JP2023162528A - Radar device - Google Patents

Abstract

To obtain a radar device capable of excellently estimating the number of incoming waves even when intensity of a received signal is increased, and calculating an incoming angle with high accuracy.SOLUTION: A radar device 100 receives a reflection signal reflected by an object by a plurality of receiving antennas 6, discriminates an eigenvalue of a correlation matrix of a complex spectrum of a frequency-analyzed reflected signal into a noise eigenvalue and a signal eigenvalue by using two threshold values, estimates the number of incoming signals from the number of discriminated signal eigenvalues, and calculates an incoming angle based on the incoming signals and the estimated signal eigenvalue.SELECTED DRAWING: Figure 4

Description

The present application relates to a radar device.

2. Description of the Related Art A radar device that uses a plurality of receivers to determine the angle of arrival of a plurality of received signals is known (see, for example, Patent Document 1). In Patent Document 1, a noise level is calculated from frequency analysis results of a plurality of received signals by a plurality of receivers, and a covariance matrix is generated by extracting a peak signal of a target object for each receiver based on the noise level. However, it is disclosed that each eigenvalue of a covariance matrix is distinguished into a signal eigenvalue and a noise eigenvalue by a threshold value, and the number of incident signals (number of arriving waves) is estimated from the number of signal eigenvalues.

Patent No. 4351266

However, in a radar device equipped with a plurality of receivers, an amplitude error and a phase error are imparted to the received signal due to electromagnetic coupling between the receivers or by components constituting the radar device. This amplitude error and phase error affect the eigenvalues, and when the received signal strength is greater than the noise level, the number of eigenvalues exceeding the threshold determined from the noise level increases, leading to a risk of erroneously estimating the number of arriving waves.

Therefore, there is a need for a technique that does not erroneously estimate the number of arriving waves of a received signal even when the received signal strength is high.

This application discloses a technique for solving the above-mentioned problems, which maintains good arriving wave number estimation accuracy when the received signal strength is low, and also achieves good accuracy when the received signal strength increases. The present invention aims to provide a radar device that can estimate the number of arriving waves and calculate the angle of arrival with high accuracy.

The radar device disclosed in this application includes:
a transmitting antenna for transmitting a transmitting signal;
a plurality of receiving antennas that receive reflected signals reflected by one or more objects;
A radar device comprising: a control unit that frequency-analyzes the reflected signals received by a plurality of the receiving antennas and calculates an angle of arrival, which is an angle with respect to the own device, for each of the objects,
The control unit includes:
a noise level calculation unit that calculates a noise level from the amplitude value of the complex spectrum obtained by the frequency analysis;
a peak value determination unit that determines a peak value based on the amplitude value of the complex spectrum obtained by the frequency analysis and the noise level calculated by the noise level calculation unit;
a correlation matrix calculation unit that calculates a correlation matrix for the complex spectrum determined to be the peak value;
an eigenvalue calculation unit that calculates an eigenvalue of the correlation matrix;
an arriving wave number estimation threshold calculation unit that calculates a threshold for discriminating the eigenvalue into a noise eigenvalue caused by noise and a signal eigenvalue corresponding to the object;
The number of arriving signals is estimated from the number of determined signal eigenvalues based on the threshold calculated by the arriving wave number estimation threshold calculation unit, and the arrival angle is determined based on the incoming signal and the estimated signal eigenvalue. an arrival angle calculation unit that calculates,
The arriving wave number estimation threshold calculation unit includes:
calculating a first arrival wave number estimation threshold based on the noise level calculated by the noise level calculation unit, and calculating a second arrival wave number estimation threshold based on the eigenvalue calculated by the eigenvalue calculation unit;
The arrival angle calculation unit is configured to estimate the number of arriving signals from the number of signal eigenvalues determined using the first arriving wave number estimation threshold and the second arrival wave number estimation threshold.

According to the radar device disclosed in the present application, since the signal eigenvalue is determined using two threshold values, good incoming wave number estimation accuracy is maintained when the received signal strength is small, and when the received signal strength becomes large. Therefore, it is possible to obtain a radar device that can estimate the number of arriving waves well and can calculate the angle of arrival with high accuracy.

1 is a schematic configuration diagram of a radar device according to Embodiment 1. FIG. FIG. 3 is a functional block diagram of a signal processing unit of the radar device according to the first embodiment. FIG. 3 is a hardware configuration diagram of a control unit of the radar device according to the first embodiment. FIG. 3 is a flowchart illustrating a processing procedure for calculating an angle of arrival in the radar device according to the first embodiment. FIG. 3 is a diagram showing amplitude calculation results of a complex spectrum in the radar device according to the first embodiment. FIG. 5A is a diagram showing the relationship between distance frequency and amplitude in FIG. 5A. FIG. 5A is a diagram showing the relationship between relative velocity frequency and amplitude in FIG. 5A. FIG. 2 is a conceptual diagram showing subarrays during correlation matrix calculation in the radar device according to the first embodiment. FIG. 3 is a diagram showing the relationship between eigenvalues and arrival angles in the radar device according to the first embodiment. 7 is a flowchart illustrating another processing procedure for calculating the angle of arrival in the radar device according to the first embodiment. FIG. FIG. 3 is a diagram showing the relationship between the eigenvalue, the arrival wave number estimation threshold, and the received signal strength in the radar device according to the first embodiment.

