A radar target detection method for complex road surfaces

Abstract

This invention discloses a radar target detection method for complex road surfaces, belonging to the field of vehicle-mounted radar technology. The method involves processing the target signal through steps such as intermediate frequency signal sampling, signal FFT operation, spectral leakage suppression, image suppression, and target search. It then filters the original target based on characteristic values ​​such as phase difference and amplitude ratio. This invention enables radar systems to effectively suppress spectral leakage by distinguishing interference signals caused by strong reflectors such as fences, overpasses, and tunnels from the target signal, reducing the number of false targets and the probability of false alarms. This results in higher target detection accuracy, greater reliability, and lower computational load for the radar system, making it applicable to complex road sections such as fences, overpasses, and tunnels.

Description

Technical Field

[0001] This invention relates to the field of vehicle radar technology, and in particular to a radar target detection method applicable to complex road surfaces. Background Technology

[0002] With the improvement of people's living standards and the advancement of technology, automobiles have become a common means of transportation. While bringing convenience, they also pose certain safety hazards. In recent years, traffic accidents have occurred frequently, largely due to drivers' poor visibility or slow reaction time. Developing a car driver assistance system that can promptly detect dangers and alert drivers could prevent many losses of life and property. Currently, various in-vehicle systems are available on the market, each with its own advantages and disadvantages. Among them, radar systems, compared to visual in-vehicle systems, are unaffected by external conditions such as weather, temperature, and humidity. Furthermore, their high precision and long-range measurement capabilities have led to the widespread application and development of in-vehicle radar systems in many areas.

[0003] In existing technologies, radar detects targets in the vehicle's environment by emitting radar signals and receiving radar feedback signals. In radar systems, phase-locked loops (PLLs) can reduce spectral leakage to some extent, but they increase hardware complexity and cost. In vehicle-mounted collision avoidance radar, directly distinguishing between large reflective objects such as fences, overpasses, and tunnels and vehicles approaching from behind is challenging, as these objects cause a rise in the symmetrical velocity spectrum. Traditional methods involve clustering, tracking, or point cloud processing of targets selected through two-dimensional FFT to reduce false alarms through target correlation, but this method requires significant computing power and memory.

[0004] Chinese Patent Publication No. CN111699404A discloses a driving assistance target acquisition method and device, radar, driving system, and vehicle. The method includes: acquiring target detection results from a radar behind the vehicle, obtaining position and speed information of each detected target behind the vehicle; then, determining filtering targets based on the position and speed information of each detected target and corresponding single-sided fence information. The filtering targets include mirror targets generated outside the fence. Thus, the filtering targets are removed from the detected targets to obtain driving assistance targets. However, the above driving assistance target acquisition method mainly targets the similarity between fence mirrors and "ghost images" in radar imaging, and primarily considers multipath radiation, mainly filtering mirror targets caused by multipath effects due to radar signal reflection. It is difficult to filter and suppress mirror interference targets caused by oncoming vehicles and large reflective objects such as fences, overpasses, and tunnels in the velocity spectrum. Furthermore, the above method lacks suppression of spectral leakage caused by DC signals in the radar system, easily misjudging DC interference signal targets as original targets, thereby increasing the number of false target samples, increasing the probability of misjudgment, and increasing the computational load of the radar system. Summary of the Invention

[0005] In order to overcome the shortcomings of the prior art, the technical problem to be solved by the present invention is to propose a radar target detection method for complex road surfaces, which can effectively suppress spectrum leakage and mirror signal interference, reduce the number of false targets and the probability of false alarms, thereby making the radar system more accurate in target detection, more reliable, and with less computation. It can be applied to complex road sections such as fences, overpasses, and tunnels.

[0006] To achieve this objective, the present invention adopts the following technical solution:

[0007] This invention provides a radar target detection method for complex road surfaces, comprising the following steps:

[0008] The radar transmits and receives the reflected intermediate frequency signal to the target, samples the intermediate frequency signal to obtain frame data containing chirp signal, and performs two FFT operations on each frame data to obtain the target's range index and velocity index.

[0009] A DC leakage region is determined on the velocity spectrum of each distance dimension. Q spectral lines are set on both sides of the DC leakage region. The DC signal within the Q spectral line interval is windowed to suppress the DC spectral line leakage component.

[0010] The maximum amplitude A is selected in the interval [2, M / 2] on the positive velocity spectrum of the distance dimension. The amplitude B of the symmetrical point of the maximum amplitude A on the negative velocity spectrum is found. The ratio R of the maximum amplitude A and amplitude B of the target in each frame is calculated. The amplitude ratio of the previous frame is R1 and the amplitude ratio of the current frame is R2. By judging the amplitude ratio R1 of the previous frame and the amplitude ratio R2 of the current frame, the spectrum is mirror suppressed.

[0011] Find P amplitude values ​​in the distance dimension. If the amplitude value exceeds a preset threshold, then perform the following: record the distance index and velocity index of the target whose amplitude value exceeds the preset threshold, calculate the difference between the real and imaginary phases and the ratio of the real and imaginary amplitudes of the target respectively. If the difference between the real and imaginary phases and the ratio of the real and imaginary amplitudes are all within the preset threshold range, then save the target and define it as a real target.

