimates the azimuth (angle) where the target exists based on the calculated angle spectrum (step S50).

<3. Calculation of extended correlation matrix>
The extension calculation unit 46 sets the reception signal of the virtual reception antenna to 0.

  The extension calculation unit 46 defines a reception signal vector X of 28 rows and 1 column. Each component of the reception signal vector X is a peak value extracted by each reception antenna and the peak extraction unit 45 of each reception antenna. The row components of the received signal vector X are arranged in the order of ch1 to ch28.

The extension calculation unit 46 generates a correlation matrix of the received signal. The correlation matrix R _XX is expressed as follows. Here, [·] ^H represents a complex conjugate transpose. The correlation matrix R _XX is a matrix of 28 rows and 28 columns.
R _XX = XX ^H

  The extended calculation unit 46 performs a first interpolation process on the correlation matrix to generate a first extended correlation matrix. The first interpolation processing includes segmentation processing and matrix element interpolation processing. The classification process includes the number of antennas arranged along the first axis direction (total number of reception antennas and virtual reception antennas) × the number of antennas arranged along the first axis direction (total number of reception antennas and virtual reception antennas). ) Is a unit, and the correlation matrix is divided into a plurality of block matrices. Here, the direction of a straight line connecting the receiving antenna of ch1, the receiving antenna of ch2, the receiving antenna of ch5, and the receiving antenna of ch7 is defined as a first axial direction. Unlike the present embodiment, the first axis direction may be the direction of a straight line connecting the ch1 reception antenna, the ch8 reception antenna, and the ch22 reception antenna.

  FIG. 5 is a conceptual diagram showing a correlation matrix. Each 7 × 7 thick frame is a block matrix. Each cell is a matrix element of the correlation matrix. A solid matrix element indicates a matrix element related to the virtual receiving antenna, and has a value of zero.

  In the matrix element interpolation process, in the block matrix including the matrix elements not related to the virtual receiving antenna (the block row example including the matrix elements that are not plain in FIG. 5), the matrix elements related to the virtual receiving antenna are the virtual receiving antennas having the same antenna interval. Is a process of interpolating with matrix elements not related to. FIG. 6 is a conceptual diagram showing the first extended correlation matrix.

Here, the interpolation will be described by taking the block matrix B1 in the first to seventh rows and the first to seventh columns of the correlation matrix as an example. The block matrix B1 is expressed as the following formula 1. Here, [•] ^* represents a complex conjugate.

  In the block matrix B1, matrix elements arranged diagonally correspond to matrix elements having the same antenna interval. Therefore, by replacing “matrix elements that are zero” with “non-zero matrix elements” in diagonally arranged matrix elements, matrix elements that are not related to virtual receiving antennas having the same antenna interval are replaced with matrix elements that are related to virtual receiving antennas. Can be interpolated.

  In the block matrix B1, there are a plurality of “non-zero matrix elements” in the diagonal line indicated by the dotted line, and therefore a plurality of interpolation methods are conceivable.

For example, an average value “a” of a plurality of “non-zero matrix elements” may be obtained, and all matrix elements of diagonal lines indicated by dotted lines may be set to “a”. In this case, the block matrix B11 of the first extended correlation matrix corresponding to the block matrix B1 is expressed as in Equation 2 below.

Further, for example, an average value a of a plurality of “non-zero matrix elements” may be obtained, and “a matrix element that is 0” in a diagonal line indicated by a dotted line may be replaced with a. In this case, the block matrix B11 of the first extended correlation matrix corresponding to the block matrix B1 is expressed as Equation 3 below.

Alternatively, for example, one of a plurality of “non-zero matrix elements” may be selected, and the “matrix element that is zero” in the diagonal line indicated by the dotted line may be replaced with the selected “non-zero matrix element”. Good. In this case, the block matrix B11 of the first extended correlation matrix corresponding to the block matrix B1 is expressed as the following Expression 4.

  Finally, the extended calculation unit 46 performs a second interpolation process on the first extended correlation matrix to generate a second extended correlation matrix. The second interpolation process includes a block matrix interpolation process. The block matrix interpolation process is a process in which, in the first extended correlation matrix, the block matrix is regarded as a matrix element and the same interpolation as the matrix element interpolation process is performed. FIG. 7 is a conceptual diagram showing the second extended correlation matrix. The second extended correlation matrix is an extended correlation matrix that is finally calculated in step S40 described above.

  By applying the second extended correlation matrix in the direction-of-arrival estimation process, the interval between adjacent antennas has been changed from a non-uniform interval to a pseudo equal interval. The side lobe of the angle calculation result can be suppressed while improving the.