Embodiments of the radar device disclosed in this application will be described below with reference to the drawings. Note that the radar device according to this embodiment is mounted on a vehicle and is assumed to detect objects existing around the vehicle, but it is assumed that the radar device according to the present embodiment is installed in a vehicle and detects objects existing around the vehicle. device, etc.). In each figure, the same reference numerals indicate the same or corresponding parts.

Embodiment 1.
Below, a radar device according to Embodiment 1 will be explained using figures.
FIG. 1 is a schematic configuration diagram of a radar device according to a first embodiment. Examples of the radar system include an FM-CW (Frequency Modulated Continuous Wave) system, an FCM (Fast Chirp Modulation) system, and a pulse Doppler system, but the first embodiment is applicable to any of the systems. In the first embodiment, an FCM frequency modulation radar device 100 (hereinafter simply referred to as "radar device") will be described as an example.

In FIG. 1, a case is illustrated in which the radar device 100 receives reflected signals as reception signals for six channels, but the number of channels is not limited to six. The radar device 100 includes a signal processing section 1, a control voltage generator 2, a VCO (Voltage Controlled Oscillator) 3, a distributor 4, a transmitting antenna 5, a receiving antenna 6, a mixer 7, an A/D converter 8, and an FFT calculation. (Fast Fourier Transform) 9. Here, the A/D converter 8, the FFT calculation section 9, and the signal processing section 1 constitute a control section 10. Note that the number of transmitting antennas 5 may be plural. In addition, the receiving antenna 6, mixer 7, A/D converter 8, and FFT calculation unit 9 each correspond to 6 channels, but for example, the receiving antenna 6 is referred to collectively as receiving antenna 6, and individually When shown, it is written as receiving antenna 6_n (n is a natural number from 1 to 6). The mixer 7, A/D converter 8, and FFT calculation section 9 are also expressed in the same manner.

Next, the operation of radar device 100 according to the first embodiment will be explained. First, the signal processing section 1 outputs a modulation start command to the control voltage generator 2. Control voltage generator 2 generates a desired voltage waveform and applies it to VCO 3 in response to the modulation start command. The VCO 3 outputs a frequency-modulated transmission signal according to the control voltage. The output transmission signal is distributed by the distributor 4 to the transmission antenna 5 and six mixers 7_1 to 7_6.

Transmission antenna 5 radiates a transmission signal toward a target object (not shown). The signals reflected by the target object are received by six receiving antennas 6_1 to 6_6 as received signals for six channels (referred to as CH1 to CH6). Note that when using a plurality of transmitting antennas, it is possible to obtain as many received signals as the product of the number of transmitting antennas and the number of receiving antennas. This technology is called MIMO (Multiple Input Multiple Output) technology, and the virtual receiving antenna formed by MIMO technology is also simply referred to as "receiving antenna." That is, in the following embodiments, the "receiving antenna" is not limited to the actually arranged "receiving antenna" but may include a "virtual receiving antenna".

Mixers 7_1 to 7_6 provided individually corresponding to the six receiving antennas 6_1 to 6_6 mix the respective received signals received by the receiving antennas 6_1 to 6_6 and the transmitted signals distributed by the distributor 4. The signals are mixed and beat signals for six channels are generated.

Each mixed beat signal is converted into digital data by A/D converters 8_1 to 8_6, respectively.

The digital data converted by the A/D converters 8_1 to 8_6 is subjected to frequency analysis using FFT by FFT calculation units 9_1 to 9_6, respectively. The complex spectra calculated by the FFT calculation units 9_1 to 9_6 and which are the frequency analysis results for six channels are input to the signal processing unit 1.

FIG. 2 is a functional block diagram showing the configuration of the signal processing section 1. As shown in FIG. The signal processing unit 1 calculates the arrival angle of the target object based on the calculated complex spectrum. A technical feature of this embodiment is that the number of arriving waves can be estimated accurately by setting a threshold for estimating the number of arriving waves in consideration of the influence of the amplitude error and phase error of each receiving antenna 6 on the eigenvalue. Therefore, as shown in FIG. 2, the signal processing unit 1 includes an amplitude calculation unit 101 that calculates the amplitude of a complex spectrum, a noise level calculation unit 102 that calculates the noise level in the amplitude of the calculated complex spectrum, and a noise level calculation unit 102 that calculates the noise level in the amplitude of the calculated complex spectrum. A peak value determination unit 103 determines the peak value from the amplitude of the spectrum; a distance and relative velocity calculation unit 104 determines the distance and relative velocity of the target object from the determined peak value; a subarray is formed from a plurality of receiving antennas; A correlation matrix calculation unit 105 calculates a correlation matrix (covariance matrix) of the determined complex spectrum, an eigenvalue calculation unit 106 calculates eigenvalues from the correlation matrix, and a threshold value is calculated for discriminating eigenvalues into signal eigenvalues and noise eigenvalues. an incoming wave number estimation threshold calculation unit 107 for estimating the number of arriving waves; an eigenvalue determination unit 108 for discriminating between signal eigenvalues and noise eigenvalues based on the incoming wave number estimation threshold; An angle calculation unit 109 is provided. Specific processing by each of the functional units 101 to 109 of the signal processing unit 1 will be described later.

