JP7335092B2 - Radar device and signal processing method - Google Patents

Description

The present invention relates to radar equipment and signal processing methods.

A radar device emits radio waves and receives radio waves (reflected waves) reflected from a target, thereby estimating the direction of arrival of the reflected waves. The method of estimating the direction of arrival is a method of calculating the direction of arrival (angle) from information on the phase difference and amplitude difference of received signals obtained by a plurality of receiving antennas that receive reflected waves.

JP-A-2006-329671

In calculating the direction of arrival (angle), the distance between the receiving antennas is very important, and greatly affects the angular accuracy and phase folding. There is a trade-off relationship between angular accuracy and phase folding. Specifically, if the receiving antenna spacing is wide, the angle accuracy of the target is improved, but phase folding is likely to occur. gets worse.

The incident angle estimating device disclosed in Patent Document 1 attempts to solve the problem of phase folding and improve the accuracy of the azimuth indicating the arrival direction of radio waves. However, the incident angle estimating device disclosed in Patent Document 1 has a problem that it cannot cope with the case where there are multiple azimuths representing arrival directions of radio waves.

SUMMARY OF THE INVENTION In view of the above problems, it is an object of the present invention to provide a technique capable of solving the problem of phase folding and improving the accuracy of the azimuth indicating the arrival direction of radio waves.

A signal processing apparatus according to the present invention is a signal processing apparatus that processes signals received by a plurality of antennas, and is based on signals received by three or more antennas included in the plurality of antennas. a direction calculation unit that separates the for each direction and calculates without phase folding; a first decomposition unit that decomposes the received signals of two or more antennas included in the plurality of antennas into complex signals for each of the directions; A first selection unit that selects one of the phase differences of the plurality of complex signals decomposed by the first decomposition unit and the phase folding of the phase differences for each direction based on the calculation result of the direction calculation unit. and , wherein the antenna spacing of the antennas used for decomposition by the first decomposition unit is wider than the antenna spacing of the antennas used for calculation by the azimuth calculation unit (first configuration).

In the signal processing device having the first configuration, the azimuth calculation unit calculates an azimuth representing an arrival direction of radio waves and phase folding of the azimuth based on received signals from three or more antennas included in the plurality of antennas. a separation unit that separates for each direction; a second decomposition unit that decomposes signals received by two or more antennas included in the plurality of antennas into complex signals for each direction; a second selection unit that selects one of the orientation and the phase fold of the orientation based on the decomposition result of the antenna spacing in the antenna used for separation in the separation unit is the second decomposition A configuration (second configuration) may be adopted in which the antenna spacing is wider than that of the antennas used for decomposition in the first decomposition section and narrower than the antenna spacing of the antennas used for decomposition in the first decomposition section.

In the signal processing device having the first or second configuration, the antenna spacing of the antennas used for decomposition in the first decomposition unit is the longest antenna spacing of the plurality of antennas (third configuration). may

In the signal processing device having any one of the first to third configurations, the plurality of antennas are a plurality of virtual antennas generated by combining a plurality of transmitting antennas and a plurality of receiving antennas (fourth configuration ).

A radar apparatus according to the present invention has a configuration (fifth configuration) including the signal processing device having any one of the first to third configurations and the plurality of antennas.

A direction-of-arrival estimation method according to the present invention is a signal processing method for processing signals received by a plurality of antennas, and represents the direction of arrival of radio waves based on signals received by three or more antennas included in the plurality of antennas. an azimuth calculation step of separating azimuths for each of the azimuths and calculating without phase folding; a decomposition step of decomposing received signals of two or more antennas included in the plurality of antennas into complex signals for each of the azimuths; a selecting step of selecting one of the phase differences of the plurality of complex signals decomposed by the decomposing step and the phase folding of the phase differences, based on the calculation result of the direction calculating step, for each of the The antenna spacing of the antennas used for decomposition in the decomposition step is wider than the antenna spacing of the antennas used for calculation in the azimuth calculation step (sixth configuration).

According to the present invention, it is possible to solve the problem of phase folding and improve the accuracy of the azimuth indicating the arrival direction of radio waves.

Diagram showing the configuration of a radar device A diagram for explaining an antenna included in a radar device. Diagram showing combinations of transmitting antennas and receiving antennas that make up each virtual antenna Functional block diagram showing a first example of the azimuth calculator Flowchart showing schematic operation of the radar device FIG. 4 is a diagram showing the relationship between the phase difference of the received signal for the distance d between the antennas and the azimuth in the azimuth calculation by the azimuth calculation unit; FIG. 10 is a diagram showing the relationship between the phase difference of the complex signal and the orientation determined by the first decomposition unit; A diagram showing the relationship between the direction and the phase difference obtained by converting the phase difference of the complex signal obtained by the first decomposition unit into the phase difference corresponding to the antenna interval d. Functional block diagram showing a second example of the azimuth calculation unit Flowchart showing details of direction calculation processing executed by the direction calculation unit in the second example A diagram showing the relationship between the phase difference of the received signal for the antenna interval of 2d and the azimuth in the azimuth separation in the separation unit. FIG. 10 is a diagram showing the relationship between the phase difference of the complex signal and the orientation determined by the second decomposition unit; A diagram showing the relationship between the direction and the phase difference obtained by converting the phase difference of the received signal obtained by the separation unit into the phase difference corresponding to the antenna interval d.

Exemplary embodiments of the invention are described in detail below with reference to the drawings.

<1. Configuration of Radar Device>
FIG. 1 is a diagram showing the configuration of a radar device 1 according to an embodiment of the present invention. The radar device 1 can be mounted on a moving object such as a vehicle, robot, aircraft, or ship. In this embodiment, the radar device 1 is mounted on a vehicle such as an automobile. Hereinafter, the vehicle on which the radar device 1 is mounted is expressed as the own vehicle.

