Multi-path ghosting target elimination method and device for radar target imaging

Abstract

The invention discloses a multipath ghosting target elimination method for radar target imaging. The method comprises the following steps that S1, a radar emits electromagnetic waves and processes echoes to obtain output signals; s2, acquiring the total number of detected targets and distance information and Doppler information of each target; s3, drawing a distance Doppler image, and carrying out binarization to obtain a binarized image; s4, carrying out target collinear detection, and judging each detected collinear group; s5, sorting the multiple groups of collinear groups obtained in the step S4; s6, carrying out the judgment of a multipath ghosting target; and S7, eliminating the multipath ghosting target from the target set T. The invention also discloses a multipath ghosting target elimination device for radar target imaging. According to the method, the first-order ghosting and the second-order ghosting can be accurately searched and eliminated at the same time, the coping capacity of the radar for different-order multipath ghosting targets is enhanced, and the accuracy and the reliability of the radar for real target imaging are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of radar signal processing, and particularly relates to a method and device for eliminating multipath ghost targets for radar target imaging. Background Art

[0002] With the rapid development of intelligent transportation systems (ITS), radar technology is increasingly widely used in traffic management, such as traffic flow monitoring, vehicle speed detection, accident warning, etc. However, in practical applications, radar signals are easily affected by multipath effects, resulting in the generation of false targets, namely "ghost" targets. Ghost targets seriously affect the accuracy and reliability of radar target imaging, interfere with the judgment of traffic management systems, reduce traffic management efficiency, and even pose potential safety hazards.

[0003] Multipath effect refers to the phenomenon that after the electromagnetic waves emitted by the radar propagate through different paths, the time delays of each component field reaching the receiving end are different, and they are superimposed on each other according to their respective phases, causing interference and distorting the original signal or generating errors. When the radar performs target imaging by transmitting and receiving electromagnetic waves, due to the inevitable presence of reflectors such as building walls, smooth road signs, and metal guardrails in the real environment, the propagation path of electromagnetic waves is diverse, which will cause the radar to output a large number of ghost targets. These ghost targets are caused by multipath propagation, and they will greatly reduce the accuracy of the radar imaging result and affect the reliability of radar target imaging in practical applications.

[0004] Due to the existence of multipath propagation, multiple useless results will appear in the output result of detecting a real target signal. If these useless results are not screened and eliminated, the radar will detect strong target signals where there should be no targets originally, that is, multipath ghost targets will appear. Usually, the number of times the signal is reflected by the reflector is the order of the corresponding multipath ghost target. In addition, due to the large attenuation caused by signal reflection, generally, signals reflected by the reflector once or twice will cause multipath ghost targets to appear.

[0005] To address the adverse effects brought about by the multipath propagation of radar signals, early research and technical solutions mainly focused on actively reducing the impact of multipath effects by changing the radar transmit waveform or adjusting the modulation method and parameters. For example, by designing the pulse waveform to reduce the mutual interference between the signal and the multipath echo, or by adjusting parameters such as waveform frequency and period, making it difficult for multipath reflections to be detected by the receiving end. In recent years, with the development of radar technology and the diversification of application scenarios, the demand for accurate target imaging and positioning has been continuously increasing, and traditional multipath suppression methods have been unable to effectively cope. In some complex scenarios, ghost targets, false echoes, and data noise caused by multipath effects have become problems that cannot be ignored. Therefore, how to effectively eliminate the false targets caused by multipath has become a research hotspot. Currently, sidelobe cancellation is usually adopted, but this multipath interference cancellation method must additionally set several auxiliary antennas, increasing the cost of hardware settings.

[0006] The Chinese patent with the publication number CN108318864B proposes a method for eliminating radar multipath ghost targets by comparing distance and angle information, but this method mainly aims at eliminating first-order multipath ghost targets. It should be noted that the existence of second-order ghost targets still seriously affects the accuracy of the imaging results. Summary of the Invention

[0007] In view of the defects existing in the above-mentioned prior art, the present invention discloses a method and device for eliminating multipath ghost targets for radar target imaging. This method can simultaneously eliminate first-order and second-order multipath ghost targets, and has the characteristics of fast and efficient, simple implementation, and low cost.

[0008] The method for eliminating multipath ghost targets for radar target imaging according to the present invention includes the following steps:

[0009] S1. After the electromagnetic wave emitted by the radar is reflected by the target, the received echo signal is processed by the radar system to obtain the output signal;

[0010] S2. Complete target detection based on the radar output signal, obtain the total number of detected targets and the distance information and Doppler information of each target, and store the target information of the target with the index number i into the data structure target i =(range i , doppler i ); where target i represents the target information of the i-th target, where range i represents the distance information of the i-th target, and where doppler i represents the Doppler information of the i-th target.

