Low slow small target detection method based on stepping ground clutter cancellation

Abstract

The invention discloses a low slow small target detection method based on stepping ground clutter cancellation, and relates to the technical field of radar detection. The method comprises the following steps: establishing a node coordinate system, and generating distance-Doppler two-dimensional data; obtaining current frame data and reference frame data, and performing ground clutter cancellation processing on the current frame data; then calculating and generating a steering vector, and performing inter-array-element beam forming on the data subjected to ground clutter elimination by using the steering vector to obtain synthesized distance-Doppler data; and obtaining target information and reference frame target information according to the synthesized distance-Doppler data, performing cancellation processing on the target information according to the reference frame target information, further performing target angle measurement, and outputting a target positioning result. According to the method, the detection probability of the radar on the low-speed small target under the influence of the ground clutter signal is improved, the loss of the real target signal in the ground clutter cancellation processing process is reduced, and the accuracy of ground clutter cancellation is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of radar detection, and particularly to a method for detecting low, slow and small targets based on stepped ground clutter cancellation. Background Art

[0002] Low, slow and small targets refer to low-altitude, slow-speed and small targets, which are characterized by low flight altitude, slow flight speed and small radar cross section. Common low, slow and small targets include multi-rotor drones, fixed-wing drones, drone swarms, tethered balloons, etc. They are easy to operate, convenient to carry and difficult to be detected and warned, posing a serious threat to the security work of key activities and areas. Due to the particularity of low, slow and small targets, they are greatly affected by ground clutter, and it is difficult for radar to accurately locate targets from ground clutter. Therefore, there is an urgent need for a method for detecting low, slow and small targets that can eliminate ground clutter. Summary of the Invention

[0003] In view of this, the present invention proposes a method for detecting low, slow and small targets based on stepped ground clutter cancellation. This method completes the elimination of ground clutter signals and improves the detection probability of radar for low, slow and small targets under the influence of ground clutter signals. By adopting a stepped reference frame selection rule, the loss of real target signals in the ground clutter cancellation process is reduced, and the accuracy of ground clutter cancellation is improved.

[0004] In order to achieve the above object, the technical solution adopted by the present invention is as follows:

[0005] A method for detecting low, slow and small targets based on stepped ground clutter cancellation, comprising the following steps:

[0006] Step 1, establish a node coordinate system for the detection radar to obtain the array element coordinates of the detection radar, and use the detection radar to receive the echo signal of the low, slow and small target to generate fast time-slow time two-dimensional data;

[0007] Step 2, perform ranging and velocity measurement processing on the fast time-slow time two-dimensional data obtained in Step 1 to generate range-Doppler two-dimensional data, and store this data to obtain the current frame data;

[0008] Step 3, set a reference frame data selection rule, and selectively store the range-Doppler data obtained in Step 2 to obtain multiple groups of reference frame data;

[0009] Step 4, perform ground clutter cancellation processing on the current frame data;

[0010] Step 5, calculate and generate a steering vector, and use this value to perform inter-element beamforming on the data after ground clutter cancellation to obtain the synthesized range-Doppler data;

[0011] Step 6, completing the threshold crossing detection of the range-Doppler data synthesized in step 5 to obtain the synthesized target information;

[0012] Step 7, generating and storing multiple sets of reference frame target information;

[0013] Step 8, performing reference frame target information elimination processing on the target information in step 6;

[0014] Step 9, complete the target angle measurement based on the target information after the reference frame target information is eliminated, and output the positioning result of the low, slow and small target.

[0015] Furthermore, the specific method of step 1 is:

[0016] S11, establish the node coordinate system XYZ in the single-transmit and single-receive scenario for the detection radar, and further obtain the array element coordinates (x n ,y n ,z n ), n=1,2,3,……,N;

[0017] S12, N array elements respectively continuously perform airspace scanning and receive data, each airspace scanning is composed of K wave positions, that is, each array element receives K frames of one-dimensional echo data after completing an airspace scan, wherein the length of one frame of echo data is 1×PM, P is the number of sampling points of the detection radar, and M is the number of pulses of the detection radar;

[0018] S13, sorting out the echo data received by each array element during the current spatial domain scanning process, and obtaining K frames of P×M fast-time-slow-time two-dimensional data corresponding to each array element.