[0012] A preferred embodiment of the present invention involves image suppression of spectral lines, including:

[0013] In each frame, by using the index of the maximum amplitude A on the positive velocity spectrum, find the amplitude B of the symmetrical point of A on the negative velocity spectrum, record the index of B as Idx and keep it, where the index of the negative velocity direction in the previous frame is recorded as Idx1 and the index of the negative velocity direction in the current frame is recorded as Idx2.

[0014] Perform an OR operation between the index Idx1 of the negative velocity direction in the previous frame and the index Idx2 of the negative velocity direction in the current frame to obtain the symmetrical point region [minIdx, maxIdx] of the suppression region. The minimum value among {Idx1, Idx2} is denoted as minIdx, and the maximum value among {Idx1, Idx2} is denoted as maxIdx.

[0015] The 8 positions extending around the symmetrical point region [minIdx, maxIdx] constitute the extension region of the suppression region;

[0016] A suitable suppression function F(R) is selected for the spectral lines. The suppression function F(R) performs mirror suppression on the spectral lines in the suppression region according to the value of R.

[0017] A preferred embodiment of the present invention involves mirror suppression of spectral lines in the suppression region, comprising:

[0018] The range of the ratio R of the maximum amplitude A to the amplitude B in each frame of the target is defined as [R]. min R max ];

[0019] 1) If both the amplitude ratio R1 of the previous frame and the amplitude ratio R2 of the current frame are greater than 0, then execute:

[0020] The suppression function F(R) is based on the amplitude ratio R2 of the current frame and R...max The magnitude relationship is used to perform mirror suppression of spectral lines in the suppression region;

[0021] 2) If the amplitude ratio R1 of the previous frame is not 0, and the amplitude ratio R2 of the current frame is 0, then execute:

[0022] The suppression function F(R) is based on the amplitude ratio R1 of the previous frame and R... max The magnitude relationship is used to perform mirror suppression of spectral lines in the suppression region;

[0023] 3) If the amplitude ratio R1 of the previous frame is 0 and the amplitude ratio R2 of the current frame is not 0, then mirror suppression is not performed.

[0024] A preferred embodiment of the present invention is that the windowing process includes:

[0025] Find spectral lines in the velocity spectrum within the interval [M-Q+1, M] that are of the same order of magnitude as those in the velocity spectrum within the interval [2, 1+Q], and find the index position of these spectral lines. Suppress spectral lines of the same order of magnitude in both the interval [M-Q+1, M] and [2, 1+Q] using a window function.

[0026] The degree of suppression of the target signal of the spectral line increases as the distance between the spectral line and the DC leakage region decreases.

[0027] The preferred technical solution of the present invention is that the window function selected for the windowing process is a (1-Gaussian window), and the number of Gaussian window points is 2Q.

[0028] A preferred embodiment of the present invention comprises calculating the difference between the real and imaginary phases of the target, and the ratio of the real to the imaginary amplitudes, including:

[0029] Based on the target's distance index and velocity index, find the FFT value of the target at that point. After performing FFT on the real part and the imaginary part of the target signal respectively, calculate the difference between the phase of the real part and the phase of the imaginary part of the target signal, and the ratio of the amplitude of the real part to the amplitude of the imaginary part.

[0030] A preferred embodiment of the present invention is that the amplitudes of adjacent distance dimensions at the same speed are added together to obtain the amplitude of the real target.

[0031] The preferred technical solution of the present invention is that the total number of targets in each frame is no more than K.

[0032] A preferred embodiment of the present invention involves sampling the intermediate frequency signal, including:

[0033] The radar emits electromagnetic waves and simultaneously receives the echo signals from the reflected targets. The intermediate frequency signal is obtained by processing the frequency difference between the echo signals and the local oscillator signals.

[0034] After sampling the intermediate frequency signal, two FFT operations are performed, including:

[0035] 1) Perform an M-point FFT operation on each chirp signal to obtain the target's distance index;

[0036] 2) Perform an N-point FFT operation on each distance dimension to obtain the target's velocity index.

[0037] A preferred embodiment of the present invention is that the method for determining spectral lines of the same order of magnitude includes:

[0038] Calculate the ratio of the amplitude of spectral line one of the velocity spectrum in the interval [M-Q+1, M] to the amplitude of spectral line two of the velocity spectrum in the interval [2, 1+Q].

[0039] If the ratio of the amplitude of spectral line 1 to the amplitude of spectral line 2 is within the preset threshold range of the amplitude ratio of spectral lines, then spectral line 1 and spectral line 2 are considered to be spectral lines of the same order of magnitude.

[0040] The beneficial effects of this invention are as follows:

[0041] The radar target detection method for complex road surfaces provided by this invention processes the target signal through steps such as intermediate frequency signal sampling, signal FFT operation, spectrum leakage suppression, image suppression, and target search. This method enables the radar system to effectively suppress spectrum leakage and distinguish the interference signals caused by strong reflectors such as fences, overpasses, and tunnels from the target signal, thereby reducing the number of false targets and the probability of false alarms. As a result, the radar system has higher target detection accuracy, stronger reliability, and lower computational load, and can be applied to complex road sections such as fences, overpasses, and tunnels. Attached Figure Description

[0042] Figure 1 This is a schematic diagram of the target detection process of the radar target detection method applied to complex road surfaces provided in a specific embodiment of the present invention.

