Space-time orthogonal projection sea clutter suppression method for APES-based sky-wave radar

Abstract

This invention provides a space-time orthogonal projection sea clutter suppression method for ground-wave radar based on APES. This invention performs steps such as space-time data matrix construction, clutter sample construction, and space-time orthogonal projection on the received echo data from high-frequency ground-wave radar. This effectively suppresses sea clutter while preserving the target signal, significantly improving the signal-to-clutter ratio. This invention addresses the different spatial and temporal distribution characteristics of targets and sea clutter in high-frequency ground-wave radar echo signals. It suppresses sea clutter through orthogonal projection in the spatial and temporal domains. The processing only uses data from the range cell where the target is located, effectively avoiding the impact of the non-stationary range dimension characteristics of sea clutter on the accuracy of clutter subspace construction. Even when the target is obscured by sea clutter, this method effectively preserves the target spectral points while suppressing sea clutter, significantly improving the signal-to-clutter ratio.

Description

Technical Field

[0001] This invention belongs to the field of sea clutter suppression and target detection technology, and in particular, it is a high-frequency ground-to-ground wave radar sea clutter suppression method based on APES (amplitude and phase estimation) spatiotemporal orthogonal projection. Background Technology

[0002] High-frequency (HF) skywave / groundwave radar is a hybrid high-frequency over-the-horizon radar, combining the advantages of long detection range of skywave radar and stable propagation channel of groundwave radar. Due to ionospheric modulation, the first-order spectral peak of sea clutter in the echo spectrum of HF skywave / groundwave radar is severely broadened. This broadened peak obscures nearby target spectral points, severely limiting the radar's detection performance against slow-moving targets such as ships. To ensure the target detection capability of HF skywave / groundwave radar, it is necessary to suppress the first-order sea clutter.

[0003] Existing sea clutter suppression algorithms for high-frequency radar systems can be mainly divided into four categories: clutter suppression algorithms based on the statistical characteristics of sea clutter, time-frequency domain transform-based sea clutter suppression algorithms, subspace decomposition-based sea clutter suppression algorithms, and sea clutter suppression algorithms based on space-time adaptive processing. Subspace decomposition methods utilize the difference in subspace clustering characteristics between sea clutter and noise background to construct two mutually orthogonal subspaces, and then remove sea clutter through orthogonal projection. Common subspace decomposition algorithms include eigenvalue decomposition (EVD), singular value decomposition (SVD), and higher-order singular value decomposition (HOSVD). To ensure good clutter suppression performance, subspace projection algorithms typically require constructing clutter samples using adjacent range gate sea clutter data. However, when ionospheric disturbances are severe, the correlation between sea clutter at adjacent range gates is not high, which will lead to inaccurate constructed clutter samples and cause the algorithm to fail. Furthermore, for targets submerged in sea clutter, it is often difficult to distinguish them well from sea clutter. Therefore, while the sea clutter is suppressed, the target echo is often significantly attenuated, leading to the failure of subsequent constant false alarm rate (CFAR) detection. How to effectively suppress clutter while preserving the target signal remains a major challenge for subspace-based methods. Summary of the Invention

[0004] The purpose of this invention is to provide a high-frequency ground-to-surface radar sea clutter suppression method based on APES (Optical Persistent Emission System) spatial-temporal orthogonal projection, addressing the problems existing in the prior art. This invention addresses the different spatial-temporal distribution characteristics of targets and sea clutter in high-frequency ground-to-surface radar echo signals. It suppresses sea clutter through orthogonal projection in the spatial-temporal domain, using only the data of the target's range cell during processing. This effectively avoids the impact of the non-stationary range dimension characteristics of sea clutter on the accuracy of clutter subspace construction. Even when the target is obscured by sea clutter, this method effectively preserves the target's spectral points while suppressing sea clutter, significantly improving the signal-to-clutter ratio.

[0005] The technical solution to achieve the objective of this invention is as follows: On the one hand, a method for suppressing sea clutter by space-time orthogonal projection ground-to-ground wave radar based on APES is provided, comprising the following steps:

[0006] Step 1: Calculate the spatiotemporal spectrum using the current range gate space-slow time data of the range cell where the target is located, and construct the spatiotemporal data matrix of the range cell using a sliding window method;

[0007] Step 2: Divide the spectral region to determine the Doppler and spatial information of targets outside the clutter region, and construct a time-domain filter based on the APES principle to remove target echoes outside the clutter region;

[0008] Step 3: Traverse the frequency points within the clutter region to construct clutter samples;

[0009] Step 4: Perform singular value decomposition on the clutter sample to construct the sea clutter subspace;

[0010] Step 5: Using orthogonal projection transformation, the original space-time data matrix is ​​orthogonally projected onto the sea clutter subspace to suppress sea clutter, and it is determined whether the current execution frequency is the target frequency.