[0019] Furthermore, the specific method of step 2 is:

[0020] S21, performing matched filtering and multi-pulse accumulation on K frames of P×M fast-time-slow-time two-dimensional data of N array elements respectively, realizing target ranging and speed measurement processing, and obtaining N×K frames of range-Doppler two-dimensional data;

[0021] S22, opening up N groups of current frame data storage space, each group storing K frames of range-Doppler two-dimensional data corresponding to one array element, recorded as current frame data of the current array element, using the storage space to store the N×K frames of range-Doppler two-dimensional data obtained in S21, that is, the current frame data of N array elements; respectively, continuously recording the frame numbers of the K frames of range-Doppler two-dimensional data obtained by each array element in the continuous spatial domain scanning;

[0022] After each new spatial scan of N array elements, the storage spaces of N groups of current frame data are cleared, and the new N×K frame range-Doppler two-dimensional data are stored correspondingly to obtain the new current frame data of N array elements.

[0023] Further, the specific manner of step 3 is as follows:

[0024] S31. For each array element, three groups of reference frame data storage spaces are allocated, and each group stores K frames of reference frame data; the initial values of the three groups of reference frame data storage spaces are all 0. Initially, the K-frame range-Doppler two-dimensional data corresponding to the first spatial scan are recorded as a group of reference frame data and stored in the first group of reference frame data storage space; for each subsequent new group of reference frame data, it is stored sequentially in the three groups of reference frame data storage spaces. When all three groups of reference frame data storage spaces are full, the first group of reference frame data storage space is cleared, the second group of reference frame data is shifted to the first group of reference frame data storage space, the third group of reference frame data is shifted to the second group of reference frame data storage space, and the third group of reference frame data storage space is vacated for storing new reference frame data;

[0025] S32. Based on the frame number of the current frame data corresponding to each array element in the current frame data storage space, calculate the corresponding bit frame number difference between it and the latest group of reference frame data in the reference frame data storage space of the corresponding array element. If the corresponding bit frame number difference is pK, then record this group of current frame data as a new group of reference frame data and store it in the reference frame data storage space corresponding to this array element.

[0026] Further, the specific manner of step 4 is as follows:

[0027] Based on the frame number of the current frame data corresponding to each array element in the current frame data storage space, calculate the corresponding bit frame number difference between it and the first group of reference frame data in the latest reference frame data storage space of the corresponding array element. If the corresponding bit frame number difference is greater than or equal to 2pK and less than 3pK, then subtract the corresponding bits of the current frame data corresponding to this array element from the first group of reference frame data in the latest reference frame data storage space of this array element, and record the result after subtraction as the K-frame ground clutter cancellation data corresponding to this array element.

[0028] Further, the specific manner of step 5 is as follows:

[0029] S51. Calculate the pointing coordinates (x, y, z) in the node coordinate system based on the received wave position pointing (azi, ele) of the echo signal. The specific calculation expression is: x = cos(azi)cos(ele), y = sin(ele), z = sin(azi)cos(ele); where azi is the azimuth angle of the echo signal, ele is the elevation angle of the echo signal, the elevation angle of the node coordinate system is defined as the angle between the echo signal and the XOZ plane, and the azimuth angle is defined as the angle between the projection of the echo signal on the XOZ plane and the X-axis;

[0030] S52. Use the wave position pointing coordinates (x, y, z) and the coordinates (x n , y n , z n ) of N array elements to calculate the steering vectors corresponding to the N array elements. The calculation expression is: where λ is the wavelength of the echo signal;

[0031] S53. Conjugate multiply the 1×N steering vector corresponding to the N array elements with the N×K clutter-canceled data to complete beamforming between array elements, and obtain the synthesized range-Doppler data of K frames of P×M.