[0043] Figure 2 This is a schematic diagram showing the location of the extended region in the mirror suppression step provided in a specific embodiment of the present invention.

[0044] Figure 3 This is a schematic diagram of the window function provided in a specific embodiment of the present invention.

[0045] Figure 4 This is a schematic diagram of the sigmoid function provided in a specific embodiment of the present invention.

[0046] Figure 5 This is a comparison of the spectral lines before and after the application of spectral leakage suppression methods in this invention.

[0047] Figure 6This is a comparison of the spectral lines before and after the application of mirror suppression in this invention.

[0048] Figure 7 This is a comparison diagram of spectral lines under ideal and actual conditions in this invention. Detailed Implementation

[0049] The technical solution of the present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0050] like Figures 1 to 7 As shown, this embodiment provides a radar target detection method for complex road surfaces. To effectively suppress spectral leakage in the radar system, this method distinguishes between interference signals caused by strong reflectors such as fences, overpasses, and tunnels and the target signal, reducing the number of false targets and the probability of false alarms. This results in higher target detection accuracy, stronger reliability, and lower computational load for the radar system, making it applicable to complex road sections such as fences, overpasses, and tunnels. Furthermore, this radar target detection method includes the following steps:

[0051] (S1): The radar transmits and receives the reflected intermediate frequency signal to the target, samples the intermediate frequency signal to obtain frame data containing chirp signal, performs two FFT operations on each frame data to obtain the target's range index and velocity index.

[0052] (S2): Determine the DC leakage region on the velocity spectrum of each distance dimension, set Q spectral lines on both sides of the DC leakage region, and perform windowing processing on the DC signal within the Q spectral line interval to suppress the DC spectral line leakage component. In this embodiment, Q can be set arbitrarily by the operator and is not limited here.

[0053] (S3): Select the maximum amplitude A in the interval [2, M / 2] of the positive velocity spectrum of the distance dimension, find the amplitude B of the symmetrical point of the maximum amplitude A in the negative velocity spectrum, and calculate the ratio R of the maximum amplitude A and amplitude B of the target in each frame. The amplitude ratio of the previous frame is R1 and the amplitude ratio of the current frame is R2. By judging the amplitude ratio R1 of the previous frame and the amplitude ratio R2 of the current frame, the spectrum is mirrored and suppressed. In this embodiment, M is the number of points corresponding to the M-point FFT operation.

[0054] (S4): Find P amplitude values ​​in the distance dimension. If the amplitude value exceeds a preset threshold, then execute: record the distance index and velocity index of the target whose amplitude value exceeds the preset threshold, and calculate the difference between the real phase and the imaginary phase and the ratio of the real amplitude to the imaginary amplitude of the target respectively. If the difference between the real phase and the imaginary phase and the ratio of the real amplitude to the imaginary amplitude are both within the preset threshold range, then save the target and define it as a real target. In this embodiment, P can be arbitrarily set by the operator and is not limited here.

[0055] In this embodiment, the radar is a single-transmit, single-receive narrow-beam frequency-modulated continuous wave radar. In step (S1), each chirp signal in this embodiment contains 128 sampling points. By performing two FFTs on the chirp signals, the target's range index and velocity index are obtained. Chirp, a term in pulse coding technology, refers to a sound similar to a bird's chirp when the carrier frequency increases linearly during pulse encoding. The phenomenon of a shift in the center wavelength during pulse transmission is also called chirp. Using FFT for signal sampling processing significantly reduces system computation and costs compared to DFT processing, while ensuring signal integrity.

[0056] In circuits, many components (such as amplifiers) have DC drift at their output, which causes interference from DC components in the spectrum. During sampling, the spectrum of the target signal is truncated into a sequence of finite length. Ideally, the spectral amplitudes at the initial and final points should be close to 0. However, due to the influence of DC signals, spectral leakage will occur at the initial and final points. The region near the initial and final points is defined as the DC leakage region. The spectral lines in this DC leakage region need to be suppressed by a window function to correct the effect of DC leakage on the spectral lines.

[0057] In vehicle radar target detection, DC leakage is common knowledge familiar to those skilled in the art. To provide a more intuitive explanation of DC leakage, specifically, as follows... Figure 7 As shown, the horizontal axis represents the number of FFT points, and the vertical axis represents the amplitude. Ideally, after performing a 128-point FFT on a chirp signal, the amplitude should approach 0 at 2 and 128. In reality, due to the influence of DC leakage, the spectral lines will rise near 2 and 128, meaning that energy leaks to both sides of the DC. This DC leakage region needs to be suppressed using a window function to correct the effect of DC leakage on the spectral lines.

[0058] In step (S2), in order to prevent the spectrum leakage on both sides of the DC from being mistaken for the target, the spectrum near the DC leakage area is weighted and suppressed by a window function. The spectrum near the DC (within a certain range) is processed by the window function to avoid the original target being mistaken for the DC due to spectrum leakage. This can effectively reduce the impact of spectrum leakage caused by the DC signal, reduce the number of false targets, and lower the probability of misjudgment.

