A clutter map design method based on slow decay idea

Abstract

The application relates to a clutter map design method based on slow-decay thought, and belongs to the technical field of radar signal processing. The method comprises the following steps: creating an overall clutter map storage space; acquiring a range-Doppler submap within a current coherent accumulation time; calculating the index value of a clutter map submap corresponding to a current beam direction in the overall clutter map; judging whether the current time is within the first several scanning periods after the radar starts working or mode switching; if yes, directly assigning the range-Doppler submap to the clutter map submap to obtain a new clutter map submap; if not, acquiring the current direction clutter map submap according to the index value, subtracting the clutter map submap from the range-Doppler submap to obtain a difference matrix, then obtaining a sign matrix through shifting and bit-and operations, accessing an update rate array with the sign matrix as an index to obtain an update rate coefficient matrix; and weighting and summing the range-Doppler submap and the clutter map submap with the update rate matrix as a weight to obtain a new clutter map submap.

Description

Technical Field

[0001] The invention belongs to the technical field of radar signal processing, and in particular relates to a clutter map design method based on a slow decay idea. Background Art

[0002] Clutter mapping is a target detection technique that uses the statistical values ​​of clutter in the slow-time domain. It assumes that clutter is statistically stationary in time. The design of clutter mapping must consider data storage space limitations. If data space is sufficient, clutter maps should be constructed in as many dimensions as possible, such as azimuth, elevation, range, frequency, and PRF. Multidimensional clutter mapping can better suppress clutter.

[0003] Generally speaking, the design of the elevation, range, frequency, and PRF dimensions is based on radar system parameters, while the design of the azimuth dimension is relatively flexible. There are two common ways to divide azimuth granules: by beam dwell and by fixed code disk. If divided by beam dwell, the azimuth scan angle during the beam dwell time = antenna rotation speed × beam dwell time. If divided by fixed code disk, the azimuth scan angle corresponding to each code disk = 360° / fixed value, where the fixed value here refers to the fixed number of azimuth granules (for example, 256 fixed azimuth granules).

[0004] Regardless of the division method used, the number of azimuth particles must be within a reasonable range. This is because if the azimuth particles are divided too finely, it will be difficult to complete all or most updates of the clutter map within a scanning cycle, resulting in poor clutter map performance. If the azimuth particles are divided too coarsely, echo data from multiple beam dwell times will fall within the same azimuth particle, causing echo data from different beam dwell times to affect each other and causing excessive iteration of the clutter map, which is not conducive to clutter suppression.

[0005] During coherent integration, the radar creates a range-Doppler submap for each accumulation period. This submap represents the amplitude of the echo in the range-Doppler dimension within the current accumulation period. Based on the current beam position, a submap corresponding to the particle in the clutter map is selected. The range-Doppler submap is weighted and summed with the clutter map submap to iteratively update the clutter map submap for the current accumulation period. The weight here represents another important parameter in clutter map design—the update rate.

[0006] The update rate selection must consider the target characteristics of interest to the radar. Typically, within a single scan cycle, the target must be able to move beyond one range quantization unit. Otherwise, the target signal will be included in the update of the current clutter map sub-image, raising the detection threshold and causing detection loss. In severe cases, this can even lead to self-shadowing. The update rate is typically the inverse of a power of 2. For example, if the update rate is 1 / 8, the new clutter map sub-image = range - Doppler sub-image × 1 / 8 + clutter map sub-image × 7 / 8.

[0007] In scenes with stable scattering features, the conventional clutter pattern design method described above generally achieves good clutter suppression and target detection results when appropriate design parameters are selected. However, in real-world scenarios, many features, such as leaves, grass, and wheat fields, often have time-varying scattering characteristics. Natural environmental changes (such as wind and rain) can cause scattering points to oscillate back and forth from a fixed position, causing random fluctuations in the echo amplitude and phase. This results in excessive clutter traces and makes suppression difficult.