  FIG. 8 shows the angle spectrum when the second extended correlation matrix (see FIG. 7) is applied, and FIG. 9 shows the angle spectrum when the correlation matrix (see FIG. 5) is applied as a comparative example. . In FIG. 8 and FIG. 9, shaded portions indicate regions where the signal level is −7 [dB] or more. By applying the second extended correlation matrix (see FIG. 7), the side lobe of the angle calculation result can be suppressed.

  Unlike this embodiment, even when the actual arrangement of receiving antennas is not the minimum redundant arrangement, the angle separation performance can be improved and the side lobe of the angle calculation result can be suppressed with the two-dimensionally arranged antenna group with a small number of antennas. However, “0 matrix elements” remain in the second extended correlation matrix. For this reason, it is desirable that the actual arrangement of the receiving antennas is a minimum redundant arrangement.

<4. Other>
Various technical features disclosed in the present specification can be variously modified within the scope of the technical creation in addition to the above-described embodiment. In addition, a plurality of embodiments and modification examples shown in the present specification may be implemented in combination within a possible range.

  For example, in the radar apparatus, instead of the above-described FCM method or FMCW method, for example, a pulse Doppler method that detects a Doppler shift as a phase change between a plurality of pulse signals instead of a beat signal frequency may be employed.

  In the above-described embodiment, the arrival direction estimation apparatus that processes the reception signal received by the two-dimensionally arranged antenna group may be an arrival direction estimation apparatus that processes the reception signal received by the three-dimensionally arranged antenna group.

  In the above-described embodiments, the on-vehicle radar device has been described. However, the present invention can also be applied to an infrastructure radar device, an aircraft monitoring radar device, and the like installed on a road or the like.

DESCRIPTION OF SYMBOLS 1 Radar apparatus 2 Transmitting part 3 Receiving part 31 Receiving antenna 4 Signal processing apparatus 46 Expansion calculating part 47 Direction calculating part

Claims (3)

An acquisition unit for acquiring a signal for each reception antenna based on a reception signal received by an antenna group in which a real reception antenna and a non-existing virtual reception antenna are two-dimensionally arranged;
A generating unit that generates a correlation matrix based on the signal for each receiving antenna acquired by the acquiring unit;
A first interpolation unit that performs a first interpolation process on the correlation matrix to generate a first extended correlation matrix;
A second interpolation unit configured to generate a second extended correlation matrix by performing a second interpolation process on the first extended correlation matrix;
An estimation unit that estimates an arrival direction of radio waves based on the second extended correlation matrix;
With
The adjacent antennas along the first axial direction of the antenna group are arranged at a constant interval,
The adjacent antennas along the second axial direction of the antenna group are arranged at the constant interval,
The first axial direction and the second axial direction are different from each other,
The first interpolation process includes:
A block matrix composed of the number of antennas arranged along the first axial direction × the number of antennas arranged along the first axial direction is defined as one unit, and the correlation matrix is divided into a plurality of the block matrices. Classification processing to classify,
A matrix element interpolation process for interpolating a matrix element related to the virtual receive antenna with a matrix element not related to the virtual receive antenna having an equal antenna interval in the block matrix including a matrix element not related to the virtual receive antenna. And
The second interpolation process includes:
In the first extended correlation matrix, the block matrix interpolation processing that performs interpolation similar to the matrix element interpolation processing by regarding the block matrix as a matrix element,
Direction of arrival estimation device.   The direction-of-arrival estimation apparatus according to claim 1, wherein the reception antennas are arranged in a least redundant manner. An acquisition step of acquiring a signal for each reception antenna based on a reception signal received by an antenna group in which a real reception antenna and a non-existing virtual reception antenna are two-dimensionally arranged;
A generating step of generating a correlation matrix based on the signal for each of the receiving antennas acquired by the acquiring step;
A first interpolation step of generating a first extended correlation matrix by performing a first interpolation process on the correlation matrix;
A second interpolation step of generating a second extended correlation matrix by performing a second interpolation process on the first extended correlation matrix;
An estimation step of estimating an arrival direction of radio waves based on the second extended correlation matrix;
With
The adjacent antennas along the first axial direction of the antenna group are arranged at a constant interval,
The adjacent antennas along the second axial direction of the antenna group are arranged at the constant interval,
The first axial direction and the second axial direction are different from each other,
The first interpolation process includes:
A block matrix composed of the number of antennas arranged along the first axial direction × the number of antennas arranged along the first axial direction is defined as one unit, and the correlation matrix is divided into a plurality of the block matrices. Classification processing to classify,
A matrix element interpolation process for interpolating a matrix element related to the virtual receive antenna with a matrix element not related to the virtual receive antenna having an equal antenna interval in the block matrix including a matrix element not related to the virtual receive antenna. And
The second interpolation process includes:
In the first extended correlation matrix, the block matrix interpolation processing that performs interpolation similar to the matrix element interpolation processing by regarding the block matrix as a matrix element,
Direction of arrival estimation method.