[0032] Further, the specific method of step 6 is as follows:

[0033] S61. Set the signal-to-noise ratio for threshold-crossing detection, compare each of the P×M elements in each frame of the synthesized range-Doppler data with this signal-to-noise ratio, and record the elements greater than this signal-to-noise ratio and their corresponding row numbers and column numbers;

[0034] S62. Convert the row numbers and column numbers recorded in S61 to obtain the range and velocity information of the low, slow, and small targets. The conversion expression is:

[0035] where row is the row number, col is the column number, c is the speed of light, T s is the sampling interval of the detection radar, PRF is the pulse repetition frequency of the detection radar, range is the range of the low, slow, and small target, and v is the velocity of the low, slow, and small target;

[0036] S63. Open a group of target information storage spaces, and the target information storage spaces store K frames of data; for the K frames of synthesized range-Doppler data of P×M, record all the elements greater than the signal-to-noise ratio, their corresponding row numbers and column numbers, and the corresponding range and velocity information of the low, slow, and small targets in each frame of the synthesized range-Doppler data as one frame of target information, so as to obtain K frames of target information and store them in the target information storage spaces;

[0037] Whenever new K frames of P×M synthesized range-Doppler data are obtained, the target information storage space is cleared, and K frames of target information corresponding to the new K frames of P×M synthesized range-Doppler data are stored.

[0038] Furthermore, the specific method of step 7 is:

[0039] S71, opening up three groups of reference frame target information storage spaces, each group storing K frame data; the initial values ​​of the three groups of reference frame target information storage spaces are all 0, and the first obtained K frame target information is recorded as a group of reference frame target information, and stored in the first group of reference frame target information storage space;

[0040] For each subsequent new reference frame target information, the information is stored backward in the three reference frame target information storage spaces in sequence. When the three reference frame target information storage spaces are all full, the first reference frame target information storage space is cleared, the second reference frame target information is shifted to the first reference frame target information storage space, and the third reference frame target information is shifted to the second reference frame target information storage space, leaving the third reference frame target information storage space empty for storing new reference frame target information.

[0041] S72: Based on the K-frame synthesized distance-Doppler data corresponding to the K-frame target information in the target information storage space, calculate the corresponding bit frame number difference between the K-frame synthesized distance-Doppler data and the latest set of reference frame target information in the reference frame target information storage space; if the corresponding bit frame number difference is pK, the current K-frame target information is recorded as a new set of reference frame target information and stored in the corresponding reference frame target information storage space.

[0042] Furthermore, the specific method of step 8 is:

[0043] Based on the frame number of the K-frame synthesized distance-Doppler data corresponding to the K-frame target information in the target information storage space, calculate the corresponding frame number difference between the K-frame synthesized distance-Doppler data and the first group of reference frame target information in the latest reference frame target information storage space. If the corresponding frame number difference is greater than or equal to 2pK and less than 3pK, compare the current K-frame target information with the corresponding frame of the first group of reference frame target information in the latest reference frame target information storage space. If the two contain the same row and column number information in a corresponding frame, then remove the information related to the row and column number in the frame target information to obtain the target information data after K frame cancellation.

[0044] Furthermore, the specific method of step 9 is:

[0045] According to the dense beam angle measurement method, using the angle measurement steering vectors at different azimuth-elevation, based on the target information data after cancellation, calculate the signal-to-noise ratio of the low, slow, and small target in each direction, and the peak position is the position of the low, slow, and small target, thereby completing the angle measurement of the low, slow, and small target and outputting the positioning result of the low, slow, and small target.

[0046] Due to the adoption of the above technical solution, the beneficial effects of the present invention compared with the prior art are as follows:

[0047] 1. The present invention respectively performs ground clutter cancellation processing on the range-Doppler two-dimensional data and multi-target detection information, completes the elimination of the ground clutter signal, and improves the detection probability of the radar for low, slow, and small targets under the influence of the ground clutter signal.