[0011] Step 6: If the current execution frequency is the target frequency, update the clutter sample and then return to step 3 to execute the next frequency. If the current execution frequency is not the target frequency, return directly to step 3 to execute the next frequency.

[0012] Step 7: After completing the frequency traversal within the clutter region, restore the clutter-suppressed space-time data to space-slow time data.

[0013] Further, step 1, which involves using the current range gate spatial-slow time data of the target's range cell to calculate the spatiotemporal spectrum and constructing the spatiotemporal data matrix of the range cell using a sliding window method, specifically includes:

[0014] Step 1-1: Let the number of array elements be M and the number of pulses be N. Then the range gate space-slow time data of the range element is an M×N dimensional data matrix, denoted as X. Perform digital beamforming on this data matrix to obtain the space-time spectrum.

[0015] Steps 1-2: For the data matrix X, each time column a of slow time-domain data is selected and stretched to obtain a space-time snapshot y. The sliding length is set to 1. This process is repeated for all data, and L = N - a + 1 is used to obtain the space-time data matrix Y = [y1, y2, ..., y]. L ] represents aM×L dimensional data.

[0016] Further, step 2 involves dividing the spectral region to determine the Doppler and spatial information of targets outside the clutter region, and constructing a time-domain filter based on the APES principle to remove target echoes outside the clutter region. Specifically, this includes:

[0017] Step 2-1: Divide the spectral region according to the space-time spectrum and determine the clutter region range, denoted as F. c ;

[0018] Step 2-2: Obtain the Doppler frequencies f1, f2, ..., f of the target outside the clutter region based on the spatiotemporal spectrum. i , where i is the number of targets outside the clutter zone, f i The doppler frequency of the i-th target outside the clutter zone is denoted by f; for a signal with doppler frequency f, its time-domain steering vector s is denoted by f. t (f) is:

[0019] s t (f)=[1,exp(j2πf),…,exp(j2π(L-1)f)] T (1)

[0020] in,[·] T This is a transpose operation;

[0021] Steps 2-3: Construct a time-domain filter based on the APES principle to remove target echoes outside the clutter region:

[0022] The target signal subspace outside the clutter region is represented as Q1:

[0023]

[0024] Where Δf represents the frequency resolution of the high-frequency ground-to-surface radar;

[0025] Based on the APES principle, the time-domain orthogonal projection matrix P1 is constructed as follows:

[0026] P1 = I - Q1 * (Q1 T Q1 * ) -1 Q1 T (3)

[0027] in,[·] * Indicates complex conjugation, [·]-1 This indicates finding the inverse;

[0028] Then, the clutter sample after removing the target signal outside the clutter region. Represented as:

[0029]

[0030] Furthermore, step 3, which involves traversing the frequency points within the clutter region to construct clutter samples, specifically includes:

[0031] The frequency points within the clutter region are traversed, and the target signal subspace constructed in each execution is Q2:

[0032] Q2 = [s t (f in -Δf / 2),s t (f in ),s t (f in +Δf / 2)] (5)

[0033] Among them, f in This indicates the frequency currently in operation within the clutter region, f in ∈F c ;

[0034] At this point, the time-domain orthogonal projection matrix P2 is:

[0035] P2 = I - Q2 * (Q2 T Q2 * ) -1 Q2 T (6)

[0036] Then the clutter sample at this time Represented as:

[0037]

[0038] Further, step 4, which involves performing singular value decomposition on the clutter sample to construct the sea clutter subspace, specifically includes:

[0039] clutter sample Decomposed into:

[0040]

[0041] Among them, ∑=diag(σ1,σ2,…σ c ), σ1≥σ2≥…≥σ c ≥0 represents a singular value matrix, c = min(aM,L), U = [u1,u2,…,u aM ] and V T=[v1,v2,…,v L Each contains the corresponding left and right singular vectors;

[0042] The large singular values ​​in the ∑ contain most of the power of the sea clutter, while the remaining small singular values ​​represent interference and noise. The singular vectors corresponding to the large singular values ​​are then spanned into the sea clutter subspace.