[0011] S3. Store the target information target corresponding to all targets i in the target set T, draw a range-Doppler image based on all the data in the target set T, and then binarize the range-Doppler image to obtain a binarized image;

[0012] S4. Perform target collinearity detection on the binarized image obtained in step S3, and judge each detected collinear group. If the number of targets in the group is not less than 3, record the number of targets corresponding to the group and the target information of each target;

[0013] S5. For the multiple groups of collinear groups obtained in step S4, sort the groups in ascending order according to the number of targets m j in each group, sort the targets in ascending order according to the distance information range i of each target; where j is the index number of the collinear group;

[0014] S6. For the multiple groups of collinear groups recorded in step S5, traverse and mark 1 group each time in ascending order of the index number j

[0015] Each time during traversal, select 3 targets in the group to judge multi-path ghost targets, and mark the multi-path ghost targets according to the judgment results;

[0016] The judgment method is specifically as follows:

[0017] Suppose the 3 selected targets are a, b, c, and in ascending order of distance information are target a 、target b 、target c ; The Doppler information of the three targets a, b, c are respectively, doppler a 、doppler b 、doppler c ; If

[0018] |(range a -range b )-(range b -range c )|<r

[0019] |(doppler a -doppler b )-(doppler b -doppler c )|<d

[0020] Both formulas hold, then target b and targetc Marked as multipath ghost targets, where r and d are preset target distance threshold and target Doppler threshold;

[0021] Otherwise, it is considered that no ghost target can be found in this judgment;

[0022] Traverse all possible combinations of 3 targets within the group until all combinations are traversed. The traversal mark of this group ends, and the next collinear grouping is traversed. When all collinear groupings are traversed, go to step S7;

[0023] S7. Eliminate all the marked multipath ghost targets from the target set T in step S6, and output the result as the real target information.

[0024] Preferably, in step S2, the specific method for target detection of the radar output signal is: perform FFT, constant false alarm detection algorithm and peak focusing on the radar output signal in sequence to obtain the total number of detected targets and the distance information and Doppler information of each target.

[0025] Preferably, step S3 is specifically: store the information target of all targets detected in each frame i into the target set T, and draw a range-Doppler image with the data of this set, and then perform binarization. Its horizontal and vertical axes are distance information and Doppler information respectively. The cells with targets are assigned a value of 1, and other cells are all assigned a value of 0.

[0026] Preferably, in step S4, the binarized image obtained in step S3 is used for collinearity detection by using the Hough transform.

[0027] A multipath ghost target elimination device for radar target imaging, comprising:

[0028] A target information acquisition unit, which is used to perform target detection on the radar output signal, obtain the range-Doppler information of all targets, and draw a binarized range-Doppler image. The output of the target information acquisition unit is used as the input of the collinear target grouping unit;

[0029] A collinear target grouping unit, which is used to perform collinearity detection on the target distribution on the binarized image, and record and sort the collinear groupings. The output of the collinear target grouping unit is used as the input of the multipath ghost target elimination unit;

[0030] A multipath ghost target elimination unit, which is used to traverse all collinear groupings, and traverse the targets within each grouping to judge and mark the multipath ghost targets. After traversal, eliminate the marked multipath ghost targets and output the real target information.

[0031] Preferably, the target information acquisition unit includes:

[0032] An FFT unit that performs fast Fourier transforms in the range dimension and the Doppler dimension on the input of the target information acquisition unit. The output of the FFT unit serves as the input to the constant false alarm detection unit.

[0033] A constant false alarm detection unit that performs constant false alarm target detection on the FFT transform result output by the FFT unit. The output of the constant false alarm detection unit serves as the input to the peak focusing unit.

[0034] A peak focusing unit that, for the constant false alarm target detection result output by the constant false alarm detection unit, retains the cell with the highest signal-to-noise ratio at each peak and sets the remaining cells to zero. The output of the peak focusing unit serves as the input to the binarization unit.

[0035] A binarization unit that assigns a value of 1 to the cells with targets in the peak focusing result and a value of 0 to the remaining cells, and outputs a binarized image. The output of the binarization unit serves as the input to the collinear target grouping unit.

[0036] Preferably, the collinear target grouping unit includes:

[0037] A Hough transform unit that performs a Hough transform on the input of the collinear target grouping unit. The output of the Hough transform unit serves as the input to the statistical sorting unit.