[0059] In order to further distinguish the interference signals caused by strong reflectors such as fences, overpasses, and tunnels from the target signals, in step (S3), the amplitude ratio of two adjacent frames of data is judged to analyze whether there are mirror reflection interference signals in the negative velocity spectrum of the target signal, and the existing mirror reflection interference signals are suppressed, thereby further reducing the number of false targets and further reducing the probability of false alarms.

[0060] Since the target will not present a single value on the spectrum, considering the continuity of the target in terms of distance and its own physical characteristics, the original target is screened through step (S4) to further find target signals with continuous characteristics, and the target is saved and output.

[0061] Through the above process, the radar system can effectively suppress spectrum leakage using this method. By distinguishing the interference signals caused by strong reflectors such as fences, overpasses, and tunnels from the target signals, the number of false targets and the probability of false alarms are reduced. As a result, the radar system has higher target detection accuracy, stronger reliability, and less computational load, and can be applied to complex road sections such as fences, overpasses, and tunnels.

[0062] Preferably, in step (S3), mirror suppression of the spectral lines includes:

[0063] In each frame, by using the index of the maximum amplitude A on the positive velocity spectrum, find the amplitude B of the symmetrical point of A on the negative velocity spectrum, record the index of B as Idx and keep it, where the index of the negative velocity direction in the previous frame is recorded as Idx1 and the index of the negative velocity direction in the current frame is recorded as Idx2.

[0064] Perform an OR operation between the index Idx1 of the negative velocity direction in the previous frame and the index Idx2 of the negative velocity direction in the current frame to obtain the symmetrical point region [minIdx, maxIdx] of the suppression region. The minimum value among {Idx1, Idx2} is denoted as minIdx, and the maximum value among {Idx1, Idx2} is denoted as maxIdx.

[0065] The 8 positions extending around the symmetrical point region [minIdx, maxIdx] constitute the extension region of the suppression region;

[0066] A suitable suppression function F(R) is selected for the spectral lines. The suppression function F(R) performs mirror suppression on the spectral lines in the suppression region according to the value of R.

[0067] Furthermore, the index of the negative velocity direction in the previous frame is denoted as Idx1, and the index of the negative velocity direction in the current frame is denoted as Idx2. The suppression region of the index is obtained by performing calculations on the negative velocity direction indices Idx1 and Idx2 of the previous frame.

[0068] The suppression region includes a symmetrical point region. The minimum value in {Idx1, Idx2} is denoted as minIdx, and the maximum value in {Idx1, Idx2} is denoted as maxIdx, resulting in a symmetrical point region of [minIdx, maxIdx]. As long as the velocity index is within the symmetrical point region [minIdx, maxIdx], the spectral lines within the symmetrical point region [minIdx, maxIdx] are mirror suppressed by specifying the appropriate suppression function F(R). The spectral lines within the symmetrical point region [minIdx, maxIdx] are mirror suppressed at the index of the negative velocity direction of the target in each frame.

[0069] The suppression region also includes the extension region. The symmetric point region [minIdx, maxIdx] extends to eight positions around it. By specifying the appropriate suppression function F(R), the spectral lines in the extension region are mirror suppressed. The spectral lines in the extension region are mirror suppressed at the index of the negative velocity direction of each frame of the target.

[0070] The process of selecting the suppression function is as follows: Based on the mirror target to be suppressed, select an appropriate suppression function F(R), and set the threshold range of the horizontal coordinate R of the suppression function F(R) to [R...]. min R max ], where R min R max These are the minimum and maximum values ​​on the interval, respectively.

[0071] The target amplitude ratio R is compared with the threshold value of the x-coordinate of the set F(R) and the image suppression is performed by selecting the suppression coefficient for the target.

[0072] When R < R max When the inhibition coefficient is F(R), the inhibition coefficient is F(R).

[0073] When R≥R max When the inhibition coefficient is F(R) max );

[0074] Through the above process, based on the relationship between the amplitude ratio R and the set threshold, the symmetrical point region and its extended region are suppressed simultaneously. This allows for precise weighted suppression of existing mirror reflection interference signals and suppression of adjacent extended regions, ensuring data continuity, improving suppression accuracy, thereby enhancing the recognition of target detection, reducing unnecessary target data processing, and lowering the system's computational load.

[0075] Preferably, in step (S3), mirror suppression of the spectral lines in the suppression region is performed, including:

[0076] The range of the ratio R of the maximum amplitude A to the amplitude B in each frame of the target is defined as [R]. min Rmax ];

[0077] 1) If both the amplitude ratio R1 of the previous frame and the amplitude ratio R2 of the current frame are greater than 0, then execute:

[0078] The suppression function F(R) is based on the amplitude ratio R2 of the current frame and R... max The magnitude relationship is used to perform mirror suppression of spectral lines in the suppression region;

[0079] 2) If the amplitude ratio R1 of the previous frame is not 0, and the amplitude ratio R2 of the current frame is 0, then execute:

[0080] The suppression function F(R) is based on the amplitude ratio R1 of the previous frame and R... max The magnitude relationship is used to perform mirror suppression of spectral lines in the suppression region;

[0081] 3) If the amplitude ratio R1 of the previous frame is 0 and the amplitude ratio R2 of the current frame is not 0, then mirror suppression is not performed.

