KR20210152911A - Method and apparatus of processing radar signal by correcting phase distortion - Google Patents

Abstract

A method and a device for processing a radar signal by correcting phase distortion are disclosed. According to one embodiment, the method includes the steps of: generating radar data based on a radar transmission signal and a radar reception signal; correcting the radar data using a correction vector; and estimating an angle of arrival corresponding to the corrected radar data using a direction matrix.

Description

METHOD AND APPARATUS OF PROCESSING RADAR SIGNAL BY CORRECTING PHASE DISTORTION

The following embodiments relate to a method and apparatus for processing a radar signal by correcting phase distortion.

The Advanced Driver Assistance System (ADAS) is a system that supports driving for the purpose of improving driver safety and convenience and avoiding dangerous situations by using sensors mounted inside or outside the vehicle.

Sensors used in ADAS may include a camera, an infrared sensor, an ultrasonic sensor, a lidar (LiDAR), and a radar (Radar). Among these, radar can reliably measure objects around the vehicle without being affected by the surrounding environment, such as weather, compared to optical-based sensors.

According to an embodiment, the radar signal processing method includes a radar transmission signal transmitted through an array antenna of a radar sensor based on a frequency modulation model and a radar reception signal received through the array antenna after the radar transmission signal is reflected by a target based on the , generating radar data; calibrating the radar data using a calibration vector for correcting a feedline error according to a difference in feedline delay between channels of the array antenna; and estimating an angle of arrival corresponding to the calibrated radar data using a direction matrix in which a phase shift of the calibrated radar data according to the frequency modulation characteristic of the frequency modulation model is reflected.

When the radar reception signal is received from a target located in front of the radar sensor, the calibration vector may correct the radar data so that phase components of the channels of the radar data have the same value.

The direction matrix may include sub-direction matrices corresponding to different sample indices, respectively. The estimating of the angle of arrival may include: obtaining a first sub-direction matrix corresponding to a first sample index from among the sub-direction matrices; and estimating a first angle of arrival of first sub-radar data corresponding to the first sample index in the calibrated radar data by using the first sub-direction matrix.

The generating of the radar data may include generating the radar data by sampling an intermediate frequency signal generated based on the radar transmission signal and the radar reception signal. The estimating of the angle of arrival may include correcting an element beam pattern (EBP) error that occurs due to an effect of a beam pattern of an antenna element of the array antenna on a beam pattern of the array antenna. .

The estimating of the angle of arrival may include: estimating an initial angle of arrival corresponding to the corrected radar data using the direction matrix; and estimating the final angle of arrival corresponding to the corrected radar data by removing the EBP error caused by the effect of the beam pattern of the antenna element of the array antenna on the beam pattern of the array antenna included in the initial angle of arrival. may include

The estimating of the final angle of arrival may include: determining an EBP error value with respect to the initial angle of arrival corresponding to the corrected radar data based on an EBP error model indicating EBP error values for each angle; and estimating the final angle of arrival corresponding to the corrected radar data by correcting the initial angle of arrival with the determined EBP error value. The EBP error model may be generated by estimating EBP error values for other angles based on EBP error values for basic angles measured through a test.

The radar transmission signal may include a chirp signal in which a carrier frequency is modulated based on the frequency modulation model. The radar reception signal may be received through reception antenna elements of the array antenna, and the channels may be formed based on the reception antenna elements. The radar signal processing method may further include estimating at least one of a distance and a speed with respect to the target based on the radar data, and based on at least one of the angle of arrival, the distance, and the speed, the radar A vehicle equipped with a signal processing device may be controlled.

According to an embodiment, the radar signal processing apparatus transmits a radar transmission signal through an array antenna based on a frequency modulation model, and when the radar transmission signal is reflected by a target, a radar receiving a radar signal through the array antenna sensor; and generating radar data based on the radar transmission signal and the radar reception signal, and correcting the radar data using a correction vector that corrects a feedline error according to a difference in feedline delay between channels of the array antenna, , a processor for estimating an angle of arrival corresponding to the calibrated radar data using a direction matrix in which a phase shift of the calibrated radar data according to a frequency modulation characteristic of the frequency modulation model is reflected.

According to an embodiment, a vehicle transmits a radar transmission signal through an array antenna based on a frequency modulation model, and when the radar transmission signal is reflected by a target, a radar sensor configured to receive a radar reception signal through the array antenna; generating radar data based on the radar transmission signal and the radar reception signal, and correcting the radar data using a correction vector that corrects a feedline error according to a difference in feedline delay between channels of the array antenna; a processor for estimating an angle of arrival corresponding to the calibrated radar data using a direction matrix in which a phase shift of the calibrated radar data according to the frequency modulation characteristic of the frequency modulation model is reflected; and a control system for controlling the vehicle based on the angle of arrival.

1 illustrates a process of recognizing a surrounding environment through a radar signal processing method according to an exemplary embodiment.
2 illustrates a configuration of a radar signal processing apparatus according to an exemplary embodiment.
3 shows a configuration of a radar sensor according to an embodiment.
4 illustrates a process of processing a radar signal through phase distortion correction according to an exemplary embodiment.
5 illustrates a feedline delay of an array antenna according to an exemplary embodiment.
6 illustrates correction of a feedline error using a correction vector according to an embodiment.
7 is a diagram illustrating receiving antenna elements of an array antenna according to an exemplary embodiment.
8 illustrates a phase change according to a carrier frequency change for each sampling point in a radar signal processing method according to an embodiment.
9 illustrates a sampling process of an intermediate frequency signal according to an exemplary embodiment.
10 illustrates correction of a frequency modulation parasitic error through phase normalization according to an embodiment.
11 and 12 illustrate correction of device beam pattern errors according to an exemplary embodiment.
13 illustrates a radar signal processing method according to an embodiment.
14 illustrates an electronic device according to an embodiment.

Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. However, since various changes may be made to the embodiments, the scope of the patent application is not limited or limited by these embodiments. It should be understood that all modifications, equivalents and substitutes for the embodiments are included in the scope of the rights.

The terms used in the examples are used for the purpose of description only, and should not be construed as limiting. The singular expression includes the plural expression unless the context clearly dictates otherwise. In this specification, terms such as "comprise" or "have" are intended to designate that a feature, number, step, operation, component, part, or a combination thereof described in the specification exists, but one or more other features It should be understood that this does not preclude the existence or addition of numbers, steps, operations, components, parts, or combinations thereof.

Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which the embodiment belongs. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the related art, and should not be interpreted in an ideal or excessively formal meaning unless explicitly defined in the present application. does not

In addition, in the description with reference to the accompanying drawings, the same components are given the same reference numerals regardless of the reference numerals, and the overlapping description thereof will be omitted. In describing the embodiment, if it is determined that a detailed description of a related known technology may unnecessarily obscure the gist of the embodiment, the detailed description thereof will be omitted.

In addition, in describing the components of the embodiment, terms such as first, second, A, B, (a), (b), etc. may be used. These terms are only for distinguishing the elements from other elements, and the essence, order, or order of the elements are not limited by the terms. When it is described that a component is "connected", "coupled" or "connected" to another component, the component may be directly connected or connected to the other component, but another component is between each component. It will be understood that may also be "connected", "coupled" or "connected".

Components included in one embodiment and components having a common function will be described using the same names in other embodiments. Unless otherwise stated, descriptions described in one embodiment may be applied to other embodiments as well, and detailed descriptions within the overlapping range will be omitted.

1 illustrates a process of recognizing a surrounding environment through a radar signal processing method according to an exemplary embodiment. Referring to FIG. 1 , the radar signal processing apparatus 110 analyzes the radar signal received from the radar sensor 111 to obtain information about the target 180 in front (eg, range, velocity). , direction, etc.) can be detected. The radar sensor 111 may be located inside or outside the radar signal processing device 110 , and the radar signal processing device 110 includes not only the radar signal received from the radar sensor 111 , but also other sensors (eg, an image sensor). etc.), it is also possible to detect information about the target 180 in front by considering the collected data together. The resolution of radar data processing can be divided into resolving power performance on the hardware side and the resolution performance on the software side. Below, the resolution performance improvement on the software side will be mainly described.