The radar device 1 is used to detect targets existing around the own vehicle, such as other vehicles, signs, guardrails, and people. The detection result of the target is output to a storage device of the own vehicle, a vehicle ECU (Electronic Control Unit) 5 that controls the behavior of the own vehicle, and the like. Target detection results are used, for example, for vehicle control such as PCS (Pre-crash Safety System) and AEBS (Advanced Emergency Braking System).

As shown in FIG. 1 , the radar device 1 includes multiple transmitters 2 , multiple receivers 3 , and a signal processor 4 . The radar device 1 further includes multiple transmitting antennas 23 and multiple receiving antennas 31 .

The radar device 1 is a so-called MIMO (Multiple-Input and Multiple-Output) radar device. However, unlike the present embodiment, the radar device 1 may be modified to have a single transmitting antenna and a plurality of receiving antennas. The radar device 1 is an FCM (Fast Chirp Modulation) type radar device. However, unlike the present embodiment, the radar device 1 may be changed to a radar device other than the FCM system such as the FMCW (Frequency Modulated Continuous Wave) system.

The transmitter 2 includes a signal generator 21 and a transmitter 22 . The signal generator 21 generates a modulated signal whose voltage changes like a sawtooth wave, and supplies the modulated signal to the oscillator 22 . The transmitter 22 generates a transmission signal, which is a chirp signal, based on the modulated signal generated by the signal generator 21 and outputs it to the transmission antenna 23 .

In this embodiment, the number of transmission antennas 23 is three. The number of transmission units 3 is also three in accordance with the number of transmission antennas 23 . However, the number of transmission antennas 23 may be other than three. The number of transmission units 3 may also be changed according to the number of transmission antennas 23 . Also, the number of transmission antennas 23 and the number of transmission units 2 do not necessarily have to match. For example, one transmission unit 2 may be provided for three transmission antennas 23, and connection between each transmission antenna 23 and the transmission unit 2 may be switched by a switch.

The three transmitting antennas 23 receive transmission signals from different transmission units 2, convert the transmission signals into transmission waves TW, and output the transmission waves TW. The transmission signals output from each of the three transmission units 2 are mutually orthogonal signals (orthogonal signals). Orthogonal means that they do not interfere with each other due to differences in time, phase, frequency, code, or the like.

The receiving section 3 includes a plurality of receiving antennas 31 and a plurality of individual receiving sections 32 . One individual receiver 32 is connected to each receiving antenna 31 . Each receiving antenna 31 receives the reflected wave RW from the target, acquires a received signal, and outputs the received signal to each individual receiving unit 32 . In this embodiment, the receiving section 3 includes two receiving antennas 31 and two individual receiving sections 32 . However, the number of receiving antennas 31 may be other than two. Also, the number of individual receivers 32 may be less than the number of reception antennas 31 by introducing a switch.

Each individual receiver 32 processes the received signal obtained by the corresponding receiving antenna 31 . The individual receiver 32 includes a mixer 33 and an A/D converter 34 . A received signal obtained by the receiving antenna 31 is sent to the mixer 33 after being amplified by a low-noise amplifier (not shown). A transmission signal from each transmitter 22 of each transmission unit 2 is input to the mixer 33 , and the transmission signal and the reception signal are mixed in the mixer 33 . Thereby, a beat signal having a beat frequency that is the difference between the frequency of each 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 .

The signal processing device 4 executes various processes based on each beat signal fetched via each A/D converter 34 . The signal processing device 4 includes a microcomputer including a CPU (Central Processing Unit), a memory 41, and the like. The signal processing device 4 stores various data to be operated on in the memory 41, which is a storage device. The memory 41 is, for example, a RAM (Random Access Memory). The signal processing device 4 includes a transmission control section 42, a conversion section 43, and a data processing section 44 as functions realized by software on a microcomputer. The transmission controller 42 controls the signal generator 21 of each transmitter 2 .

Since the reception antenna 31 receives the reflected waves from a plurality of targets in an overlapping state, the converter 43 converts the beat signal generated based on the received signal into a frequency component based on the reflected waves of each target. process to separate the In this embodiment, the transformation unit 43 separates frequency components by Fast Fourier Transform (FFT) processing. In FFT processing, reception level and phase information are calculated for each frequency point (sometimes referred to as frequency bin) set at predetermined frequency intervals. The conversion unit 43 outputs the result of FFT processing to the data processing unit 44 .

More specifically, the conversion section 43 performs two-dimensional FFT processing on the beat signals output from each A/D conversion section 34 . By performing the first FFT processing, a frequency spectrum is obtained in which peaks appear in frequency bins corresponding to the distance to the target (hereinafter sometimes referred to as distance bins). By arranging the frequency spectrum obtained by the first FFT process in time series and performing the second FFT process, a frequency spectrum in which a peak appears in the frequency bin (hereinafter sometimes referred to as velocity bin) for the Doppler frequency is obtained. The conversion unit 43 obtains a two-dimensional power spectrum with distance bins and speed bins as axes by two-dimensional FFT processing.

The data processor 44 includes a peak extractor 45 , a distance/relative velocity calculator 46 , and an azimuth calculator 47 .

The peak extraction unit 45 detects peaks from the results of FFT processing and the like in the conversion unit 43 . In this embodiment, the peak extraction unit 45 extracts peaks having a power value equal to or greater than a predetermined value based on the two-dimensional power spectrum obtained by two-dimensional FFT processing and having distance bins and speed bins as axes. In addition, in the present embodiment, the peak extraction unit 45 classifies the results of peak extraction into results for each virtual antenna. Virtual antennas will be described later.

The distance/relative velocity calculator 46 derives the distance and relative velocity to the target based on the combination of the distance bins and velocity bins identified by the peak extractor 45 as having peaks.