[0082] Further, in step (S3), by judging the amplitude ratio R1 of the previous frame and the amplitude ratio R2 of the current frame, mirror suppression is performed on the spectral lines, including:

[0083] The suppression function selected for the spectral lines is F(R), and the range of the ratio R of the maximum amplitude A and amplitude B in each frame of the target is set to [R]. min R max ];

[0084] 1) If both the amplitude ratio R1 of the previous frame and the amplitude ratio R2 of the current frame are greater than 0, then mirror suppression is performed:

[0085] The preset maximum threshold R of the target's current frame amplitude ratio R2 and the x-coordinate R of the suppression function F(R) is determined. max Based on the relationship between the magnitudes of F(R), the inhibition coefficient Y1 is obtained by performing calculations on F(R):

[0086] If R2≥R max When the inhibition coefficient is Y1=F(R), then the inhibition coefficient is Y1=F(R). max );

[0087] If R2 < R max When the inhibition coefficient is Y1=F(R2), then the inhibition coefficient is Y1=F(R2).

[0088] By comparing the current frame rate ratio R2 with the preset maximum threshold x of the x-coordinate R of the suppression function F(R). max The magnitude relationship is used to calculate the suppression function F(R) to obtain the suppression coefficient Y1. The index of the negative velocity direction of the previous frame and the index of the negative velocity direction of the current frame are ORed to obtain the suppression region. The spectral lines in the suppression region are multiplied by the suppression coefficient Y1 to achieve mirror suppression of the spectral lines.

[0089] 2) If the amplitude ratio R1 of the previous frame is not 0 and the amplitude ratio R2 of the current frame is 0, then mirror suppression is performed:

[0090] The preset maximum threshold R is determined by comparing the amplitude ratio R1 of the target in the previous frame with the x-coordinate R of the suppression function F(R). max Based on the relationship between the magnitudes of F(R), the inhibition coefficient Y2 is obtained by performing calculations on F(R):

[0091] If R1≥R max When the inhibition coefficient is Y2=F(R), then the inhibition coefficient is Y2=F(R). max );

[0092] If R1 < R max When the inhibition coefficient is Y2=F(R1), then the inhibition coefficient is Y2=F(R1).

[0093] By comparing the amplitude ratio R1 of the previous frame with the preset maximum threshold R of the x-coordinate of the suppression function F(R), max The magnitude relationship is used to calculate the suppression function F(R) to obtain the suppression coefficient Y2. The index of the negative velocity direction of the previous frame and the index of the negative velocity direction of the current frame are ORed to obtain the suppression region. The spectral lines in the suppression region are multiplied by the suppression coefficient Y2 to achieve mirror suppression of the spectral lines.

[0094] 3) If the amplitude ratio R1 of the previous frame is 0 and the amplitude ratio R2 of the current frame is not 0, then mirror suppression is not performed.

[0095] Through the above process, the amplitude ratio between the previous frame and the current frame is analyzed, and the mirror reflection interference signal on the negative velocity spectrum is suppressed to different degrees. Depending on the different amplitude ratios between the two adjacent frames, it can be determined whether to suppress, the size of the suppression weighting coefficient, and the suppression area, thereby reducing the number of false targets and further reducing the computational load of the system.

[0096] Preferably, in step (S2), the windowing process includes:

[0097] Find spectral lines in the velocity spectrum within the interval [M-Q+1, M] that are of the same order of magnitude as those in the velocity spectrum within the interval [2, 1+Q], and find the index position of these spectral lines. Suppress spectral lines of the same order of magnitude in both the interval [M-Q+1, M] and [2, 1+Q] using a window function.

[0098] The suppression method is as follows: the degree of suppression of the target signal of the spectral line increases as the distance between the spectral line and the DC leakage region decreases, that is, the closer the spectral line is to the DC, the more it is suppressed, and the farther the spectral line is from the DC, the less it is suppressed.

[0099] Through the above process, the DC component spectral leakage in the DC leakage region is suppressed. Specifically, spectral lines of the same order of magnitude are identified and weighted using the same value of the window function to ensure the uniformity of data processing for spectral lines of the same order of magnitude and reduce data processing errors.

[0100] Preferably, in step (S2), the window function selected for the windowing process is a (1-Gaussian window) with 2Q Gaussian window points. The Gaussian window is an exponential window, often used to truncate non-periodic signals, such as exponentially decaying signals. The application of exponential windows forces the response signal to better meet the periodicity requirements of the FFT transform. Therefore, using a (1-Gaussian window) can effectively suppress the DC component interference signal in the DC leakage region.

[0101] Preferably, in step (S4), calculating the difference between the real and imaginary phases of the target, and the ratio of the real to imaginary amplitudes, includes:

[0102] Based on the target's distance index and velocity index, find the FFT value of the target at that point. After performing FFT on the real part and the imaginary part of the target signal respectively, calculate the difference between the phase of the real part and the phase of the imaginary part of the target signal, and the ratio of the amplitude of the real part to the amplitude of the imaginary part.

[0103] By analyzing the difference between the real and imaginary phases of the target and the ratio of the real and imaginary amplitudes, we can further determine whether the original target conforms to the physical characteristics of the detected object, effectively screen the original target, ensure target relevance, and thus reduce the number of false targets.