For reference, in the present specification, the resolving power is the ability of the device to discriminate very small differences, for example, the minimum unit discrimination power, "resolution = (minimum discernable division unit) / (total operating range)" can indicate The smaller the resolving power value of the device, the more precise the result may be output by the device. A resolution number may also be referred to as a resolving power unit. For example, when the resolution value of the device is small, the device can discriminate smaller units, so that it is possible to output a result with improved precision having an increased resolution. Conversely, if the resolution value of the device is large, the device may not be able to distinguish small units, so that it is possible to output a result of reduced precision with a reduced resolution.

According to an embodiment, the radar signal processing apparatus 110 may be mounted on a vehicle as shown in FIG. 1 . The vehicle is based on the distance to the target 180 detected by the radar signal processing device 110, adaptive cruise control (ACC), automatic emergency braking (AEB), blind spot detection (Blind Spot Detection, BSD), Lane Change Assistance (LCA), and the like may be performed.

Furthermore, the radar signal processing apparatus 110 may generate the surrounding map 130 in addition to detecting the distance. The surrounding map 130 is a map indicating the positions of various targets that exist in the vicinity of the radar signal processing device 110, such as the target 180, and the surrounding targets may be dynamic objects such as vehicles and people, and guard rails. And it may be a static object that exists in the background, such as a traffic light.

A single scan image method may be used as a method for generating the surrounding map 130 . A single scan image method is that the radar signal processing apparatus 110 acquires a single scan image 120 from a sensor, and generates a peripheral map 130 from the acquired single scan image 120 . The single scan image 120 is an image generated from a radar signal sensed by the single radar sensor 111, and may represent distances indicated by radar signals received from an arbitrary elevation angle with relatively high resolution. . For example, in the single scan image 120 illustrated in FIG. 1 , a horizontal axis may indicate a steering angle of the radar sensor 111 , and a vertical axis may indicate a distance from the radar sensor 111 to the target 180 . However, the shape of the single scan image is not limited to the bar shown in FIG. 1 , and may be expressed in another format according to design.

The steering angle may indicate an angle corresponding to a target direction from the radar signal processing apparatus 110 toward the target 180 . For example, the steering angle may be an angle between a traveling direction of the radar signal processing apparatus 110 (or a vehicle including the radar processing apparatus 110 ) and a target direction. For reference, in the present specification, the steering angle has been mainly described based on a horizontal angle, but is not limited thereto. For example, the steering angle can also be applied to the elevation angle.

According to an embodiment, the radar signal processing apparatus 110 may acquire information on the shape of the target 180 through a multi-radar map. A multiple radar map may be generated from a combination of multiple radar scan images. For example, the radar signal processing apparatus 110 may generate the surrounding map 130 by spatio-temporally combining radar scan images acquired according to the movement of the radar sensor 111 . The surrounding map 130 may be a kind of radar image map, and may be used for pilot parking and the like.

According to an embodiment, the radar signal processing apparatus 110 may utilize direction of arrival (DoA) information to generate the surrounding map 130 . The angle of arrival information refers to information indicating a direction in which a radar signal reflected from a target is received. The radar signal processing apparatus 110 may identify a direction in which the target is present based on the radar sensor 111 using the above-described angle of arrival information. Accordingly, this angle of arrival information can be used to generate radar scan data and a surrounding map.

According to an embodiment, radar information such as distance, speed, angle of arrival, and map information about the target 180 generated by the radar signal processing device 110 controls a vehicle equipped with the radar signal processing device 110 . can be used to For example, the control of the vehicle may include speed and steering control of the vehicle such as ACC, AEB, BSD, and LCA. A control system of the vehicle may control the vehicle by using radar information directly or indirectly.

2 illustrates a configuration of a radar signal processing apparatus according to an exemplary embodiment. Referring to FIG. 2 , the radar signal processing apparatus 200 may include a radar sensor 210 and a processor 220 . The radar sensor 210 may radiate a radar signal to the outside of the radar sensor 210 , and may receive a signal reflected by the target by the radiated radar signal. In the present specification, a radar signal radiated may be referred to as a radar transmission signal, and a received signal may be referred to as a radar reception signal. The radar transmission signal may include a chirp signal in which a carrier frequency is modulated based on a frequency modulation model. The frequency of the radar transmission signal may vary within a predetermined band. For example, the frequency of a radar transmission signal may vary linearly within a predetermined band.

The radar sensor 210 may include an array antenna, and may transmit a radar transmission signal and receive a radar reception signal through the array antenna. The array antenna may include a plurality of antenna elements. According to an embodiment, multiple input multiple output (MIMO) may be implemented through a plurality of antenna elements. In this case, a plurality of MIMO channels may be formed by a plurality of antenna elements. For example, a plurality of channels corresponding to the M x N virtual antennas may be formed through the M transmit antenna elements and the N receive antenna elements. Here, radar reception signals received through each channel may have different phases according to reception directions.

Based on the radar transmission signal and the radar reception signal, radar data may be generated. For example, the radar sensor 210 transmits a radar transmission signal through an array antenna based on a frequency modulation model, and when the radar transmission signal is reflected by a target, receives a radar reception signal through the array antenna, and a radar transmission signal and an intermediate frequency (IF) signal based on the radar reception signal. The intermediate frequency signal may have a frequency corresponding to the difference between the frequency of the radar transmission signal and the frequency of the radar reception signal. The processor 220 may generate radar data through a sampling operation on the intermediate frequency signal. The radar data may correspond to raw data of an intermediate frequency.

The processor 220 may generate information about the target based on the radar data and use it. For example, the processor 220 performs range FFT (Fast Fourier Transform (FFT)), Doppler FFT (Doppler FFT), CFAR (Constant False Alarm Rate Detection), angle of arrival estimation, etc., based on radar data, It is possible to obtain information about the target, such as distance, speed, and direction. Information on such a target may be provided for various applications such as AAC, AEB, BSD, and LCA.

3 shows a configuration of a radar sensor according to an embodiment. Referring to FIG. 3 , the radar sensor 310 includes a chirp transmitter 311 , a duplexer 312 , an antenna 313 , a frequency mixer 314 , an amplifier 315 , and a spectrum analyzer 316 . can do. The radar sensor 310 may radiate a signal through the antenna 313 and receive a signal through the antenna 313 . Although one antenna 313 is illustrated in FIG. 3 , the antenna 313 may include at least one transmit antenna element and at least one receive antenna. For example, the antenna 313 may correspond to an array antenna. As an example, the antenna 313 may include three or more receive antenna elements. In this case, the receiving antenna elements may be spaced apart at the same interval.

The radar sensor 310 may be, for example, a mmWave radar, and it is possible to measure the distance to the target by analyzing the time of flight (ToF), which is the time when the emitted electric wave is reflected by the target and returns, and the change in the waveform of the radar signal. can For reference, compared to optical-based sensors including cameras, mmWave radar can detect the front regardless of changes in the external environment such as fog and rain. In addition, the mmWave radar is one of the sensors that can compensate for the disadvantages of the above-described camera because it has superior cost-performance compared to LiDAR. For example, the radar sensor 310 may be implemented as a frequency modulated continuous wave (FMCW) radar. The FMCW radar may have robust characteristics against external noise.

The chirped transmitter 311 may generate a frequency-modulated signal (FM signal, 302) whose frequency varies with time. For example, the chirped transmitter 311 may generate the frequency modulated signal 302 by performing frequency modulation according to the frequency modulation characteristic of the frequency modulation model 301 . The frequency modulated signal 302 may also be referred to as a chirped signal. In the present specification, the frequency modulation model 301 may represent a model indicating a change in a carrier frequency for a given transmission time in an arbitrary radar transmission signal. A vertical axis of the frequency modulation model 301 may indicate a carrier frequency, and a horizontal axis may indicate time. For example, the frequency modulation model 301 may have a frequency modulation characteristic of linearly changing the carrier frequency (eg, linearly increasing or linearly decreasing). As another example, the frequency modulation model 301 may have a frequency modulation characteristic that non-linearly changes the carrier frequency.