[0038] A statistical sorting unit that statistically counts the number of collinear groupings and the target information corresponding to each grouping according to the Hough transform result. Then, sorting is performed separately among and within each grouping. The output of the statistical sorting unit serves as the input to the multipath ghost target elimination unit.

[0039] Preferably, the multipath ghost target elimination unit includes:

[0040] A target traversal unit that, for all collinear groupings, selects one group in order each time. Each time, three targets in each group are selected and input to the calculation and marking unit until all targets in all groups are fully traversed.

[0041] The target traversal unit is respectively connected to the calculation and marking unit and the ghost elimination unit.

[0042] A calculation and marking unit that, for the three input targets, determines the multipath ghost targets and marks the multipath ghost targets.

[0043] A ghost elimination unit that, when the target traversal unit has finished traversing, deletes the targets marked as multipath ghosts and then outputs the final real target information.

[0044] The present invention can accurately search for and eliminate first-order ghost images and second-order ghost images, which not only enhances the radar's ability to handle multi-path ghost target of different orders, but also effectively improves the accuracy and reliability of the radar's imaging of real targets in complex environments. The elimination method of the present invention only starts from two basic physical information, namely distance and Doppler, and has the characteristics of simple implementation and fast recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 is a schematic diagram of the implementation steps of a specific embodiment of the multi-path ghost target elimination method of the present invention;

[0046] Figure 2 is a schematic diagram of a typical system structure of the radar involved in the present invention;

[0047] Figure 3 is a schematic diagram of the virtual array of the radar antenna in a specific embodiment of the present invention;

[0048] Figure 4 is a schematic diagram of a specific embodiment of the multi-path ghost target elimination device of the present invention;

[0049] Figure 5 is a schematic diagram of the main propagation paths of real targets and ghost targets in a specific embodiment of the present invention;

[0050] Figure 6 is a schematic diagram of the spatial distribution of the radar, reflector, real target, and ghost target in a specific embodiment of the present invention;

[0051] Figure 7 is a range-Doppler image obtained in a specific embodiment of the present invention;

[0052] Figure 8 is the result image after Hough transform in a specific embodiment of the present invention;

[0053] Figure 9 is a schematic diagram of the single-frame simulation result obtained in a specific embodiment of the present invention before eliminating the ghost target;

[0054] Figure 10 is Figure 9 a schematic diagram of the single-frame simulation result obtained after eliminating the ghost target;

[0055] Figure 11 is a schematic diagram of the multi-frame simulation result obtained in a specific embodiment of the present invention before eliminating the ghost target;

[0056] Figure 12 is Figure 11 a schematic diagram of the multi-frame simulation result obtained after eliminating the ghost target. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0057] To make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with the specific embodiments of the present invention and the corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0058] The present invention will be further described below in conjunction with the accompanying drawings of the specification and specific preferred embodiments, but this does not limit the scope of protection of the present invention.

[0059] As Figure 1 shown, a specific embodiment of the multi-path ghost target elimination method of the present invention is given, and the steps include:

[0060] S1. The radar transmitting antenna array transmits electromagnetic waves. After being reflected by the target, the received echo signal is received by the receiving antenna array, and the output signal is obtained through the processing of the radar system.

[0061] S2. Complete target detection according to the radar output signal, obtain the total number of detected targets and the distance information and Doppler information of each target, and store the target information of the target with index number i into the data structure target i =(range i , doppler i ); where target i represents the target information of the i-th target, where range i represents the distance information of the i-th target, and where doppler i represents the Doppler information of the i-th target.

[0062] S3. Store the target information target i corresponding to all targets into the target set T, draw a range-Doppler image based on all the data in the target set T, and then binarize the range-Doppler image to obtain a binarized image.

[0063] S4. Perform target collinearity detection on the binarized image obtained in step S3, and judge each detected collinear group. If the number of targets in the group is not less than 3, record the number of targets corresponding to the group and the information of each target.

[0064] S5. Step S4 obtains multiple groups of collinear groups, and each group contains m j targets, where j is the index number of the collinear group;

[0065] According to the number of targets mj Sort the groups from small to large, and then sort the targets within each group according to the distance information range of each target i in ascending order;

[0066] S6. For the multiple collinear groups recorded in step S5, select 1 group each time in ascending order of the index number j for traversal and marking,

[0067] Each time during traversal, select 3 targets in the group each time to judge the multipath ghost targets, and mark the multipath ghost targets according to the judgment results;

[0068] Traverse all possible combinations of 3 targets within the group until all combinations are traversed, the traversal marking of this group ends, traverse the next collinear group, and when all collinear groups are traversed, enter step S7;

[0069] S7. Eliminate all the multipath ghost targets marked in step S6 from the target set T, and output the result as the real target information.