[0008] In actual engineering, some radar systems employ sliding window processing to optimize azimuth and angle measurement performance. This inevitably causes echo data from different beam dwell times to interact with each other, and the clutter map undergoes multiple iterations within a single scan cycle, inevitably impacting clutter suppression performance. To address these challenges, we have devised a clutter map design method based on the concept of slow decay. Summary of the Invention

[0009] Technical problems to be solved

[0010] During radar target detection, ground clutter, caused by the time-varying scattering characteristics of ground objects, experiences random fluctuations in amplitude and phase, even spreading to other resolution units. This results in numerous clutter traces that are difficult to suppress. To address this issue, the present invention provides a clutter map design method based on the concept of slow decay.

[0011] Technical Solution

[0012] A clutter map design method based on the slow decay concept is characterized by the following steps:

[0013] Step 1: Create a storage space for the overall clutter map matrix CMap; the overall clutter map matrix CMap is a (NUM×M)×N matrix; where NUM is the number of azimuth particles, M is the number of Doppler channels, and N is the number of range units;

[0014] Step 2: Get the distance-Doppler sub-image Data within the current coherent accumulation time rd ; The distance-Doppler sub-graph Data rd is an M×N matrix;

[0015] Step 3: Calculate the particle index value index at the current beam azimuth az , and the clutter subgraph Data corresponding to the orientation of the particles cm The row index value index in the clutter map matrix CMap row , the clutter graph subgraph Data cm For an M×N matrix:

[0016] index az =fix(azimuth / 360*NUM),(0≤index az ≤NUM-1)

[0017] index row =index az ×M,(0≤index row ≤NUM×M-1)

[0018] Where fix(·) means rounding down, and the unit of azimuth is degree;

[0019] Step 4: Assume that the first five scanning cycles after the radar starts working or the mode is switched are the clutter map fast establishment time, determine whether the current time is within the first five scanning cycles, if so, jump to step 5, if not, jump to step 6;

[0020] Step 5: Convert the range-Doppler subgraph Data rd Directly assign to the clutter graph subgraph Data cm , get the new clutter graph subgraph Data cm_new , namely Data cm_new =Data rd , the clutter graph subgraph Data cm_new is an M×N matrix; then jump to step 7;

[0021] Step 6: Convert the range-Doppler subgraph Data rd and the clutter graph subgraph Data cm Weighted summation to obtain the new clutter graph subgraph Data cm_new ; Specific steps are as follows:

[0022] Step 6.1: Any element in the range-Doppler subgraph is a ij , where 0≤i≤M-1, 0≤j≤N-1, any element in the clutter graph subgraph is b ij , find the difference between the two amplitudes diff ij and symbol sign ij :

[0023] diff ij =a ij -b ij

[0024] sign ij =(((Uint32)diff ij )>>31)&1

[0025] Among them, Uint32(·) means that the data type is forced to be converted to unsigned 32 bits, (·)>>31 means right shift 31 bits, and (·)&1 means bitwise AND 1;

[0026] Step 6.2: Use the large update rate k1 and the small update rate k2 to form an array array = {k1, k2}, that is, array(0) = k1, array(1) = k2, and find any element kk of the update rate matrix K ij , the update rate matrix K is an M×N matrix:

[0027] kk ij =array(sign ij )

[0028] Step 6.3: Update the range-Doppler subgraph Data using the update rate matrix K rd and the clutter graph subgraph Data cm Perform weighted summation to obtain the new clutter graph subgraph Data cm_new :

[0029] Data cm_new =Data rd .*K+Data cm .*(IK)

[0030] Where “.*” indicates the multiplication of corresponding elements of the matrix, and I is the identity matrix;

[0031] Step 7: Data the new clutter subgraph cm_new Fill in its storage space in the clutter map matrix CMap and complete the update of the clutter map sub-map within the current coherent accumulation time:

[0032] CMap(index row :index row +M-1,1:N)=Data cm_new

[0033] where index row :index row +M-1 means from the index row Row to index row +M-1 rows, 1:N means from column 1 to column N;

[0034] Step 8: Next coherent integration time, obtain the new range-Doppler sub-image Data rd , repeat steps 2 to 8 to complete the update of the clutter map sub-image within each accumulation time until the radar stops scanning.

[0035] A further technical solution of the present invention is implemented using C code.