The azimuth calculation unit 47 focuses on the peaks of the same frequency bin extracted by the peak extraction unit 45 for each virtual antenna, and calculates the azimuth where the target exists (the azimuth indicating the direction of arrival of radio waves) based on the phase information of these peaks. ). When there are a plurality of peaks with different frequency bins, the azimuth calculator 47 performs azimuth estimation for each peak. For direction estimation, known methods such as MUSIC (Mutiple Signal Classification) and ESPRIT (Estimation of Signal Parameters via Rotational Invariance Techniques) are used. The details of the processing of the azimuth calculation unit 47 will be described later.

The target data including the distance to the target, the relative speed of the target, and the direction in which the target exists, which are obtained by the distance/relative speed calculation unit 46 and the direction calculation unit 47, are output to the vehicle ECU 5. .

<2. Virtual Antenna>
FIG. 2 is a diagram for explaining an antenna included in the radar device 1. FIG.

In this embodiment, as shown in FIG. 2(a), the three transmitting antennas 23 are arranged at the same antenna interval 2d along the horizontal direction. That is, the antenna spacing between adjacent transmitting antennas Tx1 and Tx2 and the antenna spacing between adjacent transmitting antennas Tx2 and Tx2 are both 2d. Note that the antenna spacing between adjacent transmitting antennas 23 does not have to be exactly the same between a plurality of sets (two sets for three transmitting antennas 23). It suffices if a plurality of sets can be regarded as identical.

In this embodiment, as shown in FIG. 2(b), the 32 receiving antennas 31 are arranged at an antenna interval d along the horizontal direction. Note that the antenna spacing between adjacent receiving antennas Rx1 and receiving antennas Rx2 may not be exactly half the antenna spacing between two adjacent transmitting antennas 23, and it is necessary to consider design errors and variations. It suffices if it can be regarded as half the antenna interval between two adjacent transmitting antennas 23 .

The antenna spacing d is the same as the half wavelength of the transmitted wave TW and the received wave RW. Note that the antenna spacing d does not have to be exactly the same as the half wavelength of the transmission wave TW and the reception wave RW. It suffices if they can be regarded as identical.

Also, the antenna spacing between the plurality of transmitting antennas 23 and the antenna spacing between the plurality of receiving antennas 31 shown in FIG. 2 are merely examples, and the antenna spacing may be changed as appropriate.

A virtual array antenna shown in FIG. 2(c) is generated by combining the three transmitting antennas 23 shown in FIG. 2(a) and the two receiving antennas 31 shown in FIG. 2(b). The virtual array antenna shown in FIG. 2(c) is composed of virtual antennas VRx1 to VRx6. The virtual antennas VRx1 to VRx6 are arranged at the same antenna spacing d along the horizontal direction. By applying the MIMO technology, it is possible to obtain virtual antennas three times the number of receiving antennas by simply increasing the number of transmitting antennas by two.

FIG. 3 is a diagram showing combinations of transmitting antennas 23 and receiving antennas 31 that constitute each of the virtual antennas VRx1 to VRx6. As shown in FIG. 3, virtual antenna VRx1 is generated by combining transmitting antenna Tx1 and receiving antenna Rx1. A virtual antenna VRx2 is generated by combining the transmitting antenna Tx1 and the receiving antenna Rx2. Virtual antenna VRx3 is generated by combining transmitting antenna Tx2 and receiving antenna Rx1. Virtual antenna VRx4 is generated by combining transmitting antenna Tx2 and receiving antenna Rx2. Virtual antenna VRx5 is generated by combining transmitting antenna Tx3 and receiving antenna Rx1. Virtual antenna VRx6 is generated by combining transmitting antenna Tx3 and receiving antenna Rx2.

That is, the reception signal of the reception antenna Rx1 includes the reception signal of the virtual antenna VRx1, the reception signal of the virtual antenna VRx3, and the reception signal of the virtual antenna VRx5, which are orthogonal to each other. The reception signal of the reception antenna Rx2 includes the reception signal of the virtual antenna VRx2, the reception signal of the virtual antenna VRx4, and the reception signal of the virtual antenna VRx6, which are orthogonal to each other. The signal processing device 4 processes received signals of a plurality of virtual antennas VRx1 to VRx6 generated by combining a plurality of transmitting antennas 23 and a plurality of receiving antennas 31. FIG.

<3. First Example of Direction Calculation Section>
FIG. 4 is a functional block diagram showing a first example of the direction calculator 47. As shown in FIG. In this example, the orientation calculation unit 47 includes an orientation calculation unit 471 , a first decomposition unit 472 and a first selection unit 473 . That is, the signal processing device 4 includes an orientation calculation section 471 , a first decomposition section 472 , and a first selection section 473 .

The azimuth calculation unit 471 separates azimuths representing arrival directions of radio waves for each azimuth based on the received signals of three or more antennas included in the plurality of antennas, and calculates the azimuths without phase folding. In this example, the azimuth calculation unit 471 separates the azimuths representing the arrival directions of radio waves for each azimuth based on the signals received by the three virtual antennas VRx1 to VRx3, and calculates them without phase folding. The details of the processing of the azimuth calculation unit 471 will be described later.

The first decomposing unit 472 decomposes the received signals of two or more antennas included in the plurality of antennas into complex signals for each direction calculated by the direction calculating unit 471 . In this example, the first decomposing unit 472 decomposes the received signals of the virtual antennas VRx1 and VRx6 into complex signals for each direction calculated by the direction calculating unit 471 . A complex signal contains phase information and amplitude information. Details of the processing of the first decomposition unit 472 will be described later.

Note that the antenna spacing of the antennas used for decomposition by the first decomposition unit 472 is wider than the antenna spacing of the antennas used for calculation by the azimuth calculation unit 471 . In this example, the antenna spacing of the antennas used for decomposition by the first decomposition unit 472 is d, whereas the antenna spacing of the antennas used for decomposition by the first decomposition unit 472 is 5d.