The frequency modulation model 301 of FIG. 3 is illustrated as having a frequency modulation characteristic of linearly increasing a frequency with time. The chirp transmitter 311 may generate a frequency modulated signal 302 having a carrier frequency according to the frequency modulation model 301 . For example, as shown in FIG. 3 , the frequency modulated signal 302 may represent a waveform in which the carrier frequency gradually increases in some sections, and may represent a waveform in which the carrier frequency gradually decreases in the remaining sections. . The chirp transmitter 311 may transmit the frequency modulated signal 302 to the duplexer 312 .

The duplexer 312 may determine a transmission path and a reception path of a signal through the antenna 313 . For example, while the radar sensor 310 emits the frequency modulated signal 302 , the duplexer 312 may form a signal path from the chirped transmitter 311 to the antenna 313 , The frequency-modulated signal 302 may be transmitted to the antenna 313 and then radiated to the outside. While the radar sensor 310 receives the reflected signal from the target, the duplexer 312 may form a signal path from the antenna 313 to the spectrum analyzer 316 . The antenna 313 may receive a received signal reflected back after the radiated signal reaches the obstacle, and the radar sensor 310 may spectrum the received signal through a signal path from the antenna 313 to the spectrum analyzer 316 . may be passed to the analyzer 316 . A signal radiated through the antenna 313 may be referred to as a radar transmission signal, and a signal received through the antenna 313 may be referred to as a radar reception signal.

The frequency mixer 314 may demodulate a linear signal (eg, an original chirped signal) before frequency modulation from the received signal. The amplifier 315 may amplify the amplitude of the demodulated linear signal.

The spectrum analyzer 316 may compare the frequency 308 of the received radar signal reflected from the target and the frequency 307 of the radar transmitted signal. For reference, the frequency 307 of the radar transmission signal may change according to a carrier frequency change indicated by the frequency modulation model 301 . The spectrum analyzer 316 may detect a frequency difference between the frequency 308 of the radar received signal and the frequency 307 of the radar transmitted signal. The frequency difference between the radar transmission signal and the radar reception signal, in the graph 309 shown in FIG. 3 , represents a constant difference during a period in which the carrier frequency linearly increases along the time axis in the frequency modulation model 301 . and is proportional to the distance between the radar sensor 310 and the target. Accordingly, the distance between the radar sensor 310 and the target may be derived from a frequency difference between the radar transmission signal and the radar reception signal. The spectrum analyzer 316 may transmit the analyzed information to a processor of the radar signal processing apparatus.

For example, the spectrum analyzer 316 may calculate the distance between the radar sensor 310 and the target according to Equation 1 below.

In Equation 1 described above, R represents the distance between the radar sensor 310 and the target. c represents the speed of light. T _chirp represents the time length of the rising period of the carrier frequency in the frequency modulation model 301 . f _IF is a frequency difference between a radar transmission signal and a radar reception signal at any point in the rising section, and may be referred to as an intermediate frequency or a beat frequency. B represents the modulation bandwidth. For reference, the intermediate frequency f _IF may be derived through Equation 2 below.

In Equation 2, f _IF denotes an intermediate frequency, and t _d denotes a round trip delay time for the target as a time difference (eg, delay time) between the emission timing of the radiated radar transmission signal and the reception timing of the radar reception signal. .

According to an embodiment, a plurality of radar sensors may be installed in various parts of the vehicle, and based on the information sensed by the plurality of radar sensors, the radar signal processing apparatus may reach a target in all directions of the vehicle. Distance, direction, and relative velocity can be calculated. The radar signal processing device may be mounted on a vehicle, and may provide various functions (eg, ACC, AEB, BSD, LCA, etc.) helpful for driving by using the calculated information.

Each of the plurality of radar sensors may radiate a radar transmission signal including a frequency-modulated chirp signal based on a frequency modulation model to the outside, and receive a signal reflected from the target. The processor of the radar signal processing apparatus may determine a distance from each of the plurality of radar sensors to the target from a frequency difference between the emitted radar transmission signal and the received radar reception signal. In addition, when the radar sensor 310 includes a plurality of channels, the processor of the radar signal processing apparatus may derive the angle of arrival of the received radar signal reflected from the target by using the phase information of the radar data.

The radar sensor 310 may use a wide bandwidth and adopt a MIMO method to meet the needs of a wide field of view (FoV) and high resolution (HR) of various applications. The distance resolution may be increased through a wide bandwidth, and the angular resolution may be increased through the MIMO method. The distance resolution may indicate how small units of distance information about the target can be discriminated, and the angular resolution may indicate how small units of which angle of arrival information about the target can be discriminated. For example, the radar sensor 210 may use a wide band such as 4 GHz, 5 GHz, or 7 GHz instead of a narrow band such as 200 MHz, 500 MHz, and 1 GHz. According to the adoption of the MIMO scheme, the possibility of phase distortion between channels may increase. Accordingly, there is a need for correcting the phase distortion of the array antenna for a wide FoV.

The processor of the radar signal processing apparatus may correct the phase distortion of the array antenna in the process of processing radar data to generate target information. According to an embodiment, the beam pattern of the array antenna includes a feedline error due to a difference in feedline delay between channels, a phase shift error of radar data according to a frequency modulation characteristic of a frequency modulation model, and a beam of the array antenna. It may be modeled as including an element beam pattern (EBP) error that occurs due to the influence of the beam pattern of the antenna element on the pattern. The processor may correct each error to remove phase distortion existing in the beam pattern of the array antenna. A correction process for each error will be described in detail later.

4 illustrates a process of processing a radar signal through phase distortion correction according to an exemplary embodiment. Referring to FIG. 4 , in step 410 , the radar signal processing apparatus acquires radar data. According to an embodiment, the radar signal processing apparatus is based on the radar transmission signal transmitted through the array antenna of the radar sensor and the radar transmission signal reflected by the target based on the frequency modulation model and the radar reception signal received through the array antenna Thus, radar data can be generated. For example, the radar signal processing apparatus may generate an intermediate frequency signal based on a radar transmission signal and a radar reception signal, and may generate radar data through a sampling operation on the intermediate frequency signal.

The radar signal processing apparatus may generate information about the target based on the radar data. For example, the radar signal processing apparatus performs distance FFT in step 430 , performs Doppler FFT in step 440 , performs CFAR in step 450 , and estimates an angle of arrival in step 440 . can do. Through this, the radar signal processing apparatus may acquire information about the target, such as distance, speed, and direction.

More specifically, radar data is three-dimensional data, with each axis representing the time it takes from when any electromagnetic wave is transmitted until it is received by the radar sensor, the change between the transmitted chirp signals in one scan, and the virtual antenna. It can respond to changes in the received chirp signals. Each axis of the radar data may be converted into a distance axis, a radial velocity axis, and an angle axis through preprocessing. The line of sight speed may indicate the relative speed of the target when the radar sensor looks at the target.

For example, the radar signal processing apparatus may process the radar data in the order of distance FFT, Doppler FFT, and angle of arrival estimation, but information corresponding to each axis of the radar data includes information that is separable from each other. Therefore, even if the FFT operation and the DBF (Digital Beamforming) operation are applied by changing the processing order, the same result may be exhibited. For reference, the angular axis may be an axis related to an azimuth angle. However, in the present specification, the description will be mainly focused on the horizontal angle, but the present invention is not limited thereto, and the angular axis may be an axis related to both a horizontal angle and an elevation angle.

For example, the radar signal processing apparatus may obtain a distance value by applying an FFT operation to a time taken from transmission and reception of electromagnetic waves in radar data to a distance FFT operation. Also, the radar signal processing apparatus may estimate an angle corresponding to the arrival direction of the radar signal reflected from the target through the angle of arrival estimation operation. For example, the radar signal processing apparatus may estimate the angle of arrival using the MUSIC algorithm, the Bartlett algorithm, the MVDR algorithm, digital beam forming (DBF), and Estimation of Signal Parameter via Rotational invariance Techniques (ESPIRT). Also, the radar signal processing apparatus may estimate a radial velocity (eg, a Doppler velocity) from a signal change between chirp signals along a Doppler axis through Doppler FFT. As a Doppler FFT operation, the radar signal processing apparatus may obtain a gaze velocity at a corresponding distance and a corresponding angle by applying an FFT operation to a signal change between chirped signals at an arbitrary distance and an arbitrary angle.