FIG. 3 is a diagram showing an example of the hardware configuration of the control unit 10 of the radar device 100. As shown in FIG. 10, the control unit 10 serves as a processing circuit, and includes an arithmetic processing unit 11 (computer) such as a DSP (Digital Signal Processor), a storage device 12 that exchanges data with the arithmetic processing unit 11, and an arithmetic processing unit 11. It includes an A/D converter 8 that inputs a beat signal, a D/A converter 14 that outputs a command value of a frequency modulation signal from an arithmetic processing unit 11 to a control voltage generator 2, a communication circuit 13, and the like.

The arithmetic processing unit 11 includes a CPU (Central Processing Unit), an ASIC (Application Specific Integrated Circuit), an IC (Integrated Circuit), an FPGA (Field Programmable Gate Array), various logic circuits, various signal processing circuits, etc. You can. Further, a plurality of arithmetic processing units 11 of the same type or different types may be provided, and each process may be shared and executed. As the storage device 12, a RAM (Random Access Memory) configured to be able to read and write data from the arithmetic processing unit 11, a ROM (Read Only Memory) configured to be able to read data from the arithmetic processing unit 11, etc. It is equipped. For example, when the radar device 100 is mounted on a vehicle, the communication circuit 13 is connected to an external control device (not shown) such as a vehicle control device via a communication line, and is connected to a CAN (Controller Area Network). Wired communication is performed based on communication protocols such as (Trademark)).

The functional units 101 to 109 of the FFT calculation unit 9, the transmission signal generation unit (not shown), and the signal processing unit 1 included in the control unit 10 are stored in a storage device 12 such as a ROM. This is realized by executing the software (program) and cooperating with each hardware of the control unit 10 such as the storage device 12, the A/D converter 8, the D/A converter 14, and the communication circuit 13. Note that the setting data used by the signal processing section 1 is stored in a storage device 12 such as a ROM as part of software (program).

Next, specific processing by the signal processing unit 1, that is, estimating the number of arriving waves with high accuracy and calculating an angle of arrival based on the estimation, will be described based on a flowchart.
FIG. 4 is a flowchart showing the process of calculating the angle of arrival in the signal processing unit 1 of the radar device 100 according to the first embodiment.
First, in step S1, the amplitude calculation unit 101 calculates the amplitude of each of the complex spectra of the six channels of beat signals input from the FFT calculation unit 9.

5A to 5C are diagrams showing amplitude calculation results of a complex spectrum in amplitude calculation section 101 of radar device 100 according to the first embodiment. 5A is a diagram showing the amplitude corresponding to the distance frequency and the relative velocity frequency, FIG. 5B is a diagram showing the relationship between the distance frequency and the amplitude in FIG. 5A, and FIG. 5C is a diagram showing the relative velocity frequency in FIG. 5A. FIG. 3 is a diagram showing the relationship between and amplitude. By calculating the square of the absolute value of each complex spectrum for each channel, amplitude values corresponding to the distance frequency and relative velocity frequency as shown in FIGS. 5A to 5C can be obtained.

Next, in step S2, the noise level calculating section 102 calculates a noise level _Pn as a representative value of the noise amplitude values. However, the representative value of the noise amplitude value shall be obtained by calculating the average value, median value, mode value, etc. As a specific calculation example, the amplitude values obtained in the previous step S1 are added for all channels, and the average value P _a of the amplitude values on the relative velocity frequency axis is determined for each distance frequency. A value obtained by multiplying P _a by a predetermined value is set as a temporary threshold T _t , amplitude values below the temporary threshold T _t are considered to be noise, and an average value P _a ′ of only the noise is determined again. The above calculation is repeated and the converged average value is defined as the noise level _Pn . In this case, the noise level P _n is calculated for each distance frequency.

Next, in step S3, the peak value determination unit 103 determines a peak signal from among the amplitude values obtained in the previous step S1. First, the peak signal discrimination threshold value A _th is calculated as shown in equation (1).
A _th = α・P _n ...(1)
Here, α is a constant determined by a predetermined false alarm probability.
An amplitude value that is larger than A _th and larger than the amplitude values of the preceding and following frequencies is determined to be the peak value corresponding to the target object.

Next, in step S4, the distance and relative velocity calculation unit 104 calculates the distance and relative velocity of the target object from the distance frequency and relative velocity frequency corresponding to the peak value obtained in step S3. Furthermore, the number of detected target objects is counted and stored in the storage device 12.

Next, in step S5, the correlation matrix calculation unit 105 calculates a correlation matrix for the complex spectrum determined to have a peak value in step S4. M arbitrary-shaped subarrays are extracted from a plurality of receiving antennas 6, and the subarrays are averaged, and an averaged correlation matrix _Rxx is calculated using the following equation (2). Here, the arbitrary shape means that a subarray is formed by a set of different antennas from the plurality of receiving antennas 6. Further, the method of forming the subarray is a method of calculating a spatial average, and is not limited to this method, but may be a time average using complex spectra at different times. For example, correlation matrices corresponding to complex spectra determined to be peak values obtained by frequency analysis from signals received from the same object at different times are computed, and the averaged correlation matrices are calculated by averaging them. Just use it.