[0048] 2. The present invention adopts a stepped reference frame selection rule, reduces the loss of the real target signal during the ground clutter cancellation process, and improves the accuracy of the ground clutter cancellation. Description of the Drawings

[0049] Figure 1 It is the overall flowchart of a method for detecting low, slow, and small targets based on stepped ground clutter cancellation in an embodiment of the present invention.

[0050] Figure 2 It is a schematic diagram of the reference frame selection rule in an embodiment of the present invention.

[0051] Figure 3 、 Figure 4 They are respectively the two-dimensional plane target positioning result diagrams corresponding to no ground clutter cancellation and with ground clutter cancellation in an embodiment of the present invention.

[0052] Figure 5 、 Figure 6 They are respectively the three-dimensional coordinate target positioning result diagrams corresponding to no ground clutter cancellation and with ground clutter cancellation in an embodiment of the present invention. Detailed Embodiment

[0053] The following further describes the content of the present invention in conjunction with the drawings and specific embodiments.

[0054] A method for detecting low, slow, and small targets based on stepped ground clutter cancellation, as Figure 1 shown, specifically includes the following steps:

[0055] S1. Set the low, slow, and small target as an unmanned aerial vehicle, establish a node coordinate system, obtain the array element coordinates, and organize the received echo signals to generate fast time-slow time two-dimensional data, which specifically includes the following steps:

[0056] S11: Establish a node coordinate system in a single-transmitter single-receiver scenario, construct the coordinates (x n , y n , zn ) where \(n = 1, 2, \cdots, 16\). The pitch angle of the node coordinate system is defined as the angle with the \(XOZ\) plane, and the azimuth angle is defined as the angle between the projection in the \(XOZ\) plane and the \(X\)-axis;

[0057] S12: The 16 array elements perform spatial scanning and receive data respectively. One spatial scan consists of 5 wave positions, that is, each array element completes one spatial scan and receives 5 frames of one-dimensional echo data;

[0058] S13: The length of one frame of echo data is \(1\times3037500\). The echo data (\(1\times3037500\)) received by each array element is sorted out to obtain two-dimensional fast time - slow time data (\(486\times6250\)).

[0059] S2: Perform ranging and velocity measurement processing on the two-dimensional data obtained in S1 to generate two-dimensional range - Doppler data, and store this data to obtain the current frame data. The specific steps are as follows:

[0060] S21: Perform matched filtering and multi-pulse accumulation on the two-dimensional data of the 16 array elements obtained in S1 respectively to achieve target ranging and velocity measurement processing, and obtain \(16\times5\) frames of two-dimensional range - Doppler data;

[0061] S22: Open up 16 groups of storage spaces for the current frame data, each group stores 5 frames of data, and use this storage space to store the \(16\times5\) frames of two-dimensional range - Doppler data obtained in S21 to obtain the current frame data.

[0062] S3: As Figure 2 shown, store the range - Doppler data obtained in S2 according to the reference frame data selection rule to generate multiple groups of reference frame data. The specific steps are as follows:

[0063] S31: For each array element, open up 3 groups of storage spaces for reference frame data, each group stores 5 frames of data; the first group of space stores the \(1^{st}\) to \(5^{th}\) frames of two-dimensional range - Doppler data, and the data of the remaining groups is initialized to 0;

[0064] S32: The step interval of each group of reference frames is 50 frames of data. Judge whether the frame number of the current frame differs from the frame number of the previous group of reference frames by 50. Take 5 frames of data that meet the interval condition as a group and store them in the storage space opened up in S31;

[0065] Specifically, in this embodiment, \(p = 10\), so \(p\times K = 50\);

[0066] S33: Update the reference frame data according to the reference frame update rule. The update rule is: update the reference frame data every 50 frames of data, that is, shift the second group of data to the first group, shift the third group of data to the second group, and so on, leaving the storage location of the third group empty for storing the new reference frame.