The CFAR technique is a technique for determining the probability that the point is a target based on the surrounding cells for a random point (eg, cell under test (CUT)). noise floor), etc.) and may be a thresholding technique based on a signal strength sensed for a corresponding point by a radar sensor and a threshold value. For example, when a signal strength sensed from a corresponding point is greater than a threshold value determined based on a signal strength sensed from a neighboring cell, the radar signal processing apparatus may determine the corresponding point as a target. Conversely, when the signal strength sensed from the corresponding point is less than or equal to the threshold, the radar signal processing apparatus may determine the corresponding point as a non-target.

According to an embodiment, the beam pattern of the array antenna of the radar sensor may be defined as in Equation 3 below.

In Equation 3, n is an index for identifying antenna elements. n may have an integer value between 1 and N. N is the total number of antenna elements. EPn(θ) is the EBP of the nth antenna element with respect to the angle of arrival θ. k ^t is the wave number at time t. d is the distance between two adjacent antenna elements. φ _n represents the feedline delay of the nth antenna element.

에 따른 피드라인 에러, k^t에 따른 FM 기생 에러(Frequency Modulation parasitic error), EPn(θ)에 따른 EBP 에러를 포함할 수 있다. 이러한 에러들은 위상 왜곡을 발생시킬 수 있다. 보다 구체적으로,

은 채널들 간에 피드라인 지연의 차이가 발생할 수 있음을 나타내며, 이러한 차이에 의해 피드라인 에러가 발생할 수 있다. 또한, k^t는 시간에 따라 파수, 파장, 주파수가 변화될 수 있음을 나타낸다.Referring to Equation 3, the beam pattern is

It may include a feedline error according to k ^t , a frequency modulation parasitic error according to k t , and an EBP error according to EPn(θ). These errors can cause phase distortion. More specifically,

indicates that a difference in feedline delay may occur between channels, and this difference may cause a feedline error. Also, k ^t indicates that the wavenumber, wavelength, and frequency may change with time.

As the radar signal processing apparatus employs the FMCW method, a frequency change according to time may occur, and a phase shift error according to a frequency change may be included in the radar data. Since a phase shift error occurs due to frequency modulation, the phase shift error may be referred to as an FM parasitic error. Also, EPn(θ) may affect an Array Factor (AF) of the array antenna to apply a deformation to the AF. Accordingly, an EBP error may occur due to EPn(θ).

In the process of processing the radar signal, the above error can be corrected. For example, error correction may include correction of the feedline error of step 420 , correction of the frequency modulation parasitic error of step 461 , and correction of the EBP error of step 462 . Hereinafter, each error correction process will be described in detail.

First, in step 420, the radar signal processing apparatus corrects a feedline error in the radar data. According to an embodiment, the radar signal processing apparatus may calibrate radar data using a calibration vector. The correction vector may be designed to correct a feedline error according to a feedline delay of channels of the array antenna.

For example, in a test environment (eg, an anechoic room), when a radar transmission signal is reflected by a target located in front of a radar sensor and the radar reception signal is received through a plurality of channels of the radar sensor, the reference of the radar reception signal Phase components with respect to channels of the data (radar data acquired in a state where the angle of the target is known in the test environment) must have the same value. For example, since the target is located in front of the radar sensor, the angle of arrival of the reference data corresponds to 0 degrees. In this case, the phase components of each channel should all have the same value corresponding to 0 degrees.

If each phase component does not have the same value, a vector for correcting each phase component to have the same value may be determined. The calibration vector may be determined through this process. Accordingly, when a calibration vector is determined in advance through this process, and a radar reception signal is received from a target located in front of the radar sensor, the calibration vector is set so that phase components related to channels of the radar data have the same value. data can be corrected.

In step 460, the radar signal processing apparatus estimates an angle of arrival. In this case, FM parasitic error correction in step 461 and EBP error correction in step 462 may be performed in relation to the angle of arrival estimation.

First, the FM parasitic error can be corrected through various methods capable of compensating for a change in frequency over time. According to an embodiment, a direction matrix may be designed to reflect a change in frequency. The direction matrix may include a plurality of direction vectors each corresponding to a predetermined angle, and each direction vector may have phase values for each channel. Each phase value may be determined mathematically. For example, assuming that a radar reception signal is received from a target of a known first angle, phase values corresponding to the corresponding radar reception signal may be calculated, and the phase values are assigned to the first direction vector of the first angle. can be Through this process, phase values of each direction vector may be calculated.

When a radar reception signal is received in the course of actual use of the radar after the direction matrix is completed, any one direction vector corresponding to the radar data is selected based on the operation based on the radar data of the radar reception signal and each direction vector of the direction matrix can be The angle of arrival of the radar reception signal may be estimated as the angle of the selected direction vector. For example, when the first direction vector is selected from the direction matrix, the angle of arrival of the radar reception signal may be estimated as the first angle. The actual use process (online) may be understood to be distinct from the design and derivation process (offline).

According to an embodiment, a change in frequency over time may be considered in the process of deriving the above-described direction matrix. For example, a separate sub-direction matrix may be designed according to a sample index. In this case, after the direction matrix is derived, the corresponding direction matrix may be selected according to the sample index of the radar data in the course of actual use of the radar. For example, it may be assumed that a first sub-direction matrix of a direction matrix is designed for a first sample index and a second sub-direction matrix of a direction matrix is designed for a second sample index.

In this case, in order to estimate the angle of arrival in the course of actual use, the first sub-direction matrix is used for the first sub-radar data of the radar data corresponding to the first sample index, and the first sub-direction matrix is used for the radar data corresponding to the second sample index. A second sub-direction matrix may be used for the second sub-radar data. As such, the FM parasitic error can be corrected through a direction matrix designed in consideration of a change in frequency. A process of deriving the direction matrix will be described in more detail later.

According to an embodiment, the EBP error may be corrected through an EBP error model. The EBP error model may represent EBP error values for each angle. For example, the radar signal processing apparatus may determine an EBP error value with respect to the initial angle of arrival corresponding to the radar data based on the EBP error model, and estimate the final angle of arrival by correcting the initial angle of arrival with the determined EBP error value. have. The EBP error model may be generated by estimating EBP error values for other angles based on EBP error values for base angles measured through a test. For example, different EBP error values may be estimated through polynomial regression (PR) or various interpolation and fitting techniques.

According to an embodiment, the feedline error, the FM parasitic error, and the EBP error may be corrected sequentially. For example, if a correction vector capable of correcting a feedline error is obtained, then a corrected direction matrix may be derived using AF from which the feedline error is removed. Then, corrected reference data is obtained by applying a correction vector to the reference data obtained with respect to the base angles, and EBP error values about the base angles can be obtained based on the calibrated reference data and the calibrated direction matrix. have. Thereafter, an EBP error model may be obtained by performing polynomial regression, interpolation, or fitting through the EBP error values for the base angles. Errors that may complicate each other can be removed relatively easily through this sequential corrective operation.

5 illustrates a feedline delay of an array antenna according to an exemplary embodiment. Referring to FIG. 5 , the radar sensor 500 may include antenna elements 1 to N of an array antenna and a power divider 510 . Neighboring antenna elements (eg, antenna elements 1, 2) are defined as being separated by a distance d. The antenna elements 1 to N each individually form an EBP, and the entire beam pattern of the array antenna may be formed through each EBP.

The power divider 510 may distribute the transmission signal to the antenna elements 1 to N. At this time, the time the transmission signal arrives at each of the antenna elements 1 to N due to a physical factor (eg, the length of the transmission line, impedance, etc.) between the power divider 510 and the antenna elements 1 to N there may be differences in For example, the time at which the transmission signal arrives at the antenna element 1 and the time at which the transmission signal arrives at the antenna element 2 may be different. Errors due to the difference of the antenna elements (1 to N) transmission time that signals travel may be referred to as respective feed line delay (φ ₁ to φ _N), the feed line delay (φ ₁ to φ _N) to the feed line This may be referred to as an error.

성분으로 반영되어 있고,

성분은 교정을 통해 제거될 수 있다.When a radar transmission signal is transmitted to each channel of the array antenna, since the phase of each channel is assumed to be the same, such a feedline error may cause phase distortion in the radar signal. Therefore, it is necessary to remove such phase distortion through correction. In Equation 3 above, the feed line error is

is reflected as an ingredient,

Components can be removed through calibration.