ここで、ｘ_ｍは複素スペクトラムからなるベクトルで、式（３）のように表される。

Here, x _m is a vector consisting of a complex spectrum, and is expressed as in equation (3).

Ｋはサブアレイ内の複素スペクトラムの総数であり、Ｈは複素共役転置、Ｔは転置を示す。

K is the total number of complex spectra in the subarray, H indicates complex conjugate transposition, and T indicates transposition.

Here, subarray extraction will be explained using an example. FIG. 6 is a conceptual diagram showing subarray extraction during correlation matrix calculation in radar device 100 according to the first embodiment. Regarding the receiving antennas 6_1 to 6_6 shown in FIG. 1, subarray_1 is composed of receiving antennas 6_1 to 6_4, subarray_2 is composed of receiving antennas 6_2 to 6_5, and subarray_3 is composed of receiving antennas 6_3 to 6_5. This is an example configured with a receiving antenna 6_6. Therefore, in FIG. 5, the number of subarrays M=3, and the total number of complex spectra in the subarrays K=4 (M and K are each natural numbers). Formation of this subarray may include not only the receiving antenna 6 but also a virtual receiving antenna as described above.

Next, in step S6, the eigenvalue calculation unit 106 analyzes the eigenvalues and eigenvectors of the correlation matrix R _xx to obtain eigenvalues λ ₁ to λ _K. However, the eigenvalues are sorted in descending order of magnitude as shown in equation (4).
λ ₁ ≧ λ ₂ ≧...≧ λ _K ... (4)

The eigenvalue λ _n corresponding to the noise level can be expressed as in equation (5) using a coefficient _cm based on the method of generating the correlation matrix R _xx .
λ _n = _cm・P _n ...(5)

In addition, when the signal strength of the reflected signal is large, each eigenvalue λ _k ′(θ) at the time of one wave incidence is a constant multiple of λ ₁ ′(θ) due to the amplitude error and phase error of the antenna, as shown in equation (6). It can be expressed as r _k (θ).

Here, θ indicates the angle of arrival. FIG. 7 is a diagram showing the relationship between eigenvalues and arrival angles. r _k (θ) shown in equation (6) varies depending on the arrival angle.

Next, in step S7, the arriving wave number estimation threshold calculation unit 107 calculates the first arriving wave number estimation threshold λ _th1 . Specifically, the first arriving wave number estimation threshold λ _th1 is calculated as shown in equation (7) using equation (5).
λ _th1 = λ _n + λ _s ...(7)
Here, λ _s is expressed by equation (8).
λ _s =c _s・P _n ...(8)
_cs is a coefficient based on the method of generating the correlation matrix _Rxx .
That is,
λ _th1 = _cm・P _n +c _s・P _n =(c _m +c _s )P _n
The first incoming wave number estimation threshold λ _th1 is a value obtained by multiplying the noise level P _n by a preset value (c _m +c _s ).

In Equation (7), by adding λ _s related to the correlation matrix R _xx , λ _th1 is set in consideration of the time variation of the noise level and the inter-channel variation. By using the first incoming wave number estimation threshold λ _th1 in equation (7), the eigenvalue determination unit 108 can discriminate the eigenvalue into a signal eigenvalue and a noise eigenvalue based on the noise level.

Next, in step S8, the arriving wave number estimation threshold calculating section 107 calculates a second arriving wave number estimation threshold λ _th2 . Specifically, equation (6) is used to calculate as shown in equation (9).
λ _th2 = r ₂ ′・λ ₁ ...(9)
Here, r ₂ ' is the maximum value of r ₂ in a certain arrival angle range, as shown in FIG. Since the second arrival wave number estimation threshold λ _th2 in equation (9) corresponds to the arrival wave number estimation threshold in a region where the signal strength is high, by using the second arrival wave number estimation threshold λ _th2 in this region, the antenna The eigenvalues can be distinguished into signal eigenvalues and noise eigenvalues by considering the amplitude error and phase error.

Next, in step S9, the eigenvalue determination unit 108 determines the eigenvalue into a signal eigenvalue and a noise eigenvalue using the first arrival wave number estimation threshold λ _th1 and the second arrival wave number estimation threshold λ _th2 . That is, the eigenvalues larger than the respective thresholds are determined to be signal eigenvalues, and the number of arriving waves, that is, the number of arriving signals is estimated from the number of determined signal eigenvalues.

Next, in step S10, the number of arriving waves, that is, the number of arriving signals, is estimated from the number of signal eigenvalues determined by the eigenvalue determining section 108. The arrival angle calculation unit 109 uses the eigenvalues corresponding to the estimated arrival signals to implement a high-resolution direction of arrival estimation method such as MUSIC (Multiple Signal Classification) or ESPRIT (Estimation Signal Parameters via Rotational Invariance Technique). It becomes possible to calculate the angle of arrival with high precision. Therefore, it is possible to obtain a highly accurate estimation result of the number of arriving waves when the received signal strength is extremely small or large, and it is possible to calculate the angle of arrival with high accuracy using the estimation result of the arriving wave.