In this example, the antenna spacing of the antennas used for decomposition by the first decomposition unit 472 is the longest antenna spacing 5d of the virtual antennas VRx1 to VRx6. By adopting the longest antenna spacing 5d of the virtual antennas VRx1 to VRx6 as the antenna spacing of the antennas used for decomposition by the first decomposition unit 472, it is possible to maximize the accuracy of the azimuth representing the arrival direction of radio waves. . However, the antenna spacing of the antennas used for decomposition by the first decomposition unit 472 may be an antenna spacing other than the longest antenna spacing of 5d among the virtual antennas VRx1 to VRx6, such as 4d.

The first selection unit 473 determines the phase differences of the plurality of complex signals decomposed by the first decomposition unit 472 and the phase differences of the phase differences based on the calculation result of the direction calculation unit 471 for each direction calculated by the direction calculation unit 471 . Select one of the phase wraps. Details of the processing of the first selection unit 473 will be described later. The azimuth calculation unit 47 estimates the azimuth obtained by converting the selection result of the first selection unit 473 as the azimuth where the target exists (the azimuth representing the arrival direction of radio waves).

The azimuth calculation unit 47 uses the “calculation result of the azimuth calculation unit 471” as the selection criterion of the first selection unit 473 . As a result, the azimuth calculation unit 47 can solve the problem of phase wrapping.

In addition, under the condition that the antenna spacing of the antennas used for decomposition by the first decomposition section 472 is wider than the antenna spacing of the antennas used for calculation by the azimuth calculation section 471, the azimuth calculation section 47 is selected as "the phase differences of the plurality of complex signals decomposed by the first decomposing unit 472 and the phase folding of the phase differences". As a result, the azimuth calculation unit 47 can improve the accuracy of the azimuth indicating the direction of arrival of radio waves, compared to the case where the calculation result of the azimuth calculation unit 471 is used as the azimuth indicating the direction of arrival of radio waves.

That is, the signal processing device 4 can solve the problem of phase folding and improve the accuracy of the azimuth indicating the arrival direction of radio waves.

FIG. 5 is a flowchart showing a schematic operation of the radar device 1. As shown in FIG. FIG. 5 mainly shows the processing by the azimuth calculation unit 47 . That is, in FIG. 5, the process of obtaining the distance to the target and the relative velocity are omitted. The radar device 1 periodically repeats the processing shown in FIG. 5 at regular time intervals.

First, the transmission antenna 23 outputs a transmission wave TW (step S1). Next, the receiving antenna 31 receives the reflected wave RW reflected by the target and acquires a received signal (step S2). Next, the signal processing device 4 acquires a predetermined number of beat signals (step S3). Next, FFT processing is performed on the beat signal acquired by the conversion unit 43 (step S4).

Next, the peak extractor 45 performs peak extraction based on the result of FFT processing (step S5). The peak extraction unit 45 classifies the peak extraction results into results for each of the virtual antennas VRx1 to VRx6.

Next, the azimuth calculation unit 471 separates azimuths representing arrival directions of radio waves for each azimuth based on the signals received by the virtual antennas VRx1 to VRx3, and calculates the azimuths without phase folding (step S6). The azimuth calculator 471 calculates a correlation matrix Rxx represented by the following equation (1).
Rxx=E[X(t) ^XH (t)] (1)
here,
X(t)=[ _x1 (t), _x2 (t), _x3 (t)] ^T (2)
is.

X(t) is the received signal vector of virtual antennas VRx1 to VRx3 at time t. E[•] indicates an expected value, H indicates a complex conjugate transpose, and T indicates a transposed matrix. x ₁ (t) is the received signal of the virtual antenna VRx1, x ₂ (t) is the received signal of the virtual antenna VRx2, and x ₃ (t) is the received signal of the virtual antenna VRx3.

The azimuth calculation unit 471 uses the correlation matrix Rxx to calculate the azimuth by known MUSIC, ESPRIT, or the like. Since the correlation matrix Rxx is generated from the received signals of the three virtual antennas VRx1 to VRx3, the azimuth calculator 471 obtains a maximum of two azimuths. Since the antenna interval d of the three virtual antennas VRx1 to VRx3 is the same as the half wavelength of the transmission wave TW and the reception wave RW, phase folding does not occur in the direction calculation by the direction calculation unit 471. FIG.

A case where the orientation calculation unit 471 calculates the orientation D1 and the orientation D2 will be described below. FIG. 6 is a diagram showing the relationship between the phase difference of the received signal for the distance d between the antennas and the azimuth in the azimuth calculation by the azimuth calculator 471 .

Returning to FIG. 5, when the azimuths are calculated by the azimuth calculation unit 471, the first decomposing unit 472 divides the reception signals of the virtual antennas VRx1 and VRx6 into It is decomposed into complex signals (step S7).
X(t)=AS(t) (3)
here,
A=[a(θ ₁ ), a(θ ₂ )] (4)
S(t)=[ _s1 (t), _s2 (t)] ^T (5)
is.

Note that a(θ ₁ ) is the mode vector of direction D1. a(θ ₂ ) is the mode vector of direction D2. θ ₁ is an angle representing direction D1. _θ2 is an angle representing the direction D2. The angle representing the azimuth is 0 [deg] when the target is positioned directly in front of the radar device 1, and becomes larger than 0 [deg] as the target deviates to the right from the front as viewed from the radar device 1. The target becomes smaller than 0 [deg] as it shifts to the left from the front when viewed from the radar device 1 . S(t) is the phase-amplitude vector of the radio wave. s ₁ (t) is the complex signal (phase amplitude signal) of direction D1. s ₂ (t) is the complex signal (phase amplitude signal) of direction D2.

Since direction D1 and direction D2 are calculated by direction calculation section 471, matrix A in equation (3) is already known. Also, the received signal vector X(t) in equation (3) is known. Therefore, the first decomposition section 472 can obtain the phase-amplitude vector S(t) based on Equation (3).