6 illustrates correction of a feedline error using a correction vector according to an embodiment. Graphs 611 and 613 of FIG. 6 correspond to reference data measured with respect to a 45 degree target in a test environment. The graph 611 shows the reference data before the feedline error is corrected, and the graph 613 shows the reference data after the feedline error is corrected. The graph 611 shows a result different from the target angle (45 degrees), and the graph 613 shows a result matching the target angle (45 degrees). The graph 612 represents a correction value for each channel of a correction vector for correcting a feedline error.

The calibration value of the calibration vector may be obtained in various ways. According to an embodiment, a calibration value may be derived from reference data of a target positioned in front of the radar sensor (thus, the angle of the target is 0 degrees). More specifically, reference data may be obtained by radiating a radar signal to a target located in front of the radar sensor in a test environment and receiving a signal reflected by the target through a plurality of channels. Since the reference data was acquired in a state where the target angle was 0 degrees, the phase components of each channel of the reference data should have the same value (eg, e ⁰ =1).

If the phase components of each channel do not have the same value, calibration values that cause the phase components to have the same value may be calculated, and a calibration vector having the corresponding calibration values may be derived. The calibration vector and calibration values may be expressed as in Equation 4 below.

In Equation 4, C denotes a calibration vector, and n denotes a channel. n may have a value of 1 to N. C _n (θ ₀ ) means a correction value for the nth channel when the angle of the target is θ _{0 .} θ ₀ may represent 0 degrees. G represents reference data. G _n (θ ₀ ) is reference data corresponding to a target of an angle of _{θ 0 , and |G n} (θ ₀ )| is an absolute value of _{G n} (θ _{0 ).} By _{dividing G n} (θ ₀ ) by |G _n (θ ₀ )| _{, the magnitude component is removed from G n} (θ ₀ ), leaving only the phase component.

When the target is positioned at 0 degrees, the phase components of each channel of the reference data must have the same value (eg, e ⁰ =1). If a feedline error is present, the value of each phase component may not be equal to one. In this case, _{a channel other than G n} (θ ₀ )/|G _n (θ ₀ )|=1 exists, and the phase component of the corresponding channel is changed to G _n (θ ₀ )/|G _n (θ ₀ )| It can be divided by 1. Accordingly, a calibration vector may be defined as in Equation 4, and a calibration value of the calibration vector may be determined by measuring a phase component of reference data of each channel.

When all of the calibration values of the calibration vector are determined and the calibration vector is derived, the feedline error may be removed from the radar data by dividing the phase component of each channel by the calibration value in the course of actual use. For example, the phase component of the first channel of the _{radar data is divided by the correction value of C 1} (θ ₀ ), the phase component of the nth channel of the radar data is divided by the correction value of C ₁ (θ ₀ ), and the radar data feedline errors can be eliminated. The beam pattern 602 of FIG. 6 represents the result of removing the feed line error component from the beam pattern 601 through the correction vector.

7 is a diagram illustrating receiving antenna elements of an array antenna according to an exemplary embodiment. Referring to FIG. 7 , the array antenna 710 may include a plurality of reception antenna elements 711 to 714 . A plurality of channels may be formed through the plurality of reception antenna elements 711 to 714 . Although not shown in FIG. 7 , the array antenna 710 may further include at least one of at least one transmit antenna element and at least one receive antenna element.

When the radar sensor implements a plurality of channels, the phase information of the radar data may indicate a phase difference between a phase of a signal received through each channel and a reference phase. The reference phase may be an arbitrary phase or may be set to a phase of one channel among a plurality of channels. For example, the radar signal processing apparatus may set a phase of a reception antenna element adjacent to a corresponding reception antenna element as a reference phase with respect to one reception antenna element.

In addition, the processor may generate a radar vector of a dimension corresponding to the number of channels of the radar sensor from the radar data. For example, in the case of a radar sensor including seven channels, the processor may generate a seven-dimensional radar vector including a phase value corresponding to each channel. The phase value corresponding to each channel may be a numerical value indicating the above-described phase difference.

For example, it is assumed that the radar sensor is configured with one transmission channel and four channels. In this case, the radar signal radiated through the transmission channel may be received through four channels after being reflected by the target. As shown in FIG. 7 , when the array antenna 710 includes a plurality of reception antenna elements 711 to 714 , the phase of a signal received by the reception antenna element 711 may be set as a reference phase. . When the radar receive signal 708 reflected from the same target is received at the array antenna 710 , an additional distance between the distance from the target to the receive antenna element 711 and the distance from the target to the receive antenna element 712 . distance) Δ can be expressed as in Equation 5 below.

In Equation 5, θ may represent an angle of arrival at which the radar reception signal 708 is received from the target. d may represent an interval between receiving antenna elements. Since the speed c of light in air is treated as a constant and c=fλ, the phase change W in the receiving antenna element 712 due to the above-described additional distance Δ can be derived as in Equation 6 below.

The phase change W may correspond to a phase difference of the signal waveform received by the reception antenna element 712 based on the signal waveform received by the reception antenna element 711 . In Equation 6 described above, f is the frequency of the radar reception signal 708 , and λ is the wavelength of the radar reception signal 708 . λ is inversely proportional to the frequency f. When the carrier frequency change due to the frequency modulation model is small, the frequency f in Equation 6 above may be regarded as _{a single initial frequency (eg, f 0 ) in the frequency modulation model.} Accordingly, when only the phase change W is determined based on the received signal, the radar signal processing apparatus may determine the angle of arrival θ.

However, if the carrier frequency of the frequency modulation model changes in a wide band (eg, bandwidth of 2 GHz or more such as 7 GHz, 5 GHz, or 7 GHz), the change in the carrier frequency cannot be ignored, so there is an error in the process of estimating target information such as the angle of arrival. can occur

8 illustrates a phase change according to a carrier frequency change for each sampling point in a radar signal processing method according to an embodiment. Hereinafter, for convenience of description, an example in which the frequency modulation model 801 represents a pattern in which the carrier frequency linearly increases during a given transmission time will be described. For example, in the frequency modulation model 801, the sampling period may be a linearly _{increasing period from the initial frequency f 0} to the last frequency f _{0 +BW increased by the bandwidth BW.} The bandwidth BW is a sampling bandwidth corresponding to the sampling period and may be smaller than the modulation bandwidth B described above in Equation (1). The radar reflection signal 802 received from the radar sensor may indicate a frequency corresponding to the aforementioned carrier frequency. The radar reflection signal 802 may be individually sensed by a plurality of receive antenna elements.

Since the radar reflection signal 802 corresponding to one chirp is reflected from the same target point, the round trip delay time may be the same. Since the round trip delay time is the same, the radar signal processing apparatus should calculate the same angle of arrival at the initial, intermediate, and end points of the corresponding radar reflection signal 802 . However, since the phase change W varies with time even within one chirp due to a change in frequency, an error (the FM parasitic error described above) may occur in the phase component for estimating the angle of arrival.

For example, the radar signal processing apparatus may include a first intermediate frequency signal 811 based on a radar reception signal and a radar transmission signal received through each of the first reception antenna element, the second reception antenna element, and the third reception antenna element. ), a second intermediate frequency signal 812 , and a third intermediate frequency signal 813 may be generated. In this case, the phase change 881 of the initial time point, the phase change 882 of the intermediate time point, and the phase change 883 of the end time point within an arbitrary chirp may be changed.

는 위상이 정규화된 레이더 데이터를 나타낸다. 위상이 정규화된 레이더 데이터의 제1 중간 주파수 신호(821), 제2 중간 주파수 신호(822), 및 제3 중간 주파수 신호(823)는, 도 8에 도시된 바와 같이 한 레이더 반사 신호(802)의 처프 주기 동안 동일한 위상 변화(891, 892, 893)를 나타낼 수 있다. 따라서 레이더 신호 처리 장치는 위상 정규화를 통해 FM 기생 에러를 제거할 수 있다.The radar signal processing apparatus according to an embodiment may generate phase-normalized radar data through phase normalization 850 of a carrier frequency. In Figure 8, y(t) represents the intermediate frequency signal,

denotes phase-normalized radar data. The first intermediate frequency signal 821 , the second intermediate frequency signal 822 , and the third intermediate frequency signal 823 of the phase-normalized radar data is a radar reflection signal 802 as shown in FIG. 8 . may exhibit the same phase changes 891, 892, and 893 during the chirp period of . Therefore, the radar signal processing apparatus may remove the FM parasitic error through phase normalization.