[0067] S4. Complete the clutter cancellation process for the current frame data in S2, which specifically includes the following steps:

[0068] Perform ground clutter cancellation on the current frame data obtained in S22 according to the ground clutter cancellation condition. The ground clutter cancellation condition is: 100 ≤ current frame number - first group reference frame number < 150. The ground clutter cancellation method is: subtract the 5-frame data of the first group reference frame from the 5-frame data of the current frame in S22, that is, subtract the data of the 1st to 5th reference frames from the data of the 101st to 150th frames in the cancellation process, and so on to complete the ground clutter cancellation process.

[0069] Specifically, the first group reference frame number refers to the latest first group reference frame number in the latest reference frame data storage space; 2×p×K = 100, 3×p×K = 150;

[0070] S5. Calculate and generate the steering vector, and use this value to perform inter-element beamforming on the data after ground clutter cancellation to obtain the synthesized range-Doppler data, which specifically includes the following steps:

[0071] S51: Calculate the wave position pointing coordinates (x, y, z) in the node coordinate system according to the received wave position pointing (azi, ele). The specific calculation expression is: x = cos(azi)cos(ele), y = sin(ele), z = sin(azi)cos(ele). Set 5 wave positions to be (-5.64, -19.35), (-23.14, -19.49), (0.06, -17.96), (-17.39, -14.92), (-22.71, -17.40) respectively. According to the above expression, the wave position pointing coordinates (x, y, z) can be obtained;

[0072] S52: Use the wave position pointing coordinates (x, y, z) and the 16 element coordinates (x n , y n , z n ) to calculate the steering vectors corresponding to the 16 elements. The calculation expression is: where λ = 0.3m;

[0073] S53: Conjugate multiply the steering vector obtained in S51 and the cancelled data obtained in S41 to complete the inter-element beamforming and obtain the synthesized range-Doppler data.

[0074] Specifically, conjugate multiply the steering vector of 1 row and 16 columns corresponding to 16 array elements with the data after clutter cancellation of 16 rows and 5 columns to complete beamforming between array elements, and obtain the synthesized range-Doppler data of 5 frames of 486×6250. The 5 frames of synthesized range-Doppler data are recorded according to the frame numbers of the 5 frames of current frame data before clutter cancellation corresponding thereto;

[0075] S6. Perform threshold crossing detection on the synthesized data in S5 to obtain the synthesized target information, which specifically includes the following steps:

[0076] S61: Set the signal-to-noise ratio for threshold crossing detection, compare the synthesized range-Doppler data with this value, the value exceeding this threshold is the target, and store the value greater than the threshold and its corresponding row number and column number;

[0077] S62: Convert the row number (row) and column number (col) stored in S61 to obtain the range and velocity information of the target. The conversion expression is: where c is the speed of light, T s = 16us, PRF = 10000, M = 6250.

[0078] S7. Generate and store target information of multiple groups of reference frames, which specifically includes the following steps:

[0079] S71: Open up storage spaces for target information of 3 groups of reference frames, each group storing 5 frames of data; the first group of spaces stores the target information of frames 1 to 5, and the data of the remaining groups is initialized to 0;

[0080] S72: The step interval of each group of reference frames is 50 frames of data. Determine whether the frame number of the current frame differs from the frame number of the previous group of reference frames by 50. Take 5 frames of information that meet the interval condition as a group and store them in the storage space opened up in S71.

[0081] S8. Perform reference frame target rejection processing on the target information in S6 and complete the update of the reference frame target information, which specifically includes the following steps:

[0082] Refer to the clutter cancellation condition in S4, and use the target information obtained in S62 to subtract the target information of the first group of reference frames in S71 to complete the reference frame target rejection processing;

[0083] S9. Complete target angle measurement and output the target positioning result, which specifically includes the following steps:

[0084] According to the dense beam angle measurement method, use the angle measurement steering vectors at different azimuth-elevation to calculate the signal-to-noise ratio of the target in each direction. The peak position is the position of the target, thereby completing the target angle measurement and outputting the target positioning result.