When obtaining the complex signal (phase amplitude signal) s _1VRx1 (t) of direction D1 at virtual antenna VRx1 and the complex signal (phase amplitude signal) s _2VRx1 (t) of direction D2 at virtual antenna VRx1, the first decomposition The unit 472 sets the virtual antenna VRx1 as the reference antenna for the mode vectors a(θ ₁ ) and a(θ ₂ ). Specifically, first decomposition section 472 substitutes the following values for received signal vector X(t) and mode vector a(θ _k ).
X(t)=[x _VRx1 (t), x _VRx6 (t)] ^T (6)
a(θ ₁ )=[1, exp{-jΛ5dsin(θ ₁ )}] ^T (7)
a(θ ₂ )=[1, exp{-jΛ5dsin(θ ₂ )}] ^T (8)
λ=2π/λ (9)

Note that x _VRx1 (t) is the received signal of virtual antenna VRx1 at time t. x _VRx6 (t) is the received signal at time t of virtual antenna VRx6. λ is the wavelength of the transmitted wave TW and the received wave RW.

By the above substitution, a complex signal (phase amplitude signal) s _1VRx1 (t) of direction D1 at virtual antenna VRx1 and a complex signal (phase amplitude signal) s _2VRx1 (t) of direction D2 at virtual antenna VRx1 are obtained.

Further, when obtaining the complex signal (phase amplitude signal) s _1VRx6 (t) of the direction D1 at the virtual antenna VRx6 and the complex signal (phase amplitude signal) s _2VRx6 (t) of the direction D2 at the virtual antenna VRx6, The 1-resolving unit 472 sets the virtual antenna VRx6 as the reference antenna for the mode vectors a(θ ₁ ) and a(θ ₂ ). Specifically, the first decomposition unit 472 substitutes the value of the above equation (6) for the received signal vector X(t), and substitutes the following values for the mode vector a(θ _k ).
a(θ ₁ )=[exp{jΛ5d sin(θ ₁ )}, 1] ^T (10)
a(θ ₂ )=[exp{jΛ5d sin(θ ₂ )}, 1] ^T (11)

By the above substitution, a complex signal (phase amplitude signal) s _1VRx6 (t) of direction D1 at virtual antenna VRx6 and a complex signal (phase amplitude signal) s _2VRx6 (t) of direction D2 at virtual antenna VRx6 are obtained.

FIG. 7 is a diagram showing the relationship between the phase difference of the complex signal obtained by the first decomposition unit 472 and the direction. In FIG. 7 , the phase difference between the complex signal s _1VRx1 (t) in direction D1 at virtual antenna VRx1 and the complex signal s _1VRx6 (t) in direction D1 at virtual antenna VRx6 and the phase folding of the phase difference are indicated by black circles. In FIG. 7 , the phase difference between the complex signal s _2VRx1 (t) in the direction D2 at the virtual antenna VRx1 and the complex signal s _2VRx6 (t) in the direction D2 at the virtual antenna VRx6 and the phase folding of the phase difference are indicated by black squares. show. Since the antenna spacing between the virtual antennas VRx1 and VRx6 is 5d, the phase difference in FIG. 7 is the phase difference for the antenna spacing of 5d.

Returning to FIG. 5, when the complex signal for each of the orientations D1 and D2 is obtained by the first decomposition unit 472, the first selection unit 473 calculates the first One is selected from the phase differences of the multiple complex signals decomposed by the 1 decomposing unit 472 and the phase folding of the phase differences (step S8).

The first selection unit 473 converts the phase difference of the complex signal obtained by the first decomposition unit 472 into a phase difference corresponding to the antenna spacing d, in order to enable comparison with the calculation result of the azimuth calculation unit 471 . FIG. 8 is a diagram showing the relationship between the azimuth and the phase difference obtained by converting the phase difference of the complex signal obtained by the first decomposition unit 472 into the phase difference corresponding to the antenna interval d.

The first selection unit 473 compares the phase difference in FIG. 6 with the phase difference shown in FIG. 8, and selects the one whose phase difference is closest to the phase difference of the black circle shown in FIG. 8 from the black circles shown in FIG. and outputs the orientation D1' of the selected black circle. Direction D1' is a corresponding direction of direction D1 and has a smaller error than direction D1.

Further, the first selection unit 473 compares the phase difference in FIG. 6 with the phase difference shown in FIG. 8, and selects the phase difference from the black squares shown in FIG. Select the closest one and output the orientation D2' of the selected black square. Direction D2' is a corresponding direction of direction D2 and has a smaller error than direction D2.

Returning to FIG. 5, when the selection by the first selection unit 473 is completed, the azimuth calculation unit 47 converts the azimuths D1′ and D2′ output from the first selection unit 473 to the azimuth where the target exists (the Azimuth representing the direction) is estimated, the estimation result is output (step S9), and the flow is terminated.

As is clear from FIGS. 6 and 8, the wider the azimuth angle (the greater the absolute value of the azimuth), the smaller the rate of change in the phase difference with respect to the rate of change in the azimuth. influence becomes greater. Therefore, the signal processing device 4 can greatly improve the accuracy of the azimuth in the wide-angle range of the azimuth.

In the azimuth calculation unit 471, if a set of antennas with an antenna interval d wider than half the wavelength of the transmission wave TW and the reception wave RW is used, phase folding occurs. For example, there is a case where the azimuth is calculated by a combination of VRx1, VRx3, and VRx5. At this time, there is still the possibility that the direction D1 is not the direction calculated by the direction calculation unit 47 but is another direction due to phase folding. The same applies to the orientation D2. In this case, before calculating the mode vector in equations (7) and (8), it is necessary to eliminate the phase folding and fix the orientation. It is preferable to use various known phase wrap determination methods to cancel the phase wrap. After canceling the phase folding and determining the orientations D1 and D2, the mode vector may be calculated by substituting θ1 and θ2 into the equations (7) and (8).