[0070] When receiving and processing the radar signal in step S1, the system structure of a typical implementation of the prior art is as Figure 2 shown, including a transmitting antenna array, a receiving antenna array, a signal source, a power divider, an amplifier, a 90° phase shifter, a mixer, a filter, AD sampling, and a digital signal processor.

[0071] The power divider receives the signal transmitted by the signal source. The output end of the power divider is connected to the input ends of the first power amplifier and the 90° phase shifter, and is also connected to the first input end of the first mixer. The output end of the power amplifier is connected to the transmitting antenna array, and the output end of the 90° phase shifter is connected to the first input end of the second mixer; the receiving antenna array is connected to the input end of the second power amplifier, and the output end of the second power amplifier is also connected to the second input ends of the first mixer and the second mixer;

[0072] Each of the first mixer and the second mixer is connected to a sampling branch, as Figure 2 shown, and the sampling branch includes a filter, a signal amplifier, and an AD sampling circuit connected in sequence;

[0073] The two sampling branches respectively output orthogonal I and Q signals through the AD sampling circuits, that is, the in-phase intermediate frequency signal I and the quadrature intermediate frequency signal Q together form a complex analytic digital baseband signal. Finally, the complex analytic digital baseband signal is used as the input of the digital signal processor for ghost target elimination.

[0074] In this embodiment, the number of transmitting antennas of the radar is 1, and the number of receiving antennas is 8. Orthogonal waveforms are transmitted at the transmitting end. As Figure 3As shown, where the radar transmitting antenna is Tx1, and the receiving antennas are Rx1 to Rx8.

[0075] In a specific embodiment, the method for target detection of the radar output signal in step S2 is specifically as follows: the radar output signal is successively subjected to FFT (Fast Fourier Transform), a constant false alarm detection algorithm, and peak focusing to complete target detection, obtaining the distance information and Doppler information of all targets, where all targets include real targets and multipath ghost targets.

[0076] In a specific embodiment, the target collinearity detection in step S4 is specifically as follows: perform a Hough transform on the binary image. The number of peaks in the transformed image is the number of collinear groups. The size of each peak is an integer and equal to the number of targets included in the collinear group. Based on the peak coordinates and the Hough transform formula, calculate the distance-Doppler information of the targets included in the collinear group.

[0077] In a specific embodiment, the method for judging multipath ghost targets in step S6 is specifically as follows:

[0078] Let the three selected targets be a, b, and c, and in ascending order of distance information be target a , target b , target c ; the Doppler information of the three targets are respectively, doppler a , doppler b , doppler c ;

[0079] If

[0080] |(range a - range b ) - (range b - range c )| < r

[0081] |(doppler a - doppler b ) - (doppler b - doppler c )| < d

[0082] both hold, then mark target b and target c as multipath ghost targets, where r and d are preset target distance thresholds and target Doppler thresholds;

[0083] Since the lowest range information and Doppler information can only be real target information, and targets above the real target are ghost targets. Theoretically, the differences in range information and Doppler information of three multipath ghost targets are equal or very small. When a small threshold is set for the difference between the differences of two adjacent ghost targets, it indicates that the differences between two adjacent ghost targets are approximately equal, which conforms to the characteristics of multipath ghost targets, and the ghost targets can be identified and eliminated.

[0084] A multipath ghost target elimination device for implementing the above multipath ghost target elimination method, as Figure 4 shown, includes:

[0085] A target information acquisition unit, which is used to perform steps S2 and S3, detect targets on the radar output signal, acquire the range-Doppler information of all targets, and draw a binarized range-Doppler image. The output of the target information acquisition unit is used as the input of the collinear target grouping unit;

[0086] A collinear target grouping unit, which is used to perform steps S4 and S5, detect collinearity on the target distribution of the binarized image, and record and sort the collinear groups. The output of the collinear target grouping unit is used as the input of the multipath ghost target elimination unit;

[0087] A multipath ghost target elimination unit, which is used to perform steps S6 and S7, traverse all collinear groups, and traverse the targets within each group to judge and mark multipath ghost targets. After traversal, the marked multipath ghost targets are eliminated to obtain the output of real target information;

[0088] Specifically, the target information acquisition unit includes:

[0089] An FFT unit, which performs fast Fourier transform (FFT) on the input of the target information acquisition unit in the range dimension and Doppler dimension. The output of the FFT unit is used as the input of the constant false alarm detection unit;

[0090] A constant false alarm detection unit, which performs constant false alarm target detection on the FFT transform result output by the FFT unit. Its output is used as the input of the peak focusing unit;

[0091] A peak focusing unit, for the constant false alarm target detection result output by the constant false alarm detection unit, retains the cell with the highest signal-to-noise ratio at each peak, and sets the remaining cells to zero. Its output is used as the input of the binarization unit;

[0092] A binarization unit, which assigns a value of 1 to the cells with targets in the peak focusing result and a value of 0 to the remaining cells, and outputs a binarized image. Its output is used as the input of the collinear target grouping unit;

[0093] The collinear target grouping unit includes:

[0094] A Hough transform unit performs a Hough transform on the input of the collinear target grouping unit, and its output serves as the input of the statistical sorting unit;

[0095] A statistical sorting unit statistically sorts the number of collinear groupings and the target information corresponding to each grouping according to the Hough transform result. Then, sorting is performed separately among and within each grouping. Its output serves as the input of the multipath ghost target elimination unit;

[0096] The multipath ghost target elimination unit includes:

[0097] A target traversal unit sequentially selects one group from all collinear groups each time. Each time, three targets in each group are selected and input into the calculation and marking unit until all targets in all groups are fully traversed. The target traversal unit is respectively connected to the calculation and marking unit and the ghost elimination unit;

[0098] A calculation and marking unit determines multipath ghost targets for the three input targets and marks the multipath ghost targets;

[0099] A ghost elimination unit deletes the targets marked as multipath ghosts after the target traversal unit finishes traversing, and then outputs the final real target information;

[0100] The above multipath ghost target elimination device is implemented through modular programming of a computer, but the embodiments are not limited thereto. The specific implementation methods of the inventive device described herein include, but are not limited to, general-purpose computers, field-programmable gate arrays (FPGAs), digital signal processors (DSPs), application-specific integrated circuits (ASICs), or other programmable logic devices. In addition, the implementation of the device described herein can be completed by software, hardware, or a combination of software and hardware. The inventive device corresponds to the aforementioned method for eliminating multipath ghost targets and has the same principle.

[0101] The following further explains and describes the principle of the multipath ghost target elimination method of the present invention based on a single moving target. First, the specific sources of multipath ghost targets are analyzed. As Figure 5 shown, when there is a reflector near a single moving target T 11 there are mainly four possible propagation paths for the radar emission signal, Figure 5 where the four parts (a), (b), (c), and (d) in

[0102] represent four paths: (a) radar - target - radar; (b) radar - target - reflector - radar; (c) radar - reflector - target - radar; (d) radar - reflector - radar - reflector - radar. When the real target moves, in addition to the real target itself, considering only the first-order and second-order cases, at most 3 multipath ghost targets will also appear, asFigure 6 As shown in the figure, the radar is represented by O. The direction of the radar beam directly to the real target is the direct view direction; part of the radar beam is reflected by the reflector at the reflection point P and then reaches the real target. The direction from the radar to the reflection point P is the mirror image direction;

[0103] Figure 6 In, the position of the ghost target is the possible virtual position. From Figure 6 It can be seen that the position of the ghost target is different from that of the real target, and the positions of each ghost target are also different.

[0104] The characteristics of each target are as follows:

[0105] (1) Real target T 11 Origin: Radar - Target - Radar;

[0106] Propagation path: O → T 11 → O;

[0107] Measured distance: 2|OT 11 |;

[0108] Echo direction: Direct view direction;

[0109] Measured speed: Depends on the real target angle and the direction of the moving speed.

[0110] (2) First - order multipath ghost target T 12 Origin: Radar - Target - Reflector - Radar;

[0111] Propagation path: O → T 11 → P → O;

[0112] Measured distance: |OT 11 | + |T 11 P| + |OP| = |OT 11 | + |OT 22 |;

[0113] Echo direction: Mirror image direction;

[0114] Measured speed: Depends on the real target angle, the direction of the moving speed and the position of the reflector.

[0115] (3) First - order multipath ghost target T 21 Origin: Radar - Reflector - Target - Radar;

[0116] Propagation path: O → P → T 11 → O;

[0117] Target ranging: |OT 11 | + |T 11 P| + |OP| = |OT 11|+|OT 22 |;

[0118] Echo direction: direct view direction;

[0119] Measured speed: depends on the true target angle, direction of motion speed, and reflector position.