[0085] Specifically, it includes the distance to the target and angle measurement to obtain the target positioning result. Among them, for each set of target information data after cancellation for each frame after the target rejection process in the reference frame, each set contains information related to row and column numbers, that is, one target detected in the data of that frame; according to the number of detected targets, the corresponding number of signal-to-noise ratio peaks is calculated in S9, which is the position of the corresponding target, and the target angle measurement is completed.

[0086] The two-dimensional plane target positioning result is as Figure 3 , Figure 4 shown. Among them Figure 3 is the target positioning result without ground clutter cancellation processing, Figure 4 is the target positioning result after the ground clutter cancellation processing of the present invention. By comparing Figure 3 , Figure 4 the two figures, it can be found that when the ground clutter cancellation processing is not performed, the target positioning result deviates greatly from the actual UAV trajectory and can hardly correspond. The present invention can significantly improve the detection probability of low, slow, and small targets in a complex ground clutter environment.

[0087] The three-dimensional coordinate target positioning result is as Figure 5 , Figure 6 shown. Among them Figure 5 is the target positioning result without ground clutter cancellation processing, Figure 6 is the target positioning result after the ground clutter cancellation processing of the present invention. By comparing Figure 5 , Figure 6 the two figures, it can be found that when the ground clutter cancellation processing is not performed, the Z coordinate positioning result is basically inaccurate and the three-dimensional positioning result appears intermittently. The present invention can significantly improve the positioning accuracy and detection probability of low, slow, and small targets in a complex ground clutter environment. Figures 3 to 6 The t(s) in

[0088] represents the observation time. Those skilled in the art will realize that the described embodiments are to help readers understand the principles of the present invention and should be understood that the protection scope of the present invention is not limited to the described embodiments. For those skilled in the art, the present invention can have various changes and modifications. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the scope of the claims of the present invention.

Claims

1. A low, slow and small target detection method based on step-by-step ground clutter cancellation, characterized in that: The following steps are involved: Step 1: Establish a node coordinate system for the detection radar, obtain the array element coordinates of the detection radar, use the detection radar to receive the echo signal of the low, slow and small target, and generate fast time-slow time two-dimensional data; Step 2, performing distance measurement and speed measurement processing on the fast time-slow time two-dimensional data obtained in step 1 to generate distance-Doppler two-dimensional data, and storing the data to obtain current frame data; Step 3, setting reference frame data selection rules, selectively storing the range-Doppler data obtained in step 2, and obtaining multiple groups of reference frame data; Step 4, performing ground clutter cancellation processing on the current frame data; Step 5, calculate and generate a steering vector, and use the value to perform inter-element beamforming on the data after ground clutter elimination to obtain synthesized range-Doppler data; Step 6, completing the threshold crossing detection of the range-Doppler data synthesized in step 5 to obtain the synthesized target information; Step 7, generating and storing multiple sets of reference frame target information; Step 8, performing reference frame target information elimination processing on the target information in step 6; Step 9, complete the target angle measurement based on the target information after the reference frame target information is eliminated, and output the positioning result of the low, slow and small target.

2. According to claim 1, a method for detecting low, slow and small targets based on step-by-step ground clutter cancellation is characterized in that: The specific method of step 1 is: S11, establish the node coordinate system XYZ in the single-transmit and single-receive scenario for the detection radar, and further obtain the array element coordinates (x n ,y n ,z n ), n=1,2,3,……,N; S12, N array elements respectively continuously perform airspace scanning and receive data, each airspace scanning is composed of K wave positions, that is, each array element receives K frames of one-dimensional echo data after completing an airspace scan, wherein the length of one frame of echo data is 1×PM, P is the number of sampling points of the detection radar, and M is the number of pulses of the detection radar; S13, sorting out the echo data received by each array element during the current spatial domain scanning process, and obtaining K frames of P×M fast-time-slow-time two-dimensional data corresponding to each array element.