<4. Second Example of Direction Calculation Section>
FIG. 9 is a functional block diagram showing a second example of the azimuth calculator 47. As shown in FIG. In this example, the orientation calculation unit 47 includes an orientation calculation unit 471 , a first decomposition unit 472 , and a first selection unit 473 as in the first example. That is, the signal processing device 4 includes an orientation calculation section 471 , a first decomposition section 472 , and a first selection section 473 .

The orientation calculation unit 471 in this example differs from the orientation calculation unit 471 in the first example in that it includes a separation unit 471a, a second decomposition unit 471b, and a second selection unit 471c.

The separation unit 471a separates, for each direction, an azimuth representing an arrival direction of radio waves and a phase folding of the azimuth based on the received signals of three or more antennas included in the plurality of antennas. In this example, the separation unit 471a separates the direction representing the arrival direction of radio waves and the phase folding of the direction for each direction based on the received signals of the three virtual antennas VRx1, VRx3, and VRx5. Details of the processing of the separation unit 471a will be described later.

The second decomposing unit 471b decomposes the reception signals of two or more antennas included in the plurality of antennas into complex signals for each direction separated by the separating unit 471a. In this example, the second decomposing unit 471b decomposes the received signals of the virtual antennas VRx1 and VRx2 into complex signals for each direction separated by the separating unit 471a. Details of the processing of the second decomposition unit 471b will be described later.

In this example, the antenna spacing of the antennas used for separation by the separation unit 471a is wider than the antenna spacing of the antennas used for decomposition by the second decomposition unit 471b, and the antenna spacing used for decomposition by the first decomposition unit 472 narrower than the antenna spacing in In this example, the antenna spacing of the antennas used for separation in the separation unit 471a is 2d, whereas the antenna spacing of the antennas used for decomposition in the second decomposition unit 471b is d. Further, in this example, the antenna spacing of the antennas used for separation in the separation unit 471a is 2d, whereas the antenna spacing of the antennas used for decomposition in the first decomposition unit 472 is 5d.

The second selection unit 471c selects one of the orientations separated by the separation unit 471a and the phase folding of the orientations based on the decomposition result of the second decomposition unit 471b for each orientation separated by the separation unit 471a. do. Details of the processing of the second selection unit 471c will be described later.

FIG. 5 is a flow chart showing the schematic operation of the radar device 1 when the azimuth calculation unit 47 in this example is used, as in the case where the azimuth calculation unit 47 in the first example is used. However, when the azimuth calculation unit 47 in this example is used, the processing contents of the azimuth calculation (step S6 in FIG. 5) are different from when the azimuth calculation unit 47 in the first example is used.

FIG. 10 is a flow chart showing the contents of the orientation calculation processing executed by the orientation calculation unit 47 in this example.

First, based on the signals received by the virtual antennas VRx1, VRx3, and VRx5, the separation unit 471a separates the directions representing the arrival directions of radio waves and the phase folding of the directions for each direction (step S61). The separation unit 471a calculates a correlation matrix Rxx' represented by the following equation (12).
Rxx'=E[X(t) ^XH (t)] (12)
here,
X(t)=[ _x1 (t), _x3 (t), _x5 (t)] ^T (13)
is.

x ₁ (t) is the received signal of the virtual antenna VRx1, x ₃ (t) is the received signal of the virtual antenna VRx3, and x ₅ (t) is the received signal of the virtual antenna VRx5.

The separation unit 471a uses the correlation matrix Rxx' to separate the azimuths by known MUSIC, ESPRIT, or the like. Since the correlation matrix Rxx' is generated from the received signals of the three virtual antennas VRx1, VRx3, and VRx5, the demultiplexing section 471a obtains a maximum of two azimuths. Since the antenna spacing 2d between the three virtual antennas VRx1, VRx3, and VRx5 is longer than half the wavelength of the transmission wave TW and the reception wave RW, phase folding occurs in the directions separated by the separation unit 471a.

A case where the separating unit 471a separates the direction of arrival of radio waves into direction D1 and direction D2 will be described below. As described above, phase folding occurs in the second example, so unlike the first example, the azimuth D1 is not determined as one azimuth. The direction D2 is also the same. FIG. 11 is a diagram showing the relationship between the phase difference of the received signal corresponding to the antenna interval of 2d and the azimuth in the azimuth separation in the separation unit 471a. As shown in the drawing, there are a plurality of orientations corresponding to the same phase difference, and the separation unit 471a calculates any one of the plurality of orientations corresponding to the phase difference.

Returning to FIG. 10, when the azimuths are separated by the separating unit 471a, the second decomposing unit 471b divides the reception signals of the virtual antennas VRx1 and VRx2 into complex signals for each of the azimuths D1 and D2 based on the following equation (14). It decomposes into signals (step S62). Equations (14) to (16) below are the same as equations (3) to (5) described above.
X(t)=AS(t) (14)
here,
A=[a(θ ₁ ), a(θ ₂ )] (15)
S(t)=[ _s1 (t), _s2 (t)] ^T (16)
is.

Note that a(θ ₁ ) is the mode vector of direction D1. a(θ ₂ ) is the mode vector of direction D2. θ ₁ is an angle representing direction D1. _θ2 is an angle representing the direction D2. The angle representing the azimuth is 0 [deg] when the target is positioned directly in front of the radar device 1, and becomes larger than 0 [deg] as the target deviates to the right from the front as viewed from the radar device 1. The target becomes smaller than 0 [deg] as it shifts to the left from the front when viewed from the radar device 1 . S(t) is the phase-amplitude vector of the radio wave. s ₁ (t) is the complex signal (phase amplitude signal) of direction D1. s ₂ (t) is the complex signal (phase amplitude signal) of direction D2.