[0104] Preferably, in step (S4), the amplitudes of adjacent distance dimensions at the same speed are added together to obtain the amplitude of the real target. The target's amplitude is saved as the sum of the amplitudes of adjacent distance dimensions at the same speed. Since the target will not present a single value in the spectrum, considering the continuity in distance, the amplitudes of adjacent distance dimensions at the same speed are added together to ensure the continuity and accuracy of data processing.

[0105] Preferably, in step (S4), the total number of targets in each frame is no more than K. Based on the accuracy of the radar targets, a maximum total number of targets to be identified per frame is set to further reduce the system's computational load.

[0106] Preferably, in step (S1), sampling the intermediate frequency signal includes:

[0107] The radar emits electromagnetic waves and simultaneously receives the echo signals from the reflected targets. The intermediate frequency signal is obtained by processing the frequency difference between the echo signals and the local oscillator signals.

[0108] After sampling the intermediate frequency signal, two FFT operations are performed, including:

[0109] 1) Perform an M-point FFT operation on each chirp signal to obtain the target's distance index;

[0110] 2) Perform an N-point FFT operation on each distance dimension to obtain the target's velocity index.

[0111] Through the above process, two FFTs are performed on the intermediate frequency signal: first, an FFT is performed on the M sampling points of each chirp signal to obtain the distance dimension information; then, an FFT is performed on the N chirp signals in each distance dimension to obtain the velocity dimension information. In this embodiment, each frame of data contains 128 chirp signals. By performing a 128-point FFT on each chirp signal, the distance index of each target is obtained. Then, FFTs are performed on different frames in the same distance dimension, that is, a 128-point FFT is performed on each distance dimension to obtain the velocity index of each target. The 128 points of the obtained distance can be modified according to the required distance resolution; similarly, the 128 points of the obtained velocity can be modified and adjusted according to the required velocity resolution. Using FFT (Fast Fourier Transform) for signal sampling processing can greatly save system computation and reduce costs while ensuring that the signal is not distorted.

[0112] Preferably, in step (S2), the method for determining spectral lines of the same order of magnitude includes:

[0113] Calculate the ratio of the amplitude of spectral line one of the velocity spectrum in the interval [M-Q+1, M] to the amplitude of spectral line two of the velocity spectrum in the interval [2, 1+Q].

[0114] If the ratio of the amplitude of spectral line 1 to the amplitude of spectral line 2 is within the preset threshold range of the amplitude ratio of spectral lines, then spectral line 1 and spectral line 2 are considered to be spectral lines of the same order of magnitude.

[0115] The amplitudes of spectral line 1 in the velocity spectrum interval [M-Q+1, M] and spectral line 2 in the interval [2, 1+Q] are compared. If the amplitude ratio of spectral line 1 and spectral line 2 meets the set threshold range, then spectral line 1 and spectral line 2 are considered to be of the same order of magnitude. Spectral lines of the same order of magnitude are identified, and uniform weighted suppression is performed using a window function to ensure consistency in the processing of spectral lines of the same order of magnitude and reduce data processing errors.

[0116] Specifically, in this embodiment, the target search and output can be achieved by setting the following parameters.

[0117] 1. Intermediate frequency signal sampling:

[0118] M = 128, meaning each chirp signal contains 128 sampling points.

[0119] 2. Perform FFT operations on the sampled data:

[0120] 1) M = 128, perform a 128-point FFT on each chirp to obtain the distance index of each target;

[0121] 2) N = 128, perform a 128-point FFT on each distance dimension to obtain the velocity index of each target.

[0122] 3. Spectrum leakage suppression:

[0123] 1) Determine the value of Q: Q = 8;

[0124] 2) Confirm the window function. The window function is defined as a (1-Gaussian window). The Gaussian window has 2Q points and 16 points. The amplitude decreases closer to the symmetry point. Figure 3 As shown;

[0125] 3) Find the spectral line in the interval [121, 128] that is of the same order of magnitude as the interval [2, 9], and find its index position. Suppress the entire interval using a window function. The specific operation is as follows:

[0126] First, start searching from point 128 of the spectrum and compare it with the amplitude of the spectrum in the interval [2, 9]. If point 128 does not match, then compare the amplitude of point 127 with the amplitude of the spectrum in the interval [2, 9]. The condition for matching the order of magnitude is: if the amplitude ratio matches the interval [0.5, 1.5], it is considered to be of the same order of magnitude.

[0127] Record the positions of spectral lines of the same order of magnitude. Here, we take spectral lines 128 and 3 (spectral line 3 represents the second spectral line excluding DC) as an example.

[0128] Suppression method: spectral line 128 multiplied by the 7th point amplitude of the window function, spectral line 127 multiplied by the 6th point amplitude of the window function, spectral line 126 multiplied by the 5th point amplitude of the window function, and so on;

[0129] In vehicle radar target detection, DC leakage is common knowledge familiar to those skilled in the art. To provide a more intuitive explanation of DC leakage, specifically, as follows... Figure 7 As shown, the horizontal axis represents the number of FFT points, and the vertical axis represents the amplitude. Ideally, after performing a 128-point FFT on a chirp signal, the amplitude should approach 0 at 2 and 128. In reality, due to the influence of DC leakage, the spectral lines will rise near 2 and 128, meaning that energy leaks to both sides of the DC. This DC leakage region needs to be suppressed using a window function to correct the effect of DC leakage on the spectral lines.