[0120] (4) Second-order multipath ghost target T 22 From: radar - reflector - radar - reflector - radar;

[0121] Propagation path: O → P → T 11 → P → O;

[0122] Target ranging: 2|T 11 P| + 2|OP|;

[0123] Echo direction: mirror direction;

[0124] Measured speed: depends on the true target angle, direction of motion speed, and reflector position.

[0125] From the above analysis, it can be seen that for one real target with one reflector in radar target imaging, three multipath ghost targets will be generated, and their distance information is as follows:

[0126] Real target T 11 : Distance is |OT 11 |, echo direction: direct view direction;

[0127] Ghost target T 12 : Distance is (|OT 11 | + |T 11 P| + |OP|) / 2 = (|OT 11 | + |OT 22 |) / 2, echo direction: mirror direction.

[0128] Ghost target T 21 : Distance is (|OT 11 | + |T 11 P| + |OP|) / 2 = (|OT 11 | + |OT 22 |) / 2, echo direction: direct view direction.

[0129] Ghost target T 22 : Distance is |T 11 P| + |OP|, echo direction: mirror direction.

[0130] Calculating the distance information of each of the above targets, it can be obtained that the distance of T 12 and the distance of T 21are equidistant, that is, the distances of the two first-order multipath ghost targets are equal. And the real target T 11 and the second-order ghost target T 22 The sum of the distances is equal to twice the distance of a single first-order multipath ghost target.

[0131] Since Doppler information characterizes the target velocity, which is approximately the first derivative of the distance information with respect to time, similarly, there are also laws similar to those of distance information for Doppler information: T 12 The Doppler information of 21 is equal to the Doppler information of T, that is, the Doppler information of the two first-order multipath ghost targets is equal. And T 11 and T 22 The sum of the Dopplers is equal to twice the Doppler of T 12 .

[0132] In step S1, the specific working process of the radar system architecture as shown in Figure 2 is as follows: The signal output by the signal source is transmitted by the transmitting antenna array after passing through the power divider and the first power amplifier. The output of the power divider enters the first mixer and the 90° phase shifter at the same time, and the output of the 90° phase shifter enters the second mixer. The output of the receiving antenna array is input into the mixers in the two sampling branches through the low-noise second power amplifier. After the output of the mixer in the sampling branch passes through the filter and the amplifier, the orthogonal I and Q signals, that is, the complex analytic digital baseband signals, can be obtained through the AD sampling circuit. Finally, the complex analytic digital baseband signal is input into the digital signal processor for subsequent processing.

[0133] The subsequent steps can be implemented by the multipath ghost target elimination device as shown in Figure 4 . The specific process of step S2 is implemented by the FFT unit, the constant false alarm detection unit, and the peak focusing unit. The specific process is as follows: The radar output signal is successively subjected to FFT, the constant false alarm detection algorithm, and peak focusing to complete target detection, and the total number of detected targets and the distance information and Doppler information of each target are obtained. Define a data structure target for storing information, and the information of the target with index i is stored in the data structure target i =(range i , doppler i ).

[0134] After the moving target is processed through steps S1 and S2, its distribution on the range and Doppler images is as shown in Figure 7 , Figure 7 . The ordinate in Figure 7 represents the distance information of the target, and the abscissa represents the velocity information of the target. It can be seen from 11 that on the range and Doppler images, the real target T 12 and T21 The second - order multipath ghost target T 22 Presents a collinear characteristic, that is, it is distributed on the same straight line and is evenly spaced. Among them, the two first - order ghost targets coincide in both the range image and the Doppler image, indicating that the range information and Doppler information of the two first - order ghost targets are both the same.

[0135] Step S3 is implemented by a binarization unit, and its specific process is as follows: The information target of all detected targets in each frame i Is stored in the target set T, and a range - Doppler image is drawn with the data of this set, and then binarization is performed. Its horizontal and vertical axes are range information and Doppler information respectively. The cells with targets are assigned a value of 1, and other cells are assigned a value of 0.

[0136] Step S4 is implemented by a Hough transform unit, and its specific process is as follows: The binarized image obtained in step S3 is subjected to collinearity detection using the Hough transform. For each detected collinear group, if the number of targets included in this group is not less than 3, record its corresponding target quantity and the information of each target.

[0137] As can be seen from the previous analysis, for each reflector, two first - order multipath ghost targets and one second - order multipath ghost target are generated. Since the first - order multipath ghost targets coincide in the Doppler image, plus the real position, when there are actually 3 targets, it means that the real position is included.