9 illustrates a sampling process of an intermediate frequency signal according to an exemplary embodiment. The radar signal processing apparatus may obtain an intermediate frequency signal based on the radar transmission signal 970 and the radar transmission signal 970 generated based on the frequency modulation model and the received radar reception signal 980 being reflected by the target. have. For example, the radar signal processing apparatus may calculate an intermediate frequency signal corresponding to a frequency difference between the frequency 907 of the radar transmission signal 970 and the frequency 908 of the radar reception signal 980 .

The radar signal processing unit cannot directly measure the _{intermediate frequency f IF .} The radar signal processing apparatus measures the signal waveform 981 of the radar reception signal 980, the signal waveform 971 of the radar transmission signal 970 given in advance, and the signal waveform 981 of the measured radar reception signal 980 An intermediate frequency signal may be generated based on . The radar signal processing apparatus may calculate the above-described intermediate frequency signal from frequency mixing (eg, multiplication) of the radar transmission signal 970 and the radar reception signal 980 .

_{y m} (t) shown in FIG. 9 may represent an intermediate frequency signal calculated for the radar reception signal 980 and the radar transmission signal 970 received by the mth reception antenna element among the M reception antenna elements. have. M may be an integer of 2 or more, and m may be an integer of 1 or more and M or less. For reference, in FIG. 9 , the radar transmission signal 970 , the radar reception signal 980 , and the intermediate frequency signal are only shown as frequencies lower than the actual frequency for better understanding, and the frequencies of each signal are limited to those shown in FIG. 9 . is not doing

The radar signal processing apparatus may obtain sampling data by sampling the intermediate frequency signal at a plurality of sampling points. The sampling data may include sampling values obtained at predetermined sampling points. For example, referring to FIG. 9 , s _m (i) may represent a value obtained by sampling a signal strength received by an m-th reception antenna element among M reception antenna elements at an i-th sampling point. i is a sample index, N is the sampling number of the intermediate frequency signal, N may be an integer greater than or equal to 1, and i may be an integer greater than or equal to 1 and less than or equal to N. According to an embodiment, radar data may be determined based on the sampling data. For example, the sampling data may be determined as radar data, or some processing may be performed on the sampling data and the result may be determined as radar data.

As shown in FIG. 9 , due to the carrier frequency change in the radar transmission signal 970 and the radar reception signal 980 , the phase change may vary for each sampling point. Due to the frequency change for each sampling point described above, as shown in FIG. 9 , the frequency and phase of the intermediate frequency signal are not fixed but may vary. For convenience of description, in FIG. 9 , a predetermined time delay between the radar transmission signal 970 and the radar reception signal 980 is shown clearly. However, since the radar signal propagates at the speed of light, this time delay is very small and the frequency difference is also very small, so the carrier frequency of the radar transmission signal 970 and the carrier frequency of the radar reception signal 980 at a given sampling point are can be considered substantially the same.

For reference, although a method of processing a radar signal is mainly described herein, the present disclosure is not limited thereto. The phase normalization described above may be generally applied to a reflected signal received by a signal processing apparatus performing a signal processing method transmitting a transmission signal whose carrier frequency varies according to a frequency modulation model, and the transmission signal being reflected from a target point. have. For example, the transmission signal and the reflected signal are signals propagated with a wave according to the carrier frequency, and may be not only a radar signal but also an ultrasonic signal, an electromagnetic wave signal, and an optical signal.

10 illustrates correction of a frequency modulation parasitic error through phase normalization according to an embodiment. As described above, according to the FMCW method, the carrier frequency may change over time, and the AF of the array antenna may be defined as in Equation 7 below by reflecting the change in the carrier frequency.

성분이 제거된 것으로 표시되어 있다. 수학식 7의 k^t는 아래 수학식 8과 같이 나타낼 수 있다.In Equation 7, n is an index for identifying antenna elements. n may have an integer value between 1 and N. N is the total number of antenna elements. k ^t is the wave number at time t. d is the distance between two adjacent antenna elements. φ _n represents the feedline delay of the nth antenna element. Equation 7 represents AF in a state in which the feedline error is removed, and

Ingredients are marked as removed. ^{k t} in Equation 7 can be expressed as Equation 8 below.

In Equation 8, f _c represents the carrier frequency, B represents the modulation bandwidth, T _chirp represents the time length of the rising section of the carrier frequency in the frequency modulation model, c represents the speed of light, λ represents the wavelength, k denotes the wave number, and t denotes the time. According to Equations 7 and 8, λ and k are dependent on time t. This is due to the frequency modulation model of the FMCW, and accordingly, an FM parasitic error may occur. When phase normalization is applied to Equations 7 and 8, Equations 9 and 10 can be derived. In Equations 9 and 10, k corresponds to a constant, so that an FM parasitic error does not occur.

The phase change W of Equation 6 may be redefined as Equation 11 below in terms of ω(τ).

In Equation 11, ω(τ) is a phase change according to τ. τ is the sample index (time index). Since the FM parasitic error occurs as the carrier frequency changes while the angle of arrival is fixed, in defining ω(τ), it can be defined as changing λ while θ is fixed. θ _GT represents a fixed angle of arrival. If it is defined that θ is changed, λ may be fixed as _{λ 0 .} Accordingly, Equation 12 can be defined. Also, Equation 13 can be defined according to Equation 12.

Based on Equation 8, it can be expressed as f(τ)=f _c +(B/T _chirp )τ, and by substituting this into Equation 13, a change in the angle of arrival according to τ can be calculated. For example, the line 1010 of FIG. 10 may be derived according to the calculation result. Lines 1010 and 1020 indicate the angle of arrival according to the sample index. The slope of the line 1010 means that the angle of arrival changes over time, and there is an FM parasitic error. Phase normalization may be performed through various methods for changing the line 1010 to the line 1020 by compensating for the change in the angle of arrival.

According to an embodiment, the FM parasitic error may be compensated by referring to a radar design parameter (eg, f(τ)). For example, a separate sub-direction matrix may be designed according to a sample index (eg, τ=0, 1, 2, ..., 511). Accordingly, the direction matrix may include sub-direction matrices corresponding to different sample indices, respectively. In this case, sub-direction matrices corresponding to the number of sample indices (eg, 512) may be derived. In this case, Equation 11 may be used to determine the phase values for each channel of the direction vectors of each sub-direction matrix.

For example, it may be assumed that a first sub-direction matrix of a direction matrix is designed for a first sample index and a second sub-direction matrix of a direction matrix is designed for a second sample index. In this case, in the actual use process, the first sub-direction matrix is used for the first sub-radar data of the radar data corresponding to the first sample index, and the second sub-radar data of the radar data corresponding to the second sample index is used. A second sub-direction matrix may be used. Hereinafter, sub-radar data and sub-direction matrix are further described.

For example, the sampling data Y may be expressed as in Equation 14 below. The sampling data Y may correspond to radar data.

In Equation 14, i is a sample index, N is a sampling number of sampling data, N may be an integer of 1 or more, and i may be an integer of 1 or more and N or less. When the antenna array of the radar sensor forms M channels, sampling data Y(i) at the i-th sampling point corresponding to an arbitrary sample index i may be expressed as in Equation 15 below. The sampling data Y(i) may correspond to sub-radar data.

In Equation 15, s _m (i) may represent a value obtained by sampling the signal strength received by the m-th sub-receiving antenna among the M sub-receiving antennas at the i-th sampling point. M may be an integer of 2 or more, and m may be an integer of 1 or more and M or less. Also, the direction matrix may be expressed as Equation 14 below, and the direction vector of the direction matrix may be expressed as Equation 17 below.