[0130] Figure 5 This is a comparison of spectral lines before and after the application of spectral leakage suppression methods in this invention. Figure 5In the figure, the horizontal axis represents the number of FFT points, the vertical axis represents the amplitude, curve 1 represents the original spectrum without the spectrum leakage suppression method of the present invention, and curve 2 represents the spectrum after the spectrum leakage suppression method of the present invention.

[0131] When detecting target information via radar, the radar first approaches the target and then moves away. After obtaining the radar's sampled data, the target signal is processed using the spectral leakage suppression method described in this embodiment to obtain curve 2. Figure 5 It can be seen that the amplitude fluctuation of curve 2 is more gradual compared to curve 1. This shows that the target signal obtained by the data processing method of radar target detection with spectrum leakage suppression in this embodiment is far superior to the target signal of radar target detection method in the prior art that does not use spectrum leakage suppression. It effectively reduces the impact of spectrum leakage caused by DC signal, reduces the number of false targets, and lowers the probability of misjudgment.

[0132] 4. Mirror suppression:

[0133] Select amplitude A within the velocity spectrum [2, 64] and find the amplitude B at the symmetrical point. Calculate the amplitude ratio R = A / B.

[0134] Considering the amplitude ratio between two adjacent frames, the suppression function F(R) is chosen as the sigmoid function (e.g., ...). Figure 4 As shown in the figure, the horizontal axis R is [1, 20]. Suppression is performed based on the value of R: if R < 20, the suppression coefficient is sigmoid(R); if R > 20, the suppression coefficient is sigmoid(20). The extended region is suppressed according to the above suppression method.

[0135] Figure 6 This is a comparison of spectral lines before and after the application of mirror suppression in this invention. Figure 6 In the figure, the horizontal axis represents the number of FFT points, the vertical axis represents the amplitude, curve 1 represents the original spectrum without the image suppression method of the present invention, and curve 2 represents the spectrum after the image suppression method of the present invention.

[0136] The image suppression method described in this embodiment is used to process the target signal, resulting in curve 2. Figure 6 It can be seen that the amplitude fluctuation of curve 2 is more gradual compared to curve 1. This demonstrates that the target signal obtained by the radar target detection method with image suppression in this embodiment is far superior to the target signal obtained by existing radar target detection methods that do not employ image suppression. By precisely weighting and suppressing the existing image reflection interference signal, the suppression accuracy is improved, thereby enhancing the target detection recognition rate, reducing unnecessary false target data processing, and lowering the system's computational load.

[0137] 5. Target search:

[0138] 1) Find 4 amplitude values ​​in each distance dimension, set the target threshold to 600, and record the velocity index and distance index of the point. The total number of targets in each frame is no more than 20.

[0139] 2) Calculate the difference between the real and imaginary phases of the target and the ratio of the real and imaginary amplitudes. The specific process is as follows: Based on the velocity index and distance index of the target, find the FFT value of the point (i.e. the value in step (1) of step 2), take the real and imaginary parts respectively to perform FFT, and calculate the difference between the real and imaginary phases and the ratio of the real and imaginary amplitudes. The phase difference range is [-95, -90], and the amplitude ratio is set to [0.8, 1.3]. Targets that meet these conditions are approaching, rather than false targets caused by the fence. The target saving method is: the amplitude of the target that meets the conditions is the sum of the amplitudes of the adjacent distance dimensions at the same velocity.

[0140] This invention has been described through preferred embodiments. Those skilled in the art will understand that various changes or equivalent substitutions can be made to these features and embodiments without departing from the spirit and scope of the invention. This invention is not limited to the specific embodiments disclosed herein; other embodiments falling within the scope of the claims are also within the protection scope of this invention.

[0141] It should be noted that any reference signs placed between parentheses in the claims should not be construed as limiting the claims. The word "comprising" does not exclude the presence of technical features or steps not listed in the claims. The word "a" or "an" preceding a technical feature does not exclude the presence of a plurality of such technical features. The invention can be implemented by means of hardware comprising several different technical features and by means of a suitably programmed computer. In a unit claim enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words can be interpreted as names.

[0142] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0143] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0144] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0145] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