[0036] A further technical solution of the present invention: In step 4, the method for determining whether the current moment is within the first five scanning cycles is to see whether a certain azimuth angle has been passed five times.

[0037] A further technical solution of the present invention is that the azimuth angle is zero degree.

[0038] A computer system, characterized in that it includes: one or more processors, and a computer-readable storage medium for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the above-mentioned method.

[0039] A computer-readable storage medium is characterized by storing computer-executable instructions, which are used to implement the above method when executed.

[0040] Beneficial effects

[0041] The present invention provides a clutter map design method based on the slow decay concept, which has the following characteristics and beneficial effects:

[0042] 1. Rapid clutter map creation processing can quickly complete the clutter map creation within the first few scanning cycles. This can greatly shorten the clutter map creation time when the radar is turned on and the working mode is switched, further improving the clutter suppression capability.

[0043] 2. Adopting the iterative update method of clutter map with fast increase and slow decrease can reduce the clutter traces caused by the fluctuation of echoes from time-varying scattering points, and also reduce the mutual influence of echo data with different beam dwell times, effectively improving the radar's clutter suppression capability and target detection performance;

[0044] 3. Using C code optimization processing, replacing judgment statement instructions with longer execution time with subtraction, shift, bitwise AND and array element reference instructions with shorter execution time can effectively reduce the execution time of C code and further improve the engineering practicality of this method. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The accompanying drawings are only for the purpose of illustrating particular embodiments and are not to be considered limiting of the present invention. Like reference symbols denote like parts throughout the drawings.

[0046] Figure 1 Schematic diagram of slow decay clutter diagram;

[0047] Figure 2 3D clutter map storage logic;

[0048] Figure 3 Schematic diagram of iterative update processing of clutter map;

[0049] Figure 4 Flowchart of the method of the present invention. DETAILED DESCRIPTION

[0050] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only intended to illustrate the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

[0051] The present invention solves the technical problem as follows: first, a multidimensional clutter map is quickly established within one or several radar scanning cycles. Then, within each subsequent coherent accumulation time, the amplitude of each quantization unit in the range-Doppler submap and the clutter map submap of the corresponding azimuth particle is determined. If the amplitude is greater than the amplitude of each quantization unit, a larger update rate is selected for the quantization unit; otherwise, a smaller update rate is selected. Then, a weighted sum is performed on the range-Doppler submap and the clutter map submap, completing an iterative update of the clutter map submap in each accumulation time, with the amplitude increasing rapidly and decreasing slowly.

[0052] Figure 1 This is the schematic diagram of the slow-fading clutter map, which includes two parts: rapid establishment and iterative updating. In the rapid establishment phase, appropriate design parameters, particularly those for the azimuth dimension, are set to enable rapid establishment of the clutter map within the first scan cycle. This rapid establishment is achieved by directly populating the range-Doppler submap with the clutter map submap corresponding to the azimuth particle. Considering that a slightly finer-resolution clutter map can only capture data for most azimuth particles within a single scan cycle, the rapid establishment of the clutter map can be extended to the first few scan cycles (e.g., the first 3-5 scan cycles).

[0053] During the iterative update phase, for each range-Doppler subimage and the clutter map subimage of the corresponding azimuth particle within the coherent accumulation time, the amplitude difference between each range-Doppler quantization unit in the two subimages is calculated. If the difference is greater than zero, the clutter map update rate for that quantization unit takes the larger value, k1; otherwise, the smaller value, k2. The update rates of all quantization units form the coefficient matrix K. The new clutter map subimage is then calculated as: range-Doppler subimage .*K + clutter map subimage .*(IK), where I is the identity matrix and K is the update rate coefficient matrix. The symbol ".*" indicates the multiplication of corresponding matrix elements.