Since direction D1 and direction D2 have been obtained by separating section 471a, matrix A in equation (14) is known. Also, the received signal vector X(t) in equation (14) is known. Therefore, the second decomposition unit 471b can obtain the phase-amplitude vector S(t) based on Equation (14).

When obtaining the complex signal (phase amplitude signal) s _1VRx1 (t) of direction D1 at virtual antenna VRx1 and the complex signal (phase amplitude signal) s _2VRx1 (t) of direction D2 at virtual antenna VRx1, the first decomposition The unit 472 sets the virtual antenna VRx1 as the reference antenna for the mode vectors a(θ ₁ ) and a(θ ₂ ). Specifically, the second decomposition unit 471b substitutes the following values for the received signal vector X(t) and the mode vector a(θ _k ).
X(t)=[x _VRx1 (t), x _VRx3 (t)] ^T (17)
a(θ ₁ )=[1, exp{-2jΛdsin(θ ₁ )}] ^T (18)
a(θ ₂ )=[1, exp{−2jΛdsin(θ ₂ )}] ^T (19)
λ=2π/λ (20)

Note that x _VRx1 (t) is the received signal of virtual antenna VRx1 at time t. x _VRx3 (t) is the received signal at time t of virtual antenna VRx3. λ is the wavelength of the transmitted wave TW and the received wave RW. In the first example described above, the phase folding is canceled before the mode vector is calculated to fix the azimuth to one direction, but this is not necessary in the second example. Therefore, when generating the signal vector X(t) and the mode vector a(θk), virtual antennas are selected according to the antenna spacing used in the separation unit 471a. For example, in the above case, the virtual antenna VRx3 is selected for the virtual antenna VRx1 for which the complex signal is to be obtained, according to the antenna spacing 2d used in the separation unit 471a.

At this time, the received signal vector X(t) is generated using the received signal of the selected virtual antenna as an element. In addition, the mode vector a(θ) is such that the element corresponding to the reference antenna (virtual antenna VRx1 in the above) is 1, and the other elements are the ideal phase differences at the positions of the other corresponding virtual antennas. Generate.

By the above substitution, a complex signal (phase amplitude signal) s _1VRx1 (t) of direction D1 at virtual antenna VRx1 and a complex signal (phase amplitude signal) s _2VRx1 (t) of direction D2 at virtual antenna VRx1 are obtained.

Further, when obtaining the complex signal (phase amplitude signal) s _1VRx2 (t) of the direction D1 at the virtual antenna VRx2 and the complex signal (phase amplitude signal) s _2VRx2 (t) of the direction D2 at the virtual antenna VRx2, The 2-factor decomposition unit 471b sets the virtual antenna VRx2 as the reference antenna for the mode vectors a(θ ₁ ) and a(θ ₂ ). Specifically, the second decomposition unit 471b substitutes the following values for the received signal vector X(t) and the mode vector a(θ _k ). Note that, as described above, the virtual antenna VRx4 with an antenna interval of 2d is selected for the virtual antenna VRx2.
X(t)=[x _VRx2 (t), x _VRx4 (t)] ^T (21)
a(θ ₁ )=[1, exp{-2jΛdsin(θ ₁ )}] ^T (22)
a(θ ₂ )=[1, exp{-2jΛdsin(θ ₂ )}] ^T (23)

By the above substitution, a complex signal (phase amplitude signal) s _1VRx2 (t) of direction D1 at virtual antenna VRx2 and a complex signal (phase amplitude signal) s _2VRx2 (t) of direction D2 at virtual antenna VRx2 are obtained.

FIG. 12 is a diagram showing the relationship between the phase difference of the complex signal obtained by the second decomposition unit 471b and the azimuth. In FIG. 12 , the phase difference between the complex signal s _1VRx1 (t) in direction D1 at virtual antenna VRx1 and the complex signal s _1VRx2 (t) in direction D1 at virtual antenna VRx2 and the phase folding of the phase difference are indicated by black circles. In FIG. 12 , the phase difference between the complex signal s _2VRx1 (t) in the direction D2 at the virtual antenna VRx1 and the complex signal s _2VRx2 (t) in the direction D2 at the virtual antenna VRx2 and the phase folding of the phase difference are indicated by black squares. show. Since the antenna spacing between the virtual antenna VRx1 and the virtual antenna VRx2 is d, the phase difference in FIG. 12 is the phase difference for the antenna spacing d.

Returning to FIG. 10, when the complex signal for each of the directions D1 and D2 is obtained by the second decomposing unit 471b, the second selecting unit 471c, based on the decomposing result of the second decomposing unit 471b, for each of the directions D1 and D2, One is selected from the directions separated by the separation unit 471a and the phase folding of the directions (step S63).

The second selection unit 471c converts the phase difference of the received signal obtained by the separation unit 471a into a phase difference corresponding to the antenna interval d so as to enable comparison with the decomposition result of the second decomposition unit 471b. FIG. 13 is a diagram showing the relationship between the azimuth and the phase difference obtained by converting the phase difference of the received signal obtained by the separation unit 471a into the phase difference corresponding to the antenna interval d.

The second selection unit 471c compares the phase difference in FIG. 11 with the phase difference shown in FIG. 13, and selects the one whose phase difference is closest to the phase difference of the black circle shown in FIG. 12 from among the black circles shown in FIG. and outputs the orientation D1 of the selected black circle.

Further, the second selection unit 471c compares the phase difference in FIG. 11 with the phase difference shown in FIG. 13, and selects the phase difference from the black squares shown in FIG. Select the closest one and output the orientation D2 of the selected black square.

Returning to FIG. 10, when the selection by the second selection unit 471c is completed, the orientation calculation unit 471 outputs the orientations D1 and D2 output from the second selection unit 471c as the orientation calculation results (step S64). End the flow.