Claims

1. A radar target detection method applied to complex road surfaces, characterized in that, It includes the following steps: The radar transmits and receives the reflected intermediate frequency signal to the target, samples the intermediate frequency signal to obtain frame data containing chirp signals, and performs two FFT operations on each frame data to obtain the range index and velocity index of the target. A DC leakage region is determined on the velocity spectrum of each distance dimension. Q spectral lines are set on both sides of the DC leakage region. The DC signal within the Q spectral line interval is windowed to suppress the DC spectral line leakage component. The maximum amplitude A is selected within the interval [2, M / 2] on the positive velocity spectrum of the distance dimension. The amplitude B of the symmetrical point of the maximum amplitude A on the negative velocity spectrum is found. The ratio R of the maximum amplitude A and amplitude B of each frame of data of the target is calculated, where the amplitude ratio of the previous frame is R1 and the amplitude ratio of the current frame is R2. By judging the amplitude ratio R1 of the previous frame and the amplitude ratio R2 of the current frame, image suppression is performed on the spectral lines. M is the number of points corresponding to the M-point FFT operation on the chirp signal. The image suppression of the spectral lines includes: In each frame of data, the index of the maximum amplitude A on the positive velocity spectrum is used to find the symmetrical point amplitude B of A on the negative velocity spectrum. The index of B is recorded as Idx and retained. The index of the negative velocity direction in the previous frame is recorded as Idx1 and the index of the negative velocity direction in the current frame is recorded as Idx2. Perform an OR operation between the index Idx1 of the negative velocity direction in the previous frame and the index Idx2 of the negative velocity direction in the current frame to obtain the symmetrical point region [minIdx, maxIdx] of the suppression region. The minimum value among {Idx1, Idx2} is denoted as minIdx, and the maximum value among {Idx1, Idx2} is denoted as maxIdx. The 8 positions extending outwards from the symmetrical point region [minIdx, maxIdx] constitute the extension region of the suppression region. A suitable suppression function F(R) is selected for the spectral lines. The suppression function F(R) performs mirror suppression on the spectral lines in the suppression region based on the value of R. The mirror suppression on the spectral lines in the suppression region includes: The range of the ratio R of the maximum amplitude A and amplitude B of each frame of data is defined as [R]. min R max ], where R max The preset maximum threshold of the abscissa R of the suppression function F(R); 1) If both the amplitude ratio R1 of the previous frame and the amplitude ratio R2 of the current frame are greater than 0, then execute: The suppression function F(R) is based on the amplitude ratio R2 of the current frame and R... max The magnitude relationship is used to perform mirror suppression of spectral lines in the suppression region; 2) If the amplitude ratio R1 of the previous frame is not 0, and the amplitude ratio R2 of the current frame is 0, then execute: The suppression function F(R) is based on the amplitude ratio R1 of the previous frame and R... max The magnitude relationship is used to perform mirror suppression of spectral lines in the suppression region; 3) If the amplitude ratio R1 of the previous frame is 0 and the amplitude ratio R2 of the current frame is not 0, then image suppression is not performed; Find P amplitude values ​​in the distance dimension. If the amplitude value exceeds a preset threshold, then perform the following: record the distance index and velocity index of the target whose amplitude value exceeds the preset threshold, calculate the difference between the real and imaginary phases and the ratio of the real and imaginary amplitudes of the target respectively. If the difference between the real and imaginary phases and the ratio of the real and imaginary amplitudes are both within the preset threshold range, then save the target and define it as a real target.

2. The radar target detection method for complex road surfaces according to claim 1, characterized in that: The windowing process includes: Find spectral lines in the velocity spectrum within the interval [M-Q+1, M] that are of the same order of magnitude as those in the velocity spectrum within the interval [2, 1+Q], and find the index position of these spectral lines. Suppress spectral lines of the same order of magnitude in both the interval [M-Q+1, M] and [2, 1+Q] using a window function. The degree of suppression of the target signal of the spectral line increases as the distance between the spectral line and the DC leakage region decreases.

3. The radar target detection method for complex road surfaces according to claim 1 or 2, characterized in that: The windowing process uses a 1-Gaussian window with 2Q Gaussian window points.

4. The radar target detection method for complex road surfaces according to claim 1, characterized in that: The calculation of the difference between the real and imaginary phases of the target, and the ratio of the real to imaginary amplitudes, includes: Based on the target's distance index and velocity index, find the FFT value of the target at that point. After performing FFT on the real part and the imaginary part of the target signal respectively, calculate the difference between the phase of the real part and the phase of the imaginary part of the target signal, and the ratio of the amplitude of the real part to the amplitude of the imaginary part.

5. The radar target detection method for complex road surfaces according to claim 1 or 4, characterized in that: The amplitudes of adjacent distance dimensions at the same velocity are added together to obtain the amplitude of the real target.

6. The radar target detection method for complex road surfaces according to claim 1 or 4, characterized in that: The total number of data targets in each frame is no more than K.

7. The radar target detection method for complex road surfaces according to claim 1, characterized in that: Sampling the intermediate frequency signal includes: The radar emits electromagnetic waves and simultaneously receives the echo signal from the reflected target. The intermediate frequency signal is obtained by performing frequency difference processing on the echo signal and the local oscillator signal. The intermediate frequency signal is sampled and subjected to two FFT operations, including: 1) Perform an M-point FFT operation on each chirp signal to obtain the distance index of the target; 2) Perform an N-point FFT operation on each of the distance dimensions to obtain the velocity index of the target.

8. The radar target detection method for complex road surfaces according to claim 2, characterized in that: The methods for determining spectral lines of the same order of magnitude include: Calculate the ratio of the amplitude of spectral line one of the velocity spectrum in the interval [M-Q+1, M] to the amplitude of spectral line two of the velocity spectrum in the interval [2, 1+Q]. If the ratio of the amplitude of spectral line 1 to the amplitude of spectral line 2 is within the preset threshold range of the amplitude ratio of spectral lines, then spectral line 1 and spectral line 2 are considered to be spectral lines of the same order of magnitude.

Images