In Equation 16, the direction matrix A _fi may be expressed as a set of direction vectors α _fi (θ _{k ).} Here, K may be an integer of 1 or more, and k may be an integer of 1 or more and K or less. For example, when the FoV of the radar is 180 degrees and K is 512, the angular resolution may be 0.35 degrees. However, the present invention is not limited thereto, and when K is 180, the angular resolution corresponds to 1 degree, and when K is 360, it may correspond to 0.5 degrees. K may be determined according to the desired angular resolution (eg, 1 degree or less). _{Here, each A fi} (eg, A _{f1 , A f2} , ..., A _fN ) according to the change of the sample index i may correspond to a sub-direction matrix.

는 i번째 샘플 인덱스에서의 캐리어 주파수에 대응하는 파장을 나타내고,

는 A_fi에서 k번째 방향 각도를 나타낼 수 있다.

는 주파수 변조 모델의 i번째 샘플 인덱스에 대응하는 캐리어 주파수에서 방향 각도

에 대응하는 방향 벡터를 나타낸다.In Equation 17, d denotes an interval between radar antenna elements.

represents the wavelength corresponding to the carrier frequency in the i-th sample index,

may represent the k-th direction angle in _{A fi .}

is the direction angle at the carrier frequency corresponding to the i-th sample index of the frequency modulation model.

represents the direction vector corresponding to .

일 확률에 대응하는 값으로서 도래각 정보를 나타낼 수 있다. 예를 들어, i번째 샘플링 데이터(서브 레이더 데이터에 대응) Y(i)와 i번째 방향 행렬(서브 방향 행렬에 대응) A_fi 간의 연산 결과는 i번째 샘플링 데이터(서브 레이더 데이터에 대응) Y(i)에 관한 도래각 정보에 해당할 수 있다.For reference, A _fi may be a K x M matrix including K rows and M columns. _{A matrix multiplication result A fi} Y(i) between the direction matrix A _fi according to Equation 16 and Y(i) according to Equation 15 may be calculated as a K×1-dimensional vector. As a result of matrix multiplication _{, the k-th row element in A fi} Y(i) is the angle of arrival of Y(i) equals the k-th steering angle.

The angle of arrival information may be represented as a value corresponding to one probability. For example, the operation result between the i-th sampling data (corresponding to the sub-radar data) Y(i) and the i-th direction matrix (corresponding to the sub-direction matrix) A _fi is the i-th sampling data (corresponding to the sub-radar data) Y( It may correspond to angle of arrival information about i).

11 and 12 illustrate correction of device beam pattern errors according to an exemplary embodiment. The beam pattern of the array antenna may be defined as in Equation 18 below.

Referring to Equation 18, after the feed line error is removed and the FM parasitic error is removed, the beam pattern of the array antenna may be classified into an Array Factor (AF) component and an Element Pattern (EP) component. For example, as described above, the feedline error can be removed through a calibration vector, and the FM parasitic error can be removed through a specialized design of the direction matrix.

Assuming that each antenna element has the same EP, that is, EP=EP ₁ =...=EP _N (where N is the number of antenna elements), graphs regarding the antenna pattern as shown in FIG. 11 are obtained can be In FIG. 11 , a graph 1110 corresponds to EP, a graph 1120 corresponds to AF, and a graph 1130 corresponds to a beam pattern (BP). Referring to FIG. 11 , a beam pattern is not completely formed according to AF, but is influenced by EP, and in this process, an error due to EP may be included in radar data. Such an error may be referred to as an EBP error.

According to an embodiment, the EBP error model may be used for correction of the EBP error. The EBP error model may represent EBP error values for each angle. The EBP error model may be generated by estimating EBP error values for other angles based on EBP error values for base angles measured through a test. For example, different EBP error values may be estimated through polynomial regression (PR) or various interpolation and fitting techniques.

For example, the basic EBP error may be determined by measuring reference data for several basic angles (eg, -40 degrees, 0 degrees, 40 degrees). A feedline error of the reference data may be corrected through a calibration vector, and an EBP error may be determined through comparison between the corrected reference data and sub-direction matrices for each sample index. For example, if the corrected reference data is obtained through a target of -60 degrees, and the angle of arrival is estimated to be -58 degrees by comparison with the sub-direction matrix, the EBP error for the angle of arrival of -58 degrees can be determined to be 2 degrees.

When the basic EBP errors with respect to some basic angles are determined as described above, an EBP error model may be obtained by performing polynomial regression or fitting based on them. Error values corresponding to the required resolution may be obtained through the EBP error model. The graphs 1210 to 1230 of FIG. 12 show the results of repeating the polynomial regression twice, three times, and four times based on the reference data, and each line of the graphs 1210 to 1230 is the EBP error model. can respond

The reason that the EBP error model can be estimated based on some basic EBP errors is that the feedline error and FM parasitic error have been removed before that. If the EBP error model is estimated based on some basic EBP errors in the presence of the feedline error and the FM parasitic error, the EBP error model cannot be successfully estimated. In this case, it is necessary to secure each reference data according to the required resolution, and since this process must be very precise, a lot of effort is required. Accordingly, the EBP error model according to the embodiments may reduce such effort.

The EBP error model can be utilized in various ways. According to an embodiment, when the initial angle of arrival is derived based on the sub-direction matrices for each sample index and the radar data calibrated by the calibration vector in the course of actual use, the matching relationship between the angle of arrival defined in the EBP error model and the EBP error is obtained. It is possible to estimate the final angle of arrival by correcting the initial angle of arrival. More specifically, the radar signal processing apparatus may determine an EBP error value matched to the initial angle of arrival based on the EBP error model, and may estimate the final angle of arrival by correcting the initial angle of arrival with the determined EBP error value. The correction of the initial angle of arrival may be performed inversely to the process of estimating the EBP error. As an example, based on the corrected radar data and sub-direction matrices, an initial angle of arrival of -58 degrees is derived, and if the EBP error for an angle of arrival of -58 degrees is 2 degrees, the final angle of arrival can be estimated as -60 degrees. .

13 illustrates a radar signal processing method according to an embodiment. Referring to FIG. 13 , the radar signal processing apparatus receives a radar transmission signal and a radar transmission signal transmitted through an array antenna of a radar sensor based on a frequency modulation model in step 1310, which are reflected by a target and received through the array antenna. Based on the radar received signal, the radar data is generated, and in step 1320, the radar data is calibrated using a correction vector that corrects the feedline error according to the difference in the feedline delay between the channels of the array antenna, step ( 1330), an angle of arrival corresponding to the corrected radar data is estimated using a direction matrix in which a phase shift of the corrected radar data according to the frequency modulation characteristic of the frequency modulation model is reflected. As mentioned above, it is not corrected until the phase shift (FM parasitic error) with the correction vector. For example, the phase shift may be corrected with sub-direction matrices for each sample index. The direction matrix in which the phase shift of the corrected radar data is reflected may correspond to these sub-direction matrices. In addition, the matters described with reference to FIGS. 1 to 12 and 14 may be applied to the radar signal processing method.

14 illustrates an electronic device according to an embodiment. Referring to FIG. 14 , the electronic device 1400 may perform the radar signal processing method described above. For example, the electronic device 1400 may functionally and/or structurally include the radar processing device 200 of FIG. 2 . The electronic device 1400 is, for example, an image processing device, a smartphone, a wearable device, a tablet computer, a netbook, a laptop, a desktop, a personal digital assistant (PDA), a head mounted display (HMD), a vehicle (eg, : autonomous driving vehicle), and a driving assistance device mounted on the vehicle.

Referring to FIG. 14 , an electronic device 1400 may include a processor 1410 , a storage device 1420 , a camera 1430 , an input device 1440 , an output device 1450 , and a network interface 1460 . . The processor 1410 , the storage device 1420 , the camera 1430 , the input device 1440 , the output device 1450 , and the network interface 1460 may communicate with each other through the communication bus 1470 .

The processor 1410 executes functions and instructions for execution in the electronic device 1400 . For example, the processor 1410 may process instructions stored in the storage device 1420 . The processor 1410 may perform one or more operations described above with reference to FIGS. 1 to 13 .

The storage device 1420 stores information or data necessary for the execution of the processor. For example, the pre-calculated phase normalization matrix may be stored in the storage device 1420 . The storage device 1420 may include a computer-readable storage medium or a computer-readable storage device. The storage device 1420 may store instructions for execution by the processor 1410 , and may store related information while the software or application is executed by the electronic device 1400 .