3. The method for detecting low, slow and small targets based on step-by-step ground clutter cancellation according to claim 2 is characterized in that: The specific method of step 2 is: S21, performing matched filtering and multi-pulse accumulation on K frames of P×M fast-time-slow-time two-dimensional data of N array elements respectively, realizing target ranging and speed measurement processing, and obtaining N×K frames of range-Doppler two-dimensional data; S22, opening up N groups of current frame data storage space, each group storing K frames of range-Doppler two-dimensional data corresponding to one array element, recorded as current frame data of the current array element, using the storage space to store the N×K frames of range-Doppler two-dimensional data obtained in S21, that is, the current frame data of N array elements; respectively, continuously recording the frame numbers of the K frames of range-Doppler two-dimensional data obtained by each array element in the continuous spatial domain scanning; Whenever N array elements perform a new airspace scan, the storage space of N groups of current frame data is cleared, and new N×K frames of range-Doppler two-dimensional data are stored accordingly to obtain new current frame data of N array elements.

4. The method for detecting low, slow and small targets based on step-by-step ground clutter cancellation according to claim 3 is characterized in that: The specific method of step 3 is: S31, for each array element, three groups of reference frame data storage spaces are opened, each group stores K frames of reference frame data; the initial values ​​of the three groups of reference frame data storage spaces are all 0, and initially the K frames of range-Doppler two-dimensional data corresponding to the first spatial domain scan are recorded as a group of reference frame data, and stored in the first group of reference frame data storage spaces; for each subsequent group of new reference frame data, the data are stored backward in the three groups of reference frame data storage spaces in sequence, and when the three groups of reference frame data storage spaces are full, the first group of reference frame data storage spaces are cleared, the second group of reference frame data are shifted to the first group of reference frame data storage spaces, and the third group of reference frame data are shifted to the second group of reference frame data storage spaces, leaving the third group of reference frame data storage spaces free for storing new reference frame data; S32, based on the frame number of the current frame data corresponding to each array element in the current frame data storage space, calculate the corresponding frame number difference between it and the latest group of reference frame data in the reference frame data storage space of the corresponding array element; if the corresponding frame number difference is pK, record the group of current frame data as a new group of reference frame data and store them in the reference frame data storage space corresponding to the array element.

5. The method for detecting low, slow and small targets based on step-by-step ground clutter cancellation according to claim 4 is characterized in that: The specific method of step 4 is: Based on the frame number of the current frame data corresponding to each array element in the current frame data storage space, calculate the corresponding bit frame number difference between the current frame data corresponding to the array element and the first group of reference frame data in the latest reference frame data storage space of the corresponding array element. If the corresponding bit frame number difference is greater than or equal to 2pK and less than 3pK, subtract the corresponding bits of the current frame data corresponding to the array element from the first group of reference frame data in the latest reference frame data storage space of the array element, and record the subtraction result as the K-frame ground clutter cancellation data corresponding to the array element.

6. The method for detecting low, slow and small targets based on step-by-step ground clutter cancellation according to claim 5 is characterized in that: The specific method of step 5 is: S51, calculate the wave direction coordinates (x, y, z) in the node coordinate system according to the received wave direction (azi, ele) of the echo signal, and the specific calculation expression is: x=cos(azi)cos(ele), y=sin(ele), z=sin(azi)cos(ele); wherein, azi is the azimuth of the echo signal, ele is the elevation angle of the echo signal, the elevation angle of the node coordinate system is defined as the angle between the echo signal and the XOZ plane, and the azimuth angle is defined as the angle between the projection of the echo signal on the XOZ plane and the X-axis; S52, using the wave position pointing coordinates (x, y, z) and the N array element coordinates (x n ,y n ,z n ), calculate the steering vector corresponding to N array elements, and the calculation expression is: Where λ is the wavelength of the echo signal; S53, conjugate multiplying the 1-row and N-column steering vectors corresponding to the N array elements with the N-row and K-column ground clutter cancellation data to complete inter-element beamforming and obtain K frames of P×M synthesized range-Doppler data.