[0054] If the program is written in C, a more optimized implementation for determining amplitude and selecting the update rate is as follows: Subtract the clutter map subgraph from the range-Doppler subgraph to obtain a difference matrix. Since the sign bits of positive and negative numbers are 0 and 1, respectively, the sign bit of each element in the difference matrix is ​​taken (implemented using shifts and bitwise ANDs) to obtain a sign matrix composed of 0s and 1s. The large update rate k1 and the small update rate k2 form an array {k1, k2}, with k1 and k2 indexed as 0 and 1, respectively. The sign matrix then becomes the index matrix for selecting the update rate. In summary, the entire implementation process transforms the relatively time-consuming judgment statement into operations involving subtraction, shifts, bitwise ANDs, and array element references, significantly improving code execution efficiency.

[0055] Figure 2 The storage logic of the three-dimensional clutter map is given. The three dimensions are: azimuth dimension, distance dimension, and frequency dimension. The storage priority is: azimuth dimension > Doppler dimension > distance dimension. The number of azimuth particles is NUM, the number of Doppler channels is M, and the number of distance units is N. The following takes the three-dimensional clutter map of the circular scanning radar as an example, combined with the attached Figure 3 The specific implementation method of slow attenuation processing of clutter images is described in detail. Considering the C code implementation, all the following array index values ​​start from 0:

[0056] Step 1: Create the overall clutter map matrix CMap ((NUM×M)×N matrix) storage space;

[0057] Step 2: Get the distance-Doppler sub-image Data within the current coherent accumulation time rd (M×N matrix);

[0058] Step 3: Calculate the particle index value index at the current beam azimuth az , and the clutter subgraph Data corresponding to the orientation of the particles cm (M×N matrix) row index value index in the clutter map matrix CMap row , where "fix(·)" means rounding down, and the azimuth unit is degrees:

[0059] index az =fix(azimuth / 360*NUM),(0≤index az ≤NUM-1)

[0060] index row =index az ×M,(0≤index row ≤NUM×M-1)

[0061] Step 4: Assuming that the first five scanning cycles after the radar starts working or switches modes are the clutter map fast building time, a simple method to determine whether the current time is within the first five scanning cycles is to see whether it passes through a certain azimuth five times (for example, five times through zero degrees). If so, jump to step 5; if not, jump to step 6.

[0062] Step 5: Convert the range-Doppler subgraph Data rd Directly assign to the clutter graph subgraph Data cm , get the new clutter graph subgraph Data cm_new (M×N matrix), that is, Data cm_new =Data rd , then skip to step 7;

[0063] Step 6: Convert the range-Doppler subgraph Data rd and the clutter graph subgraph Data cm Weighted summation to obtain the new clutter graph subgraph Data cm_new (M×N matrix), specifically perform the following steps:

[0064] Step 6.1: Any element in the range-Doppler subgraph is a ij (0≤i≤M-1, 0≤j≤N-1), any element in the clutter graph subgraph is b ij , find the difference between the two amplitudes diff ij and symbol sign ij , where "Uint32(·)" indicates that the data type is forced to be converted to unsigned 32 bits, "(·)>>31" indicates right shift by 31 bits, and "(·)&1" indicates bitwise AND:

[0065] diff ij =a ij -b ij

[0066] sign ij =(((Uint32)diff ij )>>31)&1

[0067] Step 6.2: Use the large update rate k1 and the small update rate k2 to form an array array = {k1, k2}, that is, array(0) = k1, array(1) = k2, and find any element kk of the update rate matrix K (M×N matrix) ij :

[0068] kk ij =array(sign ij )

[0069] Step 6.3: Update the range-Doppler subgraph Data using the update rate matrix K rd and the clutter graph subgraph Data cm Perform weighted summation to obtain the new clutter graph subgraph Data cm_new , where “.*” indicates the multiplication of corresponding matrix elements, and I is the identity matrix:

[0070] Data cm_new =Data rd .*K+Data cm .*(IK)

[0071] Step 7: Data the new clutter subgraph cm_new Fill in the storage space in the clutter map matrix CMap and complete the update of the clutter map sub-map within the current coherent accumulation time, where "index row :index row +M-1" means from the index row Row to index row +M-1 rows, "1:N" means from column 1 to column N:

[0072] CMap(index row :index row +M-1,1:N)=Data cm_new

[0073] Step 8: Next coherent integration time, obtain the new range-Doppler sub-image Data rd , repeat steps 2 to 8 to complete the update of the clutter map sub-image within each accumulation time until the radar stops scanning.