When the azimuth calculation unit 47 in this example is used, the accuracy of the azimuth calculated in step S6 in FIG. 5 is higher than when the azimuth calculation unit 47 in the first example is used. It is possible to reduce the possibility of making an error in the selection in .

上記の式を用いることで、算出精度を向上することができる。 Although the complex signal is calculated with two virtual antennas in Equations (17) to (19), it may be calculated by substituting the following values using three virtual antennas.

By using the above formula, the calculation accuracy can be improved.

上記の式を用いることで、算出精度を向上することができる。また、上記では３つとしたが、仮想アンテナの構成に応じ、３つ以上の仮想アンテナを用いて複素信号を算出してもよい。 Similarly, formulas (21) to (23) may also be calculated by substituting values as follows.

By using the above formula, the calculation accuracy can be improved. Although three virtual antennas are used in the above description, three or more virtual antennas may be used to calculate the complex signal according to the configuration of the virtual antennas.

<5. Notes>
The configurations of the embodiments and examples herein are merely illustrative of the present invention. The configurations of the embodiments and modifications may be changed as appropriate without departing from the technical idea of the present invention. Also, multiple embodiments and modifications may be implemented in combination within a possible range.

Although the vehicle-mounted radar system has been described above, the present invention may also be applied to infrastructure radar systems installed on roads, ship monitoring radar systems, aircraft monitoring radar systems, and the like.

All or part of the functions described above as being implemented in software by executing a program may be implemented in an electrical hardware circuit. Also, all or part of the functions described as being realized by hardware circuits may be realized by software. Also, the functions described as one block may be implemented by cooperation of software and hardware. Also, each functional block is a conceptual component. Functions executed by each functional block may be distributed among a plurality of functional blocks, or functions possessed by a plurality of functional blocks may be combined into one functional block.

1 Radar Device 4 Signal Processing Device 23 Transmitting Antenna 31 Receiving Antenna 471 Direction Calculating Unit 471a Separating Unit 471b Second Decomposing Unit 471c Second Selecting Unit 472 First Decomposing Unit 473 First Selecting Unit VRx1 to VRx6 Virtual Antenna

Claims (4)

A radar device that detects a target by transmitting radio waves and receiving radio waves reflected from the target with a plurality of antennas,
a separation unit that separates an azimuth representing an arrival direction of radio waves and phase folding of the azimuth for each of the azimuths based on signals received by three or more antennas included in the plurality of antennas;
A second decomposition that decomposes a received signal of two or more antennas included in the plurality of antennas, the antenna spacing being narrower than the antenna spacing of the antennas used for separation in the separation unit, into complex signals for each of the directions. Department and
converting the phase difference between the azimuth separated by the separating unit and the phase difference of the received signal corresponding to the phase folding of the azimuth into a phase difference corresponding to the antenna interval in the antenna used in the second decomposition unit; a second selection unit that selects a phase difference closest to the phase difference of the complex signal obtained by the second decomposition unit from among the phase differences after conversion;
an orientation calculation unit that calculates the orientation corresponding to the phase difference selected by the second selection unit as an orientation that is not the phase folding;
Two or more antennas included in the plurality of antennas, wherein the received signals of the antennas wider than the antenna spacing of the antennas used for separation in the separation unit are calculated for each of the directions calculated by the direction calculation unit a first decomposition unit that decomposes into a complex signal of
converting the phase difference of the complex signal for each of the directions decomposed by the first decomposing unit and folding of the phase difference into a phase difference corresponding to the antenna interval in the antenna used in the second decomposing unit; a first selection unit that selects the phase difference closest to the phase difference corresponding to the direction calculated by the direction calculation unit from among the subsequent phase differences;
and estimating an azimuth corresponding to the phase difference selected by the first selector as an azimuth in which the target exists. 2. The radar apparatus according to claim 1 , wherein the antenna spacing of the antennas used for decomposition in said first decomposition unit is the longest antenna spacing of said plurality of antennas. The plurality of antennas are a plurality of virtual antennas generated by combining a plurality of transmitting antennas and a plurality of receiving antennas, and the antenna spacing of the antennas used in the second decomposition unit is the plurality of virtual antennas. 3. The radar system according to claim 1 or 2 , wherein the minimum distance between . A signal processing method for processing signals received by a plurality of antennas in a radar device that detects the target by transmitting radio waves and receiving the radio waves reflected from the target by a plurality of antennas,
a separation step of separating a direction representing an arrival direction of radio waves and a phase folding of the direction for each direction based on signals received by three or more antennas included in the plurality of antennas;
A second decomposition for decomposing a received signal of two or more antennas included in the plurality of antennas, wherein the antenna spacing is narrower than the antenna spacing of the antennas used for separation in the separation step, into complex signals for each of the directions. process and
The azimuth separated in the separation step and the phase difference of the received signal corresponding to the phase folding of the azimuth are converted into a phase difference corresponding to the antenna interval in the antenna used in the second decomposition step, and each of the azimuths a second selection step of selecting the phase difference closest to the phase difference of the complex signal obtained in the second decomposition step from among the phase differences after conversion;
an orientation calculation step of calculating the orientation corresponding to the phase difference selected in the second selection step as an orientation that is not the phase folding;
Two or more antennas included in the plurality of antennas, wherein the received signal of the antennas wider than the antenna spacing of the antennas used for separation in the separation step is calculated for each of the directions calculated in the direction calculation step a first decomposition step of decomposing into a complex signal of
converting the phase difference of the complex signal for each of the directions decomposed by the first decomposition step and the folding of the phase difference into a phase difference corresponding to the antenna interval in the antenna used in the second decomposition unit; a first selection step of selecting the phase difference closest to the phase difference corresponding to the orientation calculated by the orientation calculation unit from among the subsequent phase differences;
and estimating the direction corresponding to the phase difference selected in the first selection step as the direction in which the target exists.

Images