7. The method for detecting low, slow and small targets based on step-by-step ground clutter cancellation according to claim 6 is characterized in that: The specific method of step 6 is: S61, setting a signal-to-noise ratio value for threshold detection, comparing P×M elements in each frame of synthesized range-Doppler data with the signal-to-noise ratio value, and recording elements greater than the signal-to-noise ratio value and their corresponding row and column numbers; S62, converting the row number and column number recorded in S61 to obtain the distance and speed information of the low, slow and small target. The conversion expression is: Among them, row is the row number, col is the column number, c is the speed of light, T s is the sampling interval of the detection radar, PRF is the pulse repetition frequency of the detection radar, range is the distance of the low, slow and small target, and v is the speed of the low, slow and small target; S63, opening a group of target information storage space, wherein the target information storage space stores K frames of data; for K frames of P×M synthesized range-Doppler data, all elements greater than the signal-to-noise ratio value in each frame of the synthesized range-Doppler data, their corresponding row numbers and column numbers, and the corresponding distance and speed information of the low, slow and small target are recorded as one frame of target information, thereby obtaining K frames of target information, and storing them in the target information storage space; Whenever new K frames of P×N synthesized range-Doppler data are obtained, the target information storage space is cleared, and K frames of target information corresponding to the new K frames of P×M synthesized range-Doppler data are stored.

8. The method for detecting low, slow and small targets based on step-by-step ground clutter cancellation according to claim 7 is characterized in that: The specific method of step 7 is: S71, opening up three groups of reference frame target information storage spaces, each group storing K frame data; the initial values ​​of the three groups of reference frame target information storage spaces are all 0, and the first obtained K frame target information is recorded as a group of reference frame target information, and stored in the first group of reference frame target information storage space; For each subsequent new reference frame target information, the information is stored backward in the three reference frame target information storage spaces in sequence. When the three reference frame target information storage spaces are all full, the first reference frame target information storage space is cleared, the second reference frame target information is shifted to the first reference frame target information storage space, and the third reference frame target information is shifted to the second reference frame target information storage space, leaving the third reference frame target information storage space empty for storing new reference frame target information. S72: Based on the K-frame synthesized distance-Doppler data corresponding to the K-frame target information in the target information storage space, calculate the corresponding bit frame number difference between the K-frame synthesized distance-Doppler data and the latest set of reference frame target information in the reference frame target information storage space; if the corresponding bit frame number difference is pK, the current K-frame target information is recorded as a new set of reference frame target information and stored in the corresponding reference frame target information storage space.

9. The method for detecting low, slow and small targets based on step-by-step ground clutter cancellation according to claim 8, characterized in that: The specific method of step 8 is: Based on the frame number of the K-frame synthesized distance-Doppler data corresponding to the K-frame target information in the target information storage space, calculate the corresponding frame number difference between the K-frame synthesized distance-Doppler data and the first group of reference frame target information in the latest reference frame target information storage space. If the corresponding frame number difference is greater than or equal to 2pK and less than 3pK, compare the current K-frame target information with the corresponding frame of the first group of reference frame target information in the latest reference frame target information storage space. If the two contain the same row and column number information in a corresponding frame, then remove the information related to the row and column number in the frame target information to obtain the target information data after K frame cancellation.

10. The method for detecting low, slow and small targets based on step-by-step ground clutter cancellation according to claim 9, characterized in that: The specific method of step 9 is: According to the dense beam angle measurement method, the angle measurement guidance vectors under different azimuths and elevations are used to calculate the signal-to-noise ratio of the low, slow and small targets in all directions based on the target information data after cancellation. The peak position is the location of the low, slow and small target, thereby completing the angle measurement of the low, slow and small target and outputting the positioning result of the low, slow and small target.

Images