[0043] The method for distinguishing between large and small singular values ​​is as follows: Arrange all singular values ​​from largest to smallest into a singular value spectrum, and select the point with the largest change in slope between two adjacent points as the boundary between large and small singular values; Let the number of large singular values ​​be k, k < aM, then the sea clutter subspace P C =[u1,u2,…,u k ].

[0044] Further, step 5, which involves orthogonally projecting the original space-time data matrix onto the sea clutter subspace using orthogonal projection transformation to suppress sea clutter and determining whether the current execution frequency is the target frequency, specifically includes:

[0045] Step 5-1, from the sea clutter subspace P C The projection operator P can be obtained. t for:

[0046] P t =P C (P C H P C ) -1 P C H (9)

[0047] in,[·] H Indicates conjugate transpose;

[0048] Step 5-2: Orthogonally project the space-time data matrix onto the sea clutter subspace to obtain the data matrix Y after suppressing sea clutter. S :

[0049] Y S =YP t Y (10)

[0050] Step 5-3, the data matrix Y after suppressing sea clutter S Restore the data to the empty-slow time domain and perform spatiotemporal spectrum estimation;

[0051] Step 5-4: If there is no obvious peak at the current execution frequency in the space-time spectrum after clutter suppression, i.e., no peak that meets the conditions, then the current execution frequency is not the target frequency; otherwise, the current execution frequency is the target frequency, and the Doppler frequency and azimuth of the target can be read based on the location of the peak.

[0052] Furthermore, the update of clutter samples in step 6 specifically involves:

[0053] make:

[0054]

[0055] Furthermore, after completing the frequency point traversal within the clutter region in step 7, the clutter-suppressed space-time data is restored to space-slow time data, specifically as follows:

[0056] Considering that the boundary point of the clutter region happens to be the target frequency, and that the frequency traversal within the clutter region is performed from left to right, the first frequency point on the right outside the clutter region is taken as the last frequency point at the end of the traversal. At this point, the clutter sample has been updated multiple times, and all target signals inside and outside the clutter region have been filtered out. The matrix Y obtained in steps 3 to 5 is then executed for the last time. S This is the final result after suppressing sea clutter;

[0057] Then Y can be S Restore the data to empty-slow time and perform spatiotemporal spectrum analysis

[0058] On the other hand, a space-time orthogonal projection ground-to-ground wave radar sea clutter suppression system based on APES is provided, the system comprising the following sequentially executed functions:

[0059] The first module is used to calculate the spatiotemporal spectrum using the current range gate space-slow time data of the range cell where the target is located, and to construct the spatiotemporal data matrix of the range cell using a sliding window method;

[0060] The second module is used to divide the spectrum region, determine the Doppler and spatial information of targets outside the clutter region, and construct a time-domain filter based on the APES principle to remove the echoes of targets outside the clutter region.

[0061] The third module is used to traverse the frequency points within the clutter region and construct clutter samples;

[0062] The fourth module is used to perform singular value decomposition on the clutter samples to construct the sea clutter subspace;

[0063] The fifth module is used to orthogonally project the original space-time data matrix onto the sea clutter subspace using orthogonal projection transformation to suppress sea clutter, and to determine whether the current execution frequency is the target frequency.

[0064] The sixth module is used to update the clutter sample if the current execution frequency is the target frequency, and then return to the third module to execute the next frequency. If the current execution frequency is not the target frequency, it directly returns to the third module to execute the next frequency.

[0065] The seventh module is used to restore the clutter-suppressed space-time data to space-slow time data after frequency point traversal in the clutter region is completed.

[0066] Compared with the prior art, the significant advantages of this invention are:

[0067] 1) The high-frequency ground-to-ground wave radar sea clutter suppression method based on APES spatial-temporal orthogonal projection proposed in this invention can effectively suppress sea clutter by using orthogonal projection, based on the different characteristics of the spatial and temporal distribution of the target and sea clutter.

[0068] 2) The high-frequency ground-to-ground wave radar sea clutter suppression method based on APES spatiotemporal orthogonal projection proposed in this invention constructs clutter samples using time-domain projection according to the APES principle, so that the clutter samples are as close as possible to the clutter signal without containing the target signal.