In addition, when calculating the second arrival wave number estimation threshold λ _th2 in the arriving wave number estimation threshold calculation unit 107, if it is possible to estimate (estimate) the arrival angle θ in advance, r ₂ ' is calculated according to the angle. By changing this, higher accuracy in estimating the number of arriving waves can be achieved.

As a method for estimating the angle of arrival in advance, a result obtained using a direction of arrival estimation method that does not require information on the number of arriving waves, such as a beam former method, is used as the provisional angle of arrival. Alternatively, the previous value of the arrival angle of the same object may be used as the provisional arrival angle.

As an example of changing r ₂ ', consider the case where one wave is incident when the noise level is extremely small. When the arrival angle θ=θ ₁ , the second arrival wave number estimation threshold λ _th2 (θ ₁ ) is obtained from equation (9) to equation (10).
λ _th2 (θ ₁ )=r ₂ (θ ₁ )・λ ₁ ≦ r ₂ ′・λ ₁ (10)
Therefore, the threshold value can be set smaller when r ₂ (θ ₁ ) is used than r ₂ ′, so the accuracy of estimating the number of arriving waves when two or more waves are incident near the arrival angle θ ₁ becomes higher.

Further, as shown in FIG. 8, a step S8a for calculating a third arriving wave number estimation threshold λ _th3 may be provided after step S8. That is, in step S8a, the arriving wave number estimation threshold calculation unit 107 calculates the third arriving wave number estimation threshold λ _th3 as shown in equation (11).

λ _th3 =p・λ _th1 +q・λ _th2
= p・c・P _n +q・r ₂ ′・λ ₁ ...(11)
Here, p and q are arbitrary constants, and c=c _m +c _s . The third arriving wave number estimation threshold λ _th3 in Equation (11) is a threshold obtained by weighting and adding the first arriving wave number estimation threshold λ _th1 and the second arriving wave number estimation threshold λ _th2 , respectively. By using this third arriving wave number estimation threshold λ _th3 , it is possible to distinguish the eigenvalue into a signal eigenvalue and a noise eigenvalue, regardless of the received signal strength.

As an example, consider the radio wave environment when one wave is incident. FIG. 9 shows the relationship between the eigenvalue and each arriving wave number estimation threshold with respect to the received signal strength when p, q=1. At this time, equation (11) is expressed as equation (12).
λ _th3 =c・P _n +r ₂ ′・λ ₁ >λ ₂ ...(12)

FIG. 9 is an example in which the noise level is constant, and the first arriving wave number estimation threshold λ _th1 is a constant value independent of the received signal strength. The second arriving wave number estimation threshold λ _th2 takes a larger value as the received signal strength increases. In the region where the received signal strength is small, λ _th1 > λ ₂ and λ _th3 > λ ₂ , and it is possible to estimate the number of arriving waves as 1 when one wave is incident, but if λ _th2 < λ ₂ , the number of arriving waves is 1. At the estimated threshold value λ _th2 , the number of arriving waves is not estimated to be 1. In a region where the received signal strength is high, λ _th2 > λ ₂ and λ _th3 > λ ₂ , and it is possible to estimate the number of arriving waves as 1 when one wave is incident, but when λ _th1 < λ ₂ , the number of arriving waves is 1. At the wave number estimation threshold λ _th1 , the number of arriving waves is not estimated to be 1. Furthermore, depending on the signal strength, there may be a region in which the number of arriving waves is estimated to be 2 instead of 1 for both the first arriving wave number estimation threshold λ _th1 and the second arriving wave number estimation threshold λ _th2 .

Therefore, by using the third arrival wave number estimation threshold λ _th3 , eigenvalues whose magnitude is λ ₂ or less regardless of the received signal strength can be determined as noise eigenvalues, and as shown in FIG. 9, when one wave is incident, The number of arriving waves can be estimated to be 1 (step S9).

However, in reality, there is a calculation accuracy of the arithmetic processing unit 11 and a calculation error of _Pn , so it is necessary to adjust p and q to obtain the desired accuracy.

Also in the flowchart of FIG. 8, similarly to the flowchart of FIG. By implementing a high resolution direction of arrival estimation method such as ESPRIT, the angle of arrival can be calculated with high accuracy.

Further, in step S5 of FIGS. 4 and 8, a plurality of groups of subarrays having different shapes, that is, different combinations of receiving antennas may be used. With the above-mentioned subarrays_1 to 3 as the first group, the threshold for estimating the number of arriving waves in equation (11) is calculated in the same way for the subarrays in other groups, and after comparing the estimated numbers of arriving waves, The results are integrated to determine the final number of arriving waves. By using a plurality of groups of subarrays having different shapes in this way, it is possible to prevent the accuracy of estimating the number of arriving waves from decreasing depending on the shape of the subarray.

In the flowcharts of FIGS. 4 and 8, the processes from step S5 to step S10 described above are repeated for the number of target objects, the distances, relative velocities, and angles of arrival of all target objects are calculated as target object information, and the external device ( (not shown), the calculated target object information is output.