[0074] The above description is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with this technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present invention, and these modifications or replacements should all be included in the scope of protection of the present invention.

Claims

1. A clutter map design method based on the slow decay concept, characterized by Here are the steps: Step 1: Create a storage space for the overall clutter map matrix CMap; the overall clutter map matrix CMap is a (NUM×M)×N matrix; where NUM is the number of azimuth particles, M is the number of Doppler channels, and N is the number of range units; Step 2: Get the distance-Doppler sub-image Data within the current coherent accumulation time rd ; The distance-Doppler sub-graph Data rd is an M×N matrix; Step 3: Calculate the particle index value index at the current beam azimuth az , and the clutter subgraph Data corresponding to the orientation of the particles cm The row index value index in the clutter map matrix CMap row , the clutter graph subgraph Data cm For an M×N matrix: index az =fix(azimuth / 360*NUM),(0≤index az ≤NUM-1) index row =index az ×M,(0≤index row ≤NUM×M-1) Where fix(·) means rounding down, and the unit of azimuth is degree; Step 4: Assume that the first five scanning cycles after the radar starts working or the mode is switched are the clutter map fast establishment time, determine whether the current time is within the first five scanning cycles, if so, jump to step 5, if not, jump to step 6; Step 5: Convert the range-Doppler subgraph Data rd Directly assign to the clutter graph subgraph Data cm , get the new clutter graph subgraph Data cm_new , namely Data cm_new =Data rd , the clutter graph subgraph Data cm_new is an M×N matrix; then jump to step 7; Step 6: Convert the range-Doppler subgraph Data rd and the clutter graph subgraph Data cm Weighted summation to obtain the new clutter graph subgraph Data cm_new ; Specific steps are as follows: Step 6.1: Any element in the range-Doppler subgraph is a ij , where 0≤i≤M-1, 0≤j≤N-1, any element in the clutter graph subgraph is b ij , find the difference between the two amplitudes diff ij and symbol sign ij : diff ij =a ij -b ij sign ij =(((Uint32)diff ij )>>31)&1 Among them, Uint32(·) means that the data type is forced to be converted to unsigned 32 bits, (·)>>31 means right shift 31 bits, and (·)&1 means bitwise AND 1; Step 6.2: Use the large update rate k1 and the small update rate k2 to form an array array = {k1, k2}, that is, array(0) = k1, array(1) = k2, and find any element kk of the update rate matrix K ij , the update rate matrix K is an M×N matrix: kk ij =array(sign ij ) Step 6.3: Update the range-Doppler subgraph Data using the update rate matrix K rd and the clutter graph subgraph Data cm Perform weighted summation to obtain the new clutter graph subgraph Data cm_new : Data cm_new =Data rd .*K+Data cm .*(I-K) Where ".*" indicates the multiplication of corresponding matrix elements, and I is the identity matrix; Step 7: Data the new clutter subgraph cm_new Fill in its storage space in the clutter map matrix CMap and complete the update of the clutter map sub-map within the current coherent accumulation time: CMap(index row :index row +M-1,1:N)=Data cm_new where index row :index row +M-1 means from the index row Row to index row +M-1 rows, 1:N means from column 1 to column N; Step 8: Next coherent integration time, obtain the new range-Doppler sub-image Data rd , repeat steps 2 to 8 to complete the update of the clutter map sub-image within each accumulation time until the radar stops scanning.

2. The clutter map design method based on the slow decay concept according to claim 1 is characterized in that Implemented using C code.

3. The clutter map design method based on slow decay concept according to claim 1 is characterized in that In step 4, the method for determining whether the current moment is within the first five scanning cycles is to see whether a certain azimuth angle has been passed five times.

4. The clutter map design method based on the slow decay concept according to claim 3 is characterized in that The azimuth is zero degrees.

5. A computer system, characterized in that include: One or more processors, and a computer-readable storage medium for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors are enabled to implement the method of claim 1.

6. A computer-readable storage medium, characterized in that Computer-executable instructions are stored, and when the instructions are executed, they are used to implement the method of claim 1.

Images