[0069] 3) The high-frequency ground-to-ground wave radar sea clutter suppression method based on APES spatiotemporal orthogonal projection proposed in this invention can suppress clutter using only single range data, which can effectively reduce the amount of data to be processed and effectively avoid the influence of the range dimension non-stationary characteristics of sea clutter on the accuracy of clutter subspace construction.

[0070] 4) The high-frequency ground-to-ground wave radar sea clutter suppression method based on APES spatiotemporal orthogonal projection proposed in this invention adopts a frequency point traversal method in the clutter region, which can effectively preserve the target frequency point and suppress sea clutter, and is suitable for scenarios with multiple targets.

[0071] The present invention will now be described in further detail with reference to the accompanying drawings. Attached Figure Description

[0072] Figure 1 This is a flowchart of the APES-based method for suppressing sea clutter in space-time orthogonal projection radar.

[0073] Figure 2 This is a schematic diagram of the construction of the space-time data matrix in this invention.

[0074] Figure 3 This is the spatiotemporal spectrum before sea clutter suppression in an embodiment of the present invention.

[0075] Figure 4 This is the space-time spectrum obtained when the frequency point being traversed in the clutter region is not the target frequency point.

[0076] Figure 5 This is the space-time spectrum obtained when the frequency point being traversed in the clutter region is the target frequency point.

[0077] Figure 6 The spatiotemporal spectrum after integrating all target frequency information and suppressing sea clutter using this method.

[0078] Figure 7 This is a comparison of the Doppler spectra of target 1 before and after sea clutter suppression in an example of the present invention.

[0079] Figure 8 This is a comparison chart showing the signal-to-clutter ratio improvement after three methods were used to suppress sea clutter at different Doppler frequencies in the simulated target of this invention. Detailed Implementation

[0080] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0081] It should be noted that if the embodiments of the present invention involve directional indicators (such as up, down, left, right, front, back, etc.), the directional indicators are only used to explain the relative positional relationship and movement of the components in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indicators will also change accordingly.

[0082] Furthermore, if the embodiments of this invention involve descriptions such as "first" or "second," these descriptions are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined with "first" or "second" may explicitly or implicitly include at least one of those features. Additionally, the technical solutions of the various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. If the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by this invention.

[0083] In one embodiment, combined Figure 1 This paper presents a method for suppressing sea clutter from space-time orthogonal projection ground-to-ground wave radar based on APES, comprising the following steps:

[0084] Step 1, combined Figure 2 The spatiotemporal spectrum is calculated using the current range gate space-slow time data of the range cell containing the target, and the spatiotemporal data matrix of the range cell is constructed using a sliding window method; specifically including:

[0085] Step 1-1: Let the number of array elements be M and the number of pulses be N. Then the range gate space-slow time data of the range element is an M×N dimensional data matrix, denoted as X. Perform digital beamforming on this data matrix to obtain the space-time spectrum.

[0086] Steps 1-2: For the data matrix X, each time column a of slow time-domain data is selected and stretched to obtain a space-time snapshot y. The sliding length is set to 1. This process is repeated for all data, and L = N - a + 1 is used to obtain the space-time data matrix Y = [y1, y2, ..., y]. L ] represents aM×L dimensional data.

[0087] Step 2 involves dividing the spectral region to determine the Doppler and spatial information of targets outside the clutter region, and constructing a time-domain filter based on the APES principle to remove target echoes outside the clutter region; combined with... Figure 3 Specifically, it includes:

[0088] Step 2-1: Divide the spectral region according to the space-time spectrum and determine the clutter region range, denoted as F. c ;

[0089] Step 2-2: Obtain the Doppler frequencies f1, f2, ..., f of the target outside the clutter region based on the spatiotemporal spectrum. i , where i is the number of targets outside the clutter zone, f i The doppler frequency of the i-th target outside the clutter zone is represented; since the number of time-domain snapshots of the constructed space-time data matrix is ​​L, for a signal with a doppler frequency f, its time-domain steering vector s t (f) is:

[0090] s t (f)=[1,exp(j2πf),…,exp(j2π(L-1)f)] T (12)

[0091] in,[·] T This is a transpose operation;

[0092] Steps 2-3: Construct a time-domain filter based on APES principle to remove target echoes outside the clutter zone: Since the high-frequency ground wave radar adopts bistatic mode and the transmission channel is modulated by the ionosphere, the Doppler of the target signal may also be broadened. In order to ensure that the clutter sample does not contain the target signal, a broadening model is used to construct the target signal subspace.