As described above, according to the radar apparatus 100 according to the first embodiment, good incoming wave number estimation accuracy is maintained when the received signal strength is small, and good incoming wave number estimation accuracy is maintained even when the received signal strength becomes large. The wave number can be estimated. That is, the radar device 100 according to the first embodiment receives reflected waves reflected from an object using a plurality of receiving antennas 6, performs frequency analysis on the received signals, and determines the noise level based on the obtained complex spectrum. is calculated and the peak value is determined. Then, correlation matrices corresponding to the complex spectra that are determined to be peak values, which are complex spectra obtained by frequency analysis from signals received from the same object at different times, are calculated and averaged. The eigenvalues and eigenvalue vectors of the correlation matrix are analyzed. Alternatively, for a plurality of subarrays formed by different sets of receiving antennas 6 among the plurality of receiving antennas 6, a correlation matrix is calculated for the complex spectrum belonging to the subarray for each subarray, and the correlation matrix for each subarray is calculated. Eigenvalues and eigenvalue vectors are analyzed for the correlation matrix of the subarray group obtained by averaging. Here, the receiving antennas forming the subarray may be virtual receiving antennas. In the radar device 100 according to the first embodiment, the arriving wave number estimation threshold calculation unit 107 sets the first arriving wave number estimation threshold λ _th1 for a region where the signal strength is low, taking into consideration the noise level. Furthermore, the arriving wave number estimation threshold calculation unit 107 sets a second arriving wave number estimation threshold λ _th2 based on the eigenvalue, taking into account the amplitude error and phase error of the receiving antenna, for a region where the signal strength is high. Depending on the signal strength, the signal eigenvalue is determined from the eigenvalues using either the first arrival wave number estimation threshold λ _th1 or the second arrival wave number estimation threshold λ _th2 , and using the signal eigenvalue, the arrival angle calculation unit 109 calculates the following: By estimating the number of arriving signals and performing a high-resolution direction-of-arrival estimation method using signal eigenvalues corresponding to the arriving signals, it becomes possible to calculate the angle of arrival with high precision.

In addition, the arriving wave number estimation threshold calculation unit 107 of the radar device 100 according to the first embodiment also calculates a third value that is weighted and added to the first arriving wave number estimation threshold λ _th1 and the second arriving wave number estimation threshold λ _th2 . A threshold value λ _th3 for estimating the number of arriving waves is calculated. Since the signal eigenvalue is determined from the eigenvalue using this third arrival wave number estimation threshold λ _th3 , it becomes possible to determine the signal eigenvalue from the eigenvalue with high accuracy regardless of the signal strength. Using the signal eigenvalues, the angle of arrival calculating section 109 performs a high-resolution direction of arrival estimation method, thereby making it possible to calculate the angle of arrival with even higher accuracy.

If only the threshold value is set based on the noise level, there is a risk that the number of arriving signals will be erroneously estimated in areas where the signal strength is large, resulting in a decrease in the accuracy of calculating the arrival angle. Since the threshold value λ _th2 for estimating the number of arriving waves is set, the number of arriving signals can be estimated with high accuracy even if the signal strength increases.

Although this application describes exemplary embodiments, the various features, aspects, and functions described in the embodiments are not limited to the application of particular embodiments, and may be used alone or It is applicable to the embodiments in various combinations.
Accordingly, countless variations not illustrated are envisioned within the scope of the technology disclosed herein. For example, this includes cases in which at least one component is modified, added, or omitted.

Hereinafter, various aspects of the present disclosure will be collectively described as supplementary notes.