[0138] One specific embodiment output result of the Hough transform in step S4 is as Figure 8 Shown. The intersection points of the curves in the figure are local peaks, that is, the number of intersection points is equal to the number of peaks. Therefore, there are a total of four peaks in this figure, so there are 4 collinear groups corresponding to them. Among them, the peak size at the square marked by the arrow is 3, representing that the group corresponding to this peak contains 3 targets. The sizes of the remaining 3 peaks are all 2, representing that each of them corresponds to 2 targets. According to the method of the present invention, only record the collinear group corresponding to the peak of 3, and the information of the 3 targets included in this group.

[0139] Step S5 is implemented by a statistical sorting unit, and its specific process is as follows: Sort the collinear groups obtained in S4 in ascending order of the number of targets included in each group. Inside each group, sort the targets in ascending order of range i From small to large. As can be seen from the Figure 8 Shown image, there is only 1 collinear group, which contains 3 targets.

[0140] Step S6 is implemented by a target traversal unit and a calculation and marking unit, and its detailed process is as Figure 9As shown in the figure, specifically: Each time, one group is selected for traversal in ascending order of the grouping index. In this group, three targets are selected each time to determine the multipath ghost targets. According to the judgment results, the multipath ghost targets are marked; Only when all combination methods within the current collinear group are traversed, the next group is selected. For example, if there are 5 targets in the group, there are 10 combinations of arbitrarily selecting 3. After all 10 combinations are judged, the traversal of this group ends. When all group targets are traversed, the search for multipath ghost targets ends.

[0141] Step S7 is implemented by the ghost elimination unit. Its specific process is: After eliminating all the marked multipath ghost targets from the set T in step S6, the set T is output as the real target information.

[0142] The first-frame radar imaging point cloud of a single real moving target in this embodiment is as Figure 9 and Figure 10 shown. The digital beamforming algorithm of the existing technology can be used to obtain the azimuth and elevation angle information of the target, and combined with the distance and Doppler information obtained by the present invention, the radar point cloud image is drawn.

[0143] Figure 9 is the original imaging point cloud containing all detected targets, Figure 10 is the imaging point cloud after eliminating the multipath ghost targets by the method of the present invention. The guardrail in the figure is an example of a reflector. Correspondingly, the schematic diagrams of the radar imaging point cloud accumulated for 40 frames before and after elimination are as Figure 11 and Figure 12 shown. From the results of the embodiment, it can be seen that in radar target imaging, the multipath ghost target elimination method proposed by the present invention can accurately search and eliminate first-order ghosts and second-order ghosts. This not only enhances the radar's ability to handle multipath ghost targets of different orders, but also effectively improves the accuracy and reliability of radar imaging of real targets in complex environments; In addition, the method of the present invention only starts from two basic physical information, distance and Doppler, and has the characteristics of simple implementation.

[0144] The foregoing are the preferred embodiments of the present invention. If the preferred implementation manners in each preferred embodiment are not obviously self-contradictory or based on a certain preferred implementation manner, each preferred implementation manner can be arbitrarily superimposed and combined for use. The embodiments and the specific parameters in the embodiments are only for clearly expressing the inventor's invention verification process, and are not used to limit the patent protection scope of the present invention. The patent protection scope of the present invention still takes its claims as the criterion. Any equivalent structural changes made by using the description and drawings of the present invention should be included in the protection scope of the present invention by the same token.

Claims

1. A method for eliminating multipath ghost targets in radar target imaging, characterized in that , including the following steps: S1. After the electromagnetic wave emitted by the radar is reflected by the target, the echo signal is received and processed by the radar system to obtain an output signal; S2. Perform target detection based on the radar output signal, obtain the total number of detected targets and the distance information and Doppler information of each target, and store the target information of the target with index number i into the data structure target i =(range i , doppler i ); where target i represents the target information of the i-th target, where range i represents the distance information of the i-th target, and where doppler i represents the Doppler information of the i-th target; S3. Store the target information target corresponding to all targets i in the target set T, draw a range-Doppler image based on all the data in the target set T, and then binarize the range-Doppler image to obtain a binary image; S4. Perform collinear target detection on the binarized image obtained in step S3, and judge each detected collinear group. If the number of targets in the group is not less than 3, record the number of targets corresponding to the group and the information of each target; S5. For the multiple collinear groups obtained in step S4, according to the number of targets m included in each group j sort the groups in ascending order, and then sort each group internally in ascending order according to the distance information range of each target i where j is the index number of the collinear group; S6. For the multiple groups of collinear groups recorded in step S5, select 1 group each time according to the index number j from small to large for traversal and marking, During each traversal, select 3 targets each time in the group to judge multi-path ghost targets, and mark the multi-path ghost targets according to the judgment results; The judgment method is specifically as follows: Let the three selected targets be a, b, and c, and in ascending order of distance information, they are target a , target b , target c ; The Doppler information of the three targets a, b, and c is respectively, doppler a , doppler b , doppler c ; If |(range a -range b )-(range b -range c )|<r |(doppler a -doppler b )-(doppler b -doppler c )|<d If both formulas hold, then target b and target c are marked as multipath ghost targets, where r and d are preset target distance threshold and target Doppler threshold; Otherwise, it is considered that no ghost target can be found in this judgment; Traverse all possible combinations of 3 targets within the group until all combinations are traversed, the traversal marking of this group ends, and the next group of collinear groups is traversed. When all collinear groups are traversed, go to step S7; S7. Eliminate all the multi-path ghost targets marked in step S6 from the target set T, and output the result as the real target information.