[0093] The target signal subspace outside the clutter region is represented as Q1:

[0094]

[0095] Where Δf represents the frequency resolution of the high-frequency ground-to-surface radar;

[0096] Based on the APES principle, the time-domain orthogonal projection matrix P1 is constructed as follows:

[0097] P1 = I - Q1 * (Q1 T Q1 * )-1 Q1 T (14)

[0098] in,[·] * Indicates complex conjugation, [·] -1 This indicates finding the inverse;

[0099] Then, the clutter sample after removing the target signal outside the clutter region. Represented as:

[0100]

[0101] Step 3: Traverse the frequency points within the clutter region to construct clutter samples; specifically including:

[0102] The frequency points within the clutter region are traversed, and the target signal subspace constructed in each execution is Q2:

[0103] Q2=[s t (f in -Δf / 2),s t (f in ),s t (f in +Δf / 2)] (16)

[0104] Among them, f in This indicates the frequency currently in operation within the clutter region, f in ∈F c ;

[0105] At this point, the time-domain orthogonal projection matrix P2 is:

[0106] P2 = I - Q2 * (Q2 T Q2 * ) -1 Q2 T (17)

[0107] Then the clutter sample at this time Represented as:

[0108]

[0109] Step 4: Perform singular value decomposition on the clutter sample to construct the sea clutter subspace; specifically including:

[0110] clutter sample Decomposed into:

[0111]

[0112] Among them, ∑=diag(σ1,σ2,…σ c ), σ1≥σ2≥…≥σc ≥0 represents a singular value matrix, c = min(aM,L), U = [u1,u2,…,u aM ] and V T =[v1,v2,…,v L Each contains the corresponding left and right singular vectors;

[0113] Since the energy of sea clutter is much greater than that of noise and interference, the large singular values ​​in ∑ will contain most of the power of sea clutter, while the remaining small singular values ​​represent interference and noise, etc.

[0114] The method for distinguishing between large and small singular values ​​is as follows: Arrange all singular values ​​from largest to smallest into a singular value spectrum, and select the point with the largest change in slope between two adjacent points as the boundary between large and small singular values; Let the number of large singular values ​​be k, k < aM, then the sea clutter subspace P C =[u1,u2,…,u k ].

[0115] Step 5: Using orthogonal projection transformation, the original space-time data matrix is ​​orthogonally projected onto the sea clutter subspace to suppress sea clutter, and it is determined whether the current execution frequency is the target frequency; specifically including:

[0116] Step 5-1, from the sea clutter subspace P C The projection operator P can be obtained. t for:

[0117] P t =P C (P C H P C ) -1 P C H (20) in,[·] H Indicates conjugate transpose;

[0118] Step 5-2: Orthogonally project the space-time data matrix onto the sea clutter subspace to obtain the data matrix Y after suppressing sea clutter. S :

[0119] Y S =YP t Y (21)

[0120] Step 5-3, the data matrix Y after suppressing sea clutter S Restore the data to the empty-slow time domain and perform spatiotemporal spectrum estimation;

[0121] Step 5-4: If there is no obvious peak at the current execution frequency in the space-time spectrum after clutter suppression, i.e., no peak that meets the conditions, then the current execution frequency is not the target frequency; otherwise, the current execution frequency is the target frequency, and the Doppler frequency and azimuth of the target can be read based on the location of the peak.

[0122] Step 6: If the current execution frequency is the target frequency, update the clutter sample, then return to step 3 to execute the next frequency. If the current execution frequency is not the target frequency, directly return to step 3 to execute the next frequency. Specifically, updating the clutter sample in step 6 involves:

[0123] make:

[0124]

[0125] Step 7: After completing the frequency traversal within the clutter region, restore the clutter-suppressed space-time data to space-slow time data, specifically as follows:

[0126] Considering that the boundary point of the clutter region happens to be the target frequency, and that the frequency traversal within the clutter region is performed from left to right, the first frequency point on the right outside the clutter region is taken as the last frequency point at the end of the traversal. At this point, the clutter sample has been updated multiple times, and all target signals inside and outside the clutter region have been filtered out. The matrix Y obtained in steps 3 to 5 is then executed for the last time. S This is the final result after suppressing sea clutter;

[0127] Then Y can be S Restore the data to empty-slow time and perform spatiotemporal spectrum analysis.