The camera 1430 may capture an image composed of a plurality of image frames. For example, the camera 1430 may generate a frame image.

The input device 1440 may receive an input from the user through tactile, video, audio, or touch input. Input device 1440 may include a keyboard, mouse, touch screen, microphone, or any other device capable of detecting input from a user and communicating the detected input.

The output device 1450 may provide an output of the electronic device 1400 to the user through a visual, auditory, or tactile channel. Output device 1450 may include, for example, a display, a touch screen, a speaker, a vibration generating device, or any other device capable of providing output to a user. The network interface 1460 may communicate with an external device through a wired or wireless network. According to an embodiment, the output device 1450 provides a result of processing the radar data to the user using at least one of visual information, auditory information, and haptic information. can

For example, when the electronic device 1400 is mounted in a vehicle, the electronic device 1400 may visualize the radar image map through the display. As another example, the electronic device 1400 may change at least one of speed, acceleration, and steering of a vehicle equipped with the device 1400 based on angle of arrival information, distance information, and/or a radar image map. However, the present invention is not limited thereto, and the electronic device 1400 may perform functions such as ACC, AEB, BSD, LCA, and ego-localization. The electronic device 1400 may structurally and/or functionally include a control system for controlling the vehicle.

The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, etc. alone or in combination. The program instructions recorded on the medium may be specially designed and configured for the embodiment, or may be known and available to those skilled in the art of computer software. Examples of the computer-readable recording medium include magnetic media such as hard disks, floppy disks and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic such as floppy disks. - includes magneto-optical media, and hardware devices specially configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions include not only machine language codes such as those generated by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

Software may comprise a computer program, code, instructions, or a combination of one or more thereof, which configures a processing device to operate as desired or is independently or collectively processed You can command the device. The software and/or data may be any kind of machine, component, physical device, virtual equipment, computer storage medium or apparatus, to be interpreted by or to provide instructions or data to the processing device. , or may be permanently or temporarily embody in a transmitted signal wave. The software may be distributed over networked computer systems and stored or executed in a distributed manner. Software and data may be stored in one or more computer-readable recording media.

As described above, although the embodiments have been described with reference to the limited drawings, those skilled in the art may apply various technical modifications and variations based on the above. For example, the described techniques are performed in an order different from the described method, and/or the described components of the system, structure, apparatus, circuit, etc. are combined or combined in a different form than the described method, or other components Or substituted or substituted by equivalents may achieve an appropriate result.

Therefore, other implementations, other embodiments, and equivalents to the claims are also within the scope of the following claims.

Claims (20)

generating radar data based on a radar transmission signal transmitted through an array antenna of a radar sensor based on a frequency modulation model and a radar reception signal reflected by a target and received through the array antenna;
calibrating the radar data using a calibration vector for correcting a feedline error according to a difference in feedline delay between channels of the array antenna; and
estimating an angle of arrival corresponding to the corrected radar data using a direction matrix in which a phase shift of the corrected radar data according to the frequency modulation characteristic of the frequency modulation model is reflected
A radar signal processing method comprising a. According to claim 1,
When the radar reception signal is received from a target located in front of the radar sensor, the calibration vector corrects the radar data so that phase components of the channels of the radar data have the same value.
Radar signal processing method. According to claim 1,
The direction matrix includes sub-direction matrices corresponding to different sample indices, respectively,
Radar signal processing method. 4. The method of claim 3,
The step of estimating the angle of arrival is
obtaining a first sub-direction matrix corresponding to a first sample index from among the sub-direction matrices; and
estimating a first angle of arrival of first sub-radar data corresponding to the first sample index in the corrected radar data by using the first sub-direction matrix,
Radar signal processing method. According to claim 1,
The step of generating the radar data is
Sampling an intermediate frequency signal generated based on the radar transmission signal and the radar reception signal to generate the radar data,
Radar signal processing method. According to claim 1,
The step of estimating the angle of arrival is
and correcting an element beam pattern (EBP) error that occurs due to an effect of a beam pattern of an antenna element of the array antenna on a beam pattern of the array antenna. According to claim 1,
The step of estimating the angle of arrival is
estimating an initial angle of arrival corresponding to the corrected radar data using the direction matrix; and
estimating a final angle of arrival corresponding to the corrected radar data by removing an EBP error caused by an effect of a beam pattern of an antenna element of the array antenna on a beam pattern of the array antenna included in the initial angle of arrival
Including, radar signal processing method. 8. The method of claim 7,
The step of estimating the final angle of arrival is
determining an EBP error value with respect to the initial angle of arrival corresponding to the corrected radar data based on an EBP error model indicating EBP error values for each angle; and
estimating the final angle of arrival corresponding to the corrected radar data by correcting the initial angle of arrival with the determined EBP error value
Including, radar signal processing method. 9. The method of claim 8,
The EBP error model is
Generated by estimating EBP error values for other angles based on EBP error values for basic angles measured through a test,
Radar signal processing method. According to claim 1,
The radar transmission signal includes a chirp signal in which a carrier frequency is modulated based on the frequency modulation model,
Radar signal processing method. 11. The method of claim 10,
The radar reception signal is received through reception antenna elements of the array antenna, and the channels are formed based on the reception antenna elements,
Radar signal processing method. According to claim 1,
further estimating at least one of a distance and a speed with respect to the target based on the radar data;
further comprising,
a vehicle equipped with a radar signal processing device is controlled based on at least one of the angle of arrival, the distance, and the speed;
Radar signal processing method. 13. A computer-readable recording medium storing one or more computer programs including instructions for performing the method of any one of claims 1 to 12. a radar sensor for transmitting a radar transmission signal through an array antenna based on a frequency modulation model, and receiving a radar reception signal through the array antenna when the radar transmission signal is reflected by a target; and
generating radar data based on the radar transmission signal and the radar reception signal, and correcting the radar data using a correction vector that corrects a feedline error according to a difference in feedline delay between channels of the array antenna; A processor for estimating an angle of arrival corresponding to the corrected radar data using a direction matrix in which a phase shift of the corrected radar data according to the frequency modulation characteristic of the frequency modulation model is reflected
A radar signal processing device comprising a. 15. The method of claim 14,
When the radar reception signal is received from a target located in front of the radar sensor, the calibration vector corrects the radar data so that phase components of the channels of the radar data have the same value.
Radar signal processing unit. 15. The method of claim 14,
The direction matrix is
Each includes sub-direction matrices corresponding to different sample indices,
the processor is
Obtaining a first sub-direction matrix corresponding to a first sample index from among the sub-direction matrices, and using the first sub-direction matrix, a first sub-radar corresponding to the first sample index in the corrected radar data estimating the first angle of arrival of the data;
Radar signal processing unit. 15. The method of claim 14,
the processor is
Correcting an element beam pattern (Element Beam Pattern, EBP) error that occurs due to the effect of the beam pattern of the antenna element of the array antenna on the beam pattern of the array antenna,
Radar signal processing unit. a radar sensor for transmitting a radar transmission signal through an array antenna based on a frequency modulation model, and receiving a radar reception signal through the array antenna when the radar transmission signal is reflected by a target;
generating radar data based on the radar transmission signal and the radar reception signal, and correcting the radar data using a correction vector that corrects a feedline error according to a difference in feedline delay between channels of the array antenna; a processor for estimating an angle of arrival corresponding to the calibrated radar data using a direction matrix in which a phase shift of the calibrated radar data according to the frequency modulation characteristic of the frequency modulation model is reflected; and
A control system that controls the vehicle based on the angle of arrival
vehicle including. 19. The method of claim 18,
The direction matrix is
Each includes sub-direction matrices corresponding to different sample indices,
the processor is
Obtaining a first sub-direction matrix corresponding to a first sample index from among the sub-direction matrices, and using the first sub-direction matrix, a first sub-radar corresponding to the first sample index in the corrected radar data estimating the first angle of arrival of the data;
vehicle. 19. The method of claim 18,
the processor is
Correcting an element beam pattern (Element Beam Pattern, EBP) error that occurs due to the effect of the beam pattern of the antenna element of the array antenna on the beam pattern of the array antenna,
vehicle.

Images