2. The multipath ghost target elimination method according to claim 1, wherein In step S2, the specific method for target detection of the radar output signal is: perform FFT, constant false alarm detection algorithm, and peak focusing on the radar output signal in sequence to obtain the total number of detected targets and the distance information and Doppler information of each target.

3. The multipath ghost target elimination method according to claim 1, characterized in that The specific steps of S3 are as follows: Store the information target of all detected targets in each frame i into the target set T, and draw a range-Doppler image with the data of this set, and then perform binarization. Its horizontal and vertical axes are range information and Doppler information respectively. The cells with targets are assigned a value of 1, and the other cells are all assigned a value of 0.

4. The multipath ghost target elimination method according to claim 1, wherein In step S4, the Hough transform is used for collinear detection of the binarized image obtained in step S3.

5. A multipath ghost target elimination device for radar target imaging, characterized in that, It includes: A target information acquisition unit, which is used to perform target detection on the radar output signal, obtain the range-Doppler information of all targets, and draw a binarized range-Doppler image. The output of the target information acquisition unit is used as the input of the collinear target grouping unit; A collinear target grouping unit, which is used to perform collinear detection on the target distribution on the binarized image, and record and sort the collinear groups. The output of the collinear target grouping unit is used as the input of the multi-path ghost target elimination unit; A multi-path ghost target elimination unit, which is used to traverse all collinear groups, and traverse the targets within each group to judge and mark multi-path ghost targets. After traversal, eliminate the marked multi-path ghost targets and output the real target information.

6. The multipath ghost target elimination device according to claim 5, characterized in that The target information acquisition unit includes: An FFT unit, which performs fast Fourier transform on the input of the target information acquisition unit in the range dimension and Doppler dimension. The output of the FFT unit is used as the input of the constant false alarm detection unit; A constant false alarm detection unit, which performs constant false alarm target detection on the FFT transform result output by the FFT unit; the output of the constant false alarm detection unit is used as the input of the peak focusing unit; A peak focusing unit, for the constant false alarm target detection result output by the constant false alarm detection unit, retains the cell with the highest signal-to-noise ratio at each peak, and sets the other cells to zero; the output of the peak focusing unit is used as the input of the binarization unit; A binarization unit, which assigns a value of 1 to the cells with targets in the peak focusing result and a value of 0 to the other cells, and outputs a binarized image. The output of the binarization unit is used as the input of the collinear target grouping unit.

7. The multi-path ghost target elimination device according to claim 5, wherein The collinear target grouping unit includes: A Hough transform unit that performs a Hough transform on the input of the collinear target grouping unit, and the output of the Hough transform unit serves as the input of the statistical sorting unit; A statistical sorting unit that statistically sorts the number of collinear groupings and the target information corresponding to each grouping according to the Hough transform result. Then, sorting is performed separately among and within each grouping; the output of the statistical sorting unit serves as the input of the multipath ghost target elimination unit.

8. The multipath ghost target elimination device according to claim 5, characterized in that The multipath ghost target elimination unit includes: A target traversal unit that, for all collinear groupings, sequentially selects one group each time. For each group, three targets are selected each time and input to the calculation and marking unit until all targets in all groups are fully traversed; The target traversal unit is respectively connected to the calculation and marking unit and the ghost elimination unit; A calculation and marking unit that determines multipath ghost targets for the three input targets and marks the multipath ghost targets; A ghost elimination unit that, when the target traversal unit has completed traversal, deletes the targets marked as multipath ghosts and then outputs the final real target information.

Images