(Additional note 1)
a transmitting antenna for transmitting a transmitting signal;
a plurality of receiving antennas that receive reflected signals reflected by one or more objects;
A radar device comprising: a control unit that frequency-analyzes the reflected signals received by a plurality of the receiving antennas and calculates an angle of arrival, which is an angle with respect to the own device, for each of the objects,
The control unit includes:
a noise level calculation unit that calculates a noise level from the amplitude value of the complex spectrum obtained by the frequency analysis;
a peak value determination unit that determines a peak value based on the amplitude value of the complex spectrum obtained by the frequency analysis and the noise level calculated by the noise level calculation unit;
a correlation matrix calculation unit that calculates a correlation matrix for the complex spectrum determined to be the peak value;
an eigenvalue calculation unit that calculates an eigenvalue of the correlation matrix;
an arriving wave number estimation threshold calculation unit that calculates a threshold for discriminating the eigenvalue into a noise eigenvalue caused by noise and a signal eigenvalue corresponding to the object;
The number of arriving signals is estimated from the number of determined signal eigenvalues based on the threshold calculated by the arriving wave number estimation threshold calculation unit, and the arrival angle is determined based on the incoming signal and the estimated signal eigenvalue. an arrival angle calculation unit that calculates,
The arriving wave number estimation threshold calculation unit includes:
calculating a first arrival wave number estimation threshold based on the noise level calculated by the noise level calculation unit, and calculating a second arrival wave number estimation threshold based on the eigenvalue calculated by the eigenvalue calculation unit;
The arrival angle calculation unit is a radar device that estimates the number of arriving signals from the number of signal eigenvalues determined using the first arriving wave number estimation threshold and the second arrival wave number estimation threshold.
(Additional note 2)
The radar device according to supplementary note 1, wherein the arriving wave number estimation threshold calculation unit calculates the first arriving wave number estimation threshold by multiplying the noise level by a first value set in advance.
(Additional note 3)
Supplementary note 1, wherein the arriving wave number estimation threshold calculation unit calculates the second arrival wave number estimation threshold by multiplying the largest eigenvalue among the eigenvalues calculated by the eigenvalue calculation unit by a second value set in advance. or the radar device described in 2.
(Additional note 4)
The radar device according to supplementary note 3, wherein the angle of arrival of the reflected signal is estimated in advance, and the second value uses a value corresponding to the previously estimated angle of arrival of the reflected signal.
(Appendix 5)
The first arrival wave number estimation threshold is used to determine the signal eigenvalue in a region where the signal strength of the reflected signal is small, and the second arrival wave number estimation threshold is used to determine the reflected signal in a region where the signal intensity of the reflected signal is small. The radar device according to any one of appendices 1 to 4, which is used to determine the signal eigenvalue in a region where the signal strength is large.
(Appendix 6)
The arriving wave number estimation threshold calculation unit calculates a third arriving wave number estimation threshold obtained by weighting and adding the first arriving wave number estimation threshold and the second arriving wave number estimation threshold, respectively,
The arrival angle calculation unit estimates the number of arriving signals from the number of signal eigenvalues determined using the third arrival wave number estimation threshold, and calculates the arrival angle. The radar device described in .
(Appendix 7)
The correlation matrix calculation unit calculates a correlation matrix for a complex spectrum that belongs to the subarray and is determined to have the peak value, for each of a plurality of subarrays formed by different sets of reception antennas among the plurality of reception antennas, averaging the obtained correlation matrices corresponding to the plurality of subarrays;
The radar device according to any one of Supplementary Notes 1 to 6, wherein the eigenvalue calculation unit calculates the eigenvalue of the averaged correlation matrix.
(Appendix 8)
A plurality of sub-arrays formed by different sets of receiving antennas among the plurality of receiving antennas are referred to as a first group array, and the first group array is defined as a plurality of receiving antennas in which the plurality of receiving antennas are combined in different ways. A plurality of subarrays formed in a group are set as a second group array,
The arrival angle calculation unit is
The radar device according to supplementary note 7, wherein the number of the arriving signals is estimated for each group array, and the final angle of arrival is determined based on the estimated number of the plurality of arriving signals.
(Appendix 9)
The correlation matrix calculation unit calculates a correlation matrix for each of the plurality of complex spectra determined to have the peak value, which is obtained by frequency-analyzing the plurality of reflected signals reflected from the same object and received at different times. calculate and average the obtained correlation matrix,
The radar device according to any one of Supplementary Notes 1 to 6, wherein the eigenvalue calculation unit calculates the eigenvalue of the averaged correlation matrix.

1: Signal processing unit, 2: Control voltage generator, 3: VCO, 4: Distributor, 5: Transmitting antenna, 6, 6_1 to 6_6: Receiving antenna, 7, 7_1 to 7_6: Mixer, 8, 8_1 to 8_6: A/D converter, 9, 9_1 to 9_6: FFT calculation unit, 10: Control unit, 11: Processing unit, 12: Storage device, 13: Communication circuit, 14: D/A converter, 100: Radar device, 101: Amplitude calculation unit, 102: Noise level calculation unit, 103: Peak value determination unit, 104: Distance and relative velocity calculation unit, 105: Correlation matrix calculation unit, 106: Eigenvalue calculation unit, 107: Arrival wave number estimation threshold calculation unit, 108 : Eigenvalue determination unit, 109: Arrival angle calculation unit.

The radar device disclosed in this application includes:
a transmitting antenna for transmitting a transmitting signal;
a plurality of receiving antennas that receive reflected signals reflected by one or more objects;
A radar device comprising: a control unit that frequency-analyzes the reflected signals received by a plurality of the receiving antennas and calculates an angle of arrival, which is an angle with respect to the own device, for each of the objects,
The control unit includes:
a noise level calculation unit that calculates a noise level from the amplitude value of the complex spectrum obtained by the frequency analysis;
a peak value determination unit that determines a peak value based on the amplitude value of the complex spectrum obtained by the frequency analysis and the noise level calculated by the noise level calculation unit;
a correlation matrix calculation unit that calculates a correlation matrix for the complex spectrum determined to be the peak value;
an eigenvalue calculation unit that calculates an eigenvalue of the correlation matrix;
an arriving wave number estimation threshold calculation unit that calculates a threshold for discriminating the eigenvalue into a noise eigenvalue caused by noise and a signal eigenvalue corresponding to the object;
The number of arriving signals is estimated from the number of determined signal eigenvalues based on the threshold calculated by the arriving wave number estimation threshold calculation unit, and the arrival angle is determined based on the incoming signal and the estimated signal eigenvalue. an arrival angle calculation unit that calculates,
The arriving wave number estimation threshold calculation unit includes:
calculating a first arrival wave number estimation threshold based on the noise level calculated by the noise level calculation unit, and calculating a second arrival wave number estimation threshold based on the eigenvalue calculated by the eigenvalue calculation unit;
The arrival angle calculation unit determines the signal eigenvalue based on the first arrival wave number estimation threshold in a region where the signal strength of the reflected signal is small, and the arrival angle calculation unit determines the signal eigenvalue based on the first arrival wave number estimation threshold, and In a region where the signal strength is high, the signal eigenvalues are determined based on the second arriving wave number estimation threshold, and the number of the arriving signals is estimated from the number of determined signal eigenvalues.

FIG. 3 is a diagram showing an example of the hardware configuration of the control unit 10 of the radar device 100. As shown in FIG. 3 , the control unit 10 serves as a processing circuit, and includes an arithmetic processing device 11 (computer) such as a DSP (Digital Signal Processor), a storage device 12 that exchanges data with the arithmetic processing device 11, and an arithmetic processing device 11. It includes an A/D converter 8 that inputs a beat signal, a D/A converter 14 that outputs a command value of a frequency modulation signal from an arithmetic processing unit 11 to a control voltage generator 2, a communication circuit 13, and the like.

Claims (9)

a transmitting antenna for transmitting a transmitting signal;
a plurality of receiving antennas that receive reflected signals reflected by one or more objects;
A radar device comprising: a control unit that frequency-analyzes the reflected signals received by a plurality of the receiving antennas and calculates an angle of arrival, which is an angle with respect to the own device, for each of the objects,
The control unit includes:
a noise level calculation unit that calculates a noise level from the amplitude value of the complex spectrum obtained by the frequency analysis;
a peak value determination unit that determines a peak value based on the amplitude value of the complex spectrum obtained by the frequency analysis and the noise level calculated by the noise level calculation unit;
a correlation matrix calculation unit that calculates a correlation matrix for the complex spectrum determined to be the peak value;
an eigenvalue calculation unit that calculates an eigenvalue of the correlation matrix;
an arriving wave number estimation threshold calculation unit that calculates a threshold for discriminating the eigenvalue into a noise eigenvalue caused by noise and a signal eigenvalue corresponding to the object;
The number of arriving signals is estimated from the number of determined signal eigenvalues based on the threshold calculated by the arriving wave number estimation threshold calculation unit, and the arrival angle is determined based on the incoming signal and the estimated signal eigenvalue. an arrival angle calculation unit that calculates,
The arriving wave number estimation threshold calculation unit includes:
calculating a first arrival wave number estimation threshold based on the noise level calculated by the noise level calculation unit, and calculating a second arrival wave number estimation threshold based on the eigenvalue calculated by the eigenvalue calculation unit;
The arrival angle calculation unit is a radar device that estimates the number of arriving signals from the number of signal eigenvalues determined using the first arriving wave number estimation threshold and the second arrival wave number estimation threshold. The radar device according to claim 1, wherein the arriving wave number estimation threshold calculation unit calculates the first arriving wave number estimation threshold by multiplying the noise level by a first value set in advance. 2. The arriving wave number estimation threshold calculation unit calculates the second arrival wave number estimation threshold by multiplying the largest eigenvalue among the eigenvalues calculated by the eigenvalue calculation unit by a preset second value. 1. The radar device according to 1. The radar device according to claim 3, wherein the angle of arrival of the reflected signal is estimated in advance, and the second value uses a value corresponding to the previously estimated angle of arrival of the reflected signal. The first arrival wave number estimation threshold is used to determine the signal eigenvalue in a region where the signal strength of the reflected signal is small, and the second arrival wave number estimation threshold is used to determine the reflected signal in a region where the signal intensity of the reflected signal is small. The radar device according to any one of claims 1 to 4, wherein the radar device is used to determine the signal eigenvalue in a region where the signal intensity is large. The arriving wave number estimation threshold calculation unit calculates a third arriving wave number estimation threshold obtained by weighting and adding the first arriving wave number estimation threshold and the second arriving wave number estimation threshold, respectively,
4. The angle of arrival calculation unit estimates the number of arriving signals from the number of signal eigenvalues determined using the third arrival wave number estimation threshold, and calculates the angle of arrival. The radar device described in . The correlation matrix calculation unit calculates a correlation matrix for a complex spectrum that belongs to the subarray and is determined to have the peak value, for each of a plurality of subarrays formed by different sets of reception antennas among the plurality of reception antennas, averaging the obtained correlation matrices corresponding to the plurality of subarrays;
The radar device according to any one of claims 1 to 3, wherein the eigenvalue calculation unit calculates an eigenvalue of the averaged correlation matrix. A plurality of sub-arrays formed by different sets of receiving antennas among the plurality of receiving antennas are referred to as a first group array, and the first group array is defined as a plurality of receiving antennas in which the plurality of receiving antennas are combined in different ways. A plurality of subarrays formed in a group are set as a second group array,
The arrival angle calculation unit is
8. The radar device according to claim 7, wherein the number of the arriving signals is estimated for each of the array groups, and the final angle of arrival is determined based on the estimated number of the plurality of arriving signals. The correlation matrix calculation unit calculates a correlation matrix for each of the plurality of complex spectra determined to have the peak value, which is obtained by frequency-analyzing the plurality of reflected signals reflected from the same object and received at different times. calculate and average the obtained correlation matrix,
The radar device according to any one of claims 1 to 3, wherein the eigenvalue calculation unit calculates an eigenvalue of the averaged correlation matrix.

Images