[0128] In one embodiment, a space-time orthogonal projection ground-to-ground wave radar sea clutter suppression system based on APES is provided, the system comprising the following sequentially executed functions:

[0129] The first module is used to calculate the spatiotemporal spectrum using the current range gate space-slow time data of the range cell where the target is located, and to construct the spatiotemporal data matrix of the range cell using a sliding window method;

[0130] The second module is used to divide the spectrum region, determine the Doppler and spatial information of targets outside the clutter region, and construct a time-domain filter based on the APES principle to remove the echoes of targets outside the clutter region.

[0131] The third module is used to traverse the frequency points within the clutter region and construct clutter samples;

[0132] The fourth module is used to perform singular value decomposition on the clutter samples to construct the sea clutter subspace;

[0133] The fifth module is used to orthogonally project the original space-time data matrix onto the sea clutter subspace using orthogonal projection transformation to suppress sea clutter, and to determine whether the current execution frequency is the target frequency.

[0134] The sixth module is used to update the clutter sample if the current execution frequency is the target frequency, and then return to the third module to execute the next frequency. If the current execution frequency is not the target frequency, it directly returns to the third module to execute the next frequency.

[0135] The seventh module is used to restore the clutter-suppressed space-time data to space-slow time data after frequency point traversal in the clutter region is completed.

[0136] Specific limitations regarding the APES-based space-time orthogonal projection terrestrial-wave radar sea clutter suppression system can be found in the limitations section above regarding the APES-based space-time orthogonal projection terrestrial-wave radar sea clutter suppression method, and will not be repeated here. Each module in the aforementioned APES-based space-time orthogonal projection terrestrial-wave radar sea clutter suppression system can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the computer device's memory as software, allowing the processor to call and execute the corresponding operations of each module.

[0137] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements:

[0138] Step 1: Calculate the spatiotemporal spectrum using the current range gate space-slow time data of the range cell where the target is located, and construct the spatiotemporal data matrix of the range cell using a sliding window method;

[0139] Step 2: Divide the spectral region to determine the Doppler and spatial information of targets outside the clutter region, and construct a time-domain filter based on the APES principle to remove target echoes outside the clutter region;

[0140] Step 3: Traverse the frequency points within the clutter region to construct clutter samples;

[0141] Step 4: Perform singular value decomposition on the clutter sample to construct the sea clutter subspace;

[0142] Step 5: Using orthogonal projection transformation, the original space-time data matrix is ​​orthogonally projected onto the sea clutter subspace to suppress sea clutter, and it is determined whether the current execution frequency is the target frequency.

[0143] Step 6: If the current execution frequency is the target frequency, update the clutter sample and then return to step 3 to execute the next frequency. If the current execution frequency is not the target frequency, return directly to step 3 to execute the next frequency.

[0144] Step 7: After completing the frequency traversal within the clutter region, restore the clutter-suppressed space-time data to space-slow time data.

[0145] For specific limitations on each step, please refer to the limitations on the APES-based space-time orthogonal projection ground-to-ground wave radar sea clutter suppression method mentioned above, which will not be repeated here.

[0146] As a specific example, in one embodiment, the invention is further verified and described in detail. Several existing methods are selected for comparison, demonstrating that the present method has a better signal-to-noise ratio improvement.

[0147] In this embodiment, measured clutter data plus simulated targets are used. The number of array elements is 8, the number of pulses is 600, and 50 columns are selected at a time during sliding windowing. Simulated target 1 has an azimuth of 110° and a Doppler frequency of -0.44Hz; simulated target 2 has an azimuth of 84° and a Doppler frequency of 0.60Hz. Based on the spatiotemporal spectrum obtained in step 1 ( Figure 3 It can be seen that the Doppler range of the negative sea clutter region is -0.45Hz to -0.35Hz, and the Doppler range of the positive sea clutter region is 0.27Hz to 0.37Hz. Target 1 is submerged in the negative sea clutter region and is difficult to detect and distinguish. Target 2 is outside the sea clutter region and has a Doppler frequency of 0.60Hz. Figure 4 This demonstrates the case where the frequency point being traversed during clutter frequency traversal is not the target frequency point. Figure 5 This demonstrates the case where the frequency point being traversed during clutter frequency traversal is the target frequency point. Figure 6 The image shows the spatiotemporal spectrum after all steps have been completed and sea clutter has been suppressed. The target is circled in white. The target's azimuth and Doppler information are consistent with the settings, proving the effectiveness of this algorithm.

[0148] Figure 7 The image shows a comparison of the Doppler spectra at angle 1 of the target before and after sea clutter suppression. Two existing methods are used for comparison. Method 1 represents the high-frequency ground-to-surface radar sea clutter suppression method based on APES spatiotemporal orthogonal projection described in this invention; Method 2 represents the High-Order Singular Value Decomposition (HOSVD) method; and Method 3 represents the conventional spatiotemporal orthogonal projection (ST-OP) method. From the perspective of clutter suppression, the method described in this invention, in addition to preserving the target, achieves clutter suppression of more than 13dB in both positive and negative clutter regions, while the other two methods both leave clutter residue in the clutter region, especially the ST-OP algorithm, which has more clutter residue in the positive clutter region.

[0149] Figure 8 The signal-to-clutter ratio (SNR) improvement after sea clutter suppression using three methods at different Doppler frequencies is presented in the simulation results. It can be seen that regardless of whether the target is outside or inside the sea clutter region, the SNR can be improved by more than 12 dB using the method described in this invention, proving the stability of the algorithm. In contrast, the HOSVD and ST-OP algorithms only improve the SNR by about 7 dB after clutter suppression, which is lower than the method described in this invention.

[0150] In summary, compared with traditional high-frequency ground-to-surface radar sea clutter suppression methods, this invention addresses the different spatial and temporal distribution characteristics of high-frequency ground-to-surface radar echo signals in the target and sea clutter. It utilizes the APES principle to construct clutter samples through temporal projection, and then suppresses sea clutter through orthogonal projection in the spatial and temporal domains. The sea clutter suppression method is effective, feasible, reliable, and achieves a higher signal-to-clutter ratio.

[0151] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention without departing from its spirit and scope should be included within the protection scope of the present invention.

Claims

1. A method for suppressing sea clutter in space-time orthogonal projection radar based on APES, characterized in that, Includes the following steps: Step 1: Calculate the spatiotemporal spectrum using the current range gate space-slow time data of the range cell where the target is located, and construct the spatiotemporal data matrix of the range cell using a sliding window method; Step 2: Divide the spectral region to determine the Doppler and spatial information of targets outside the clutter region, and construct a time-domain filter based on the APES principle to remove target echoes outside the clutter region; Step 3: Traverse the frequency points within the clutter region to construct clutter samples; Step 4: Perform singular value decomposition on the clutter sample to construct the sea clutter subspace; Step 5: Using orthogonal projection transformation, the original space-time data matrix is ​​orthogonally projected onto the sea clutter subspace to suppress sea clutter, and it is determined whether the current execution frequency is the target frequency. Step 6: If the current execution frequency is the target frequency, update the clutter sample and then return to step 3 to execute the next frequency. If the current execution frequency is not the target frequency, return directly to step 3 to execute the next frequency. Step 7: After completing the frequency traversal within the clutter region, restore the clutter-suppressed space-time data to space-slow time data; Step 2 involves dividing the spectral region to determine the Doppler and spatial information of targets outside the clutter region, and constructing a time-domain filter based on the APES principle to remove target echoes outside the clutter region. Specifically, this includes: Step 2-1: Divide the spectral region according to the spatiotemporal spectrum and determine the clutter region range, denoted as . ; Step 2-2: Obtain the Doppler frequency of the target outside the clutter region based on the spatiotemporal spectrum. , The number of targets outside the clutter zone. Let represent the Doppler frequency of the target outside the clutter region; for a Doppler frequency of , The signal, its time-domain steering vector for: （1） in, This is a transpose operation; Steps 2-3: Construct a time-domain filter based on the APES principle to remove target echoes outside the clutter region: The target signal subspace outside the clutter region is represented as: : （2） in, This indicates the frequency resolution of the high-frequency ground-wave radar; Based on the APES principle, construct the time-domain orthogonal projection matrix. for: （3） in, Indicates complex conjugation. This indicates finding the inverse; Then, the clutter sample after removing the target signal outside the clutter region. Represented as: (4) Step 3, which involves traversing the frequency points within the clutter region to construct clutter samples, specifically includes: The frequency points within the clutter region are traversed, and the target signal subspace constructed in each execution is... : (5) in, This indicates the frequency currently in operation within the clutter region. ; At this point, the time-domain orthogonal projection matrix for: (6) Then the clutter sample at this time Represented as: (7) The clutter sample update described in step 6 is specifically as follows: make: (8) Step 1, which involves using the current range gate space-slow time data of the target's range cell to calculate the spatiotemporal spectrum and constructing the spatiotemporal data matrix of the range cell using a sliding window method, specifically includes: Step 1-1, let the number of array elements be M and the number of pulses be N, then the range gate space-slow time data of the range element is: A 3D data matrix, denoted as Digital beamforming is then performed on the data matrix to obtain the spatiotemporal spectrum; Steps 1-2, for the data matrix Each time you select Straightening the slow time-domain data yields a space-time snapshot. Set the sliding length to 1, slide over all data sequentially, and record... The space-time data matrix is ​​obtained. , for Dimensional data.

2. The method for suppressing sea clutter by space-time orthogonal projection ground-to-ground wave radar based on APES according to claim 1, characterized in that, Step 4, which involves performing singular value decomposition on the clutter sample to construct the sea clutter subspace, specifically includes: clutter sample Decomposed into: (9) in, It is a singular value matrix. , and Each contains the corresponding left and right singular vectors; Large singular values ​​contain most of the power of sea clutter, while the remaining small singular values ​​represent interference and noise. The singular vectors corresponding to the large singular values ​​are then spanned into a sea clutter subspace. The method for distinguishing between large and small singular values ​​is as follows: Arrange all singular values ​​from largest to smallest into a singular value spectrum, and select the point with the largest change in slope between adjacent points as the boundary between large and small singular values; the number of large singular values ​​is denoted as... Then the sea clutter subspace .

3. The method for suppressing sea clutter by space-time orthogonal projection ground-to-ground wave radar based on APES according to claim 2, characterized in that, Step 5 describes the use of orthogonal projection transformation to orthogonally project the original space-time data matrix onto the sea clutter subspace to suppress sea clutter, and to determine whether the current execution frequency is the target frequency. Specifically, this includes: Step 5-1, from the sea clutter subspace Obtain the projection operator for: (10) in, Indicates conjugate transpose; Step 5-2: Orthogonally project the space-time data matrix onto the sea clutter subspace to obtain the data matrix after suppressing sea clutter. : (11) Step 5-3, the data matrix after suppressing sea clutter. Restore the data to the empty-slow time domain and perform spatiotemporal spectrum estimation; Step 5-4: If there is no obvious peak at the current execution frequency in the space-time spectrum after clutter suppression, i.e., no peak that meets the conditions, then the current execution frequency is not the target frequency; otherwise, the current execution frequency is the target frequency, and the Doppler frequency and azimuth of the target are read according to the position of the peak.

4. The method for suppressing sea clutter by space-time orthogonal projection ground-to-ground wave radar based on APES according to claim 3, characterized in that, After completing the frequency traversal within the clutter region in step 7, the clutter-suppressed space-time data is restored to space-slow time data, specifically as follows: Considering that the boundary point of the clutter region happens to be the target frequency, and that the frequency traversal within the clutter region is performed from left to right, the first frequency point on the right outside the clutter region is taken as the last frequency point at the end of the traversal. At this point, the clutter sample has been updated multiple times, and all target signals inside and outside the clutter region have been filtered out. The matrix obtained in steps 3 to 5 is then executed for the last time. This is the final result after suppressing sea clutter; Soon Restore the data to empty-slow time and perform spatiotemporal spectrum analysis.

5. A space-time orthogonal projection ground-to-ground wave radar sea clutter suppression system based on the method of any one of claims 1 to 4, characterized in that, The system includes sequential execution of: The first module is used to calculate the spatiotemporal spectrum using the current range gate space-slow time data of the range cell where the target is located, and to construct the spatiotemporal data matrix of the range cell using a sliding window method; The second module is used to divide the spectrum region, determine the Doppler and spatial information of targets outside the clutter region, and construct a time-domain filter based on the APES principle to remove the echoes of targets outside the clutter region. The third module is used to traverse the frequency points within the clutter region and construct clutter samples; The fourth module is used to perform singular value decomposition on the clutter samples to construct the sea clutter subspace; The fifth module is used to orthogonally project the original space-time data matrix onto the sea clutter subspace using orthogonal projection transformation to suppress sea clutter, and to determine whether the current execution frequency is the target frequency. The sixth module is used to update the clutter sample if the current execution frequency is the target frequency, and then return to the third module to execute the next frequency. If the current execution frequency is not the target frequency, it directly returns to the third module to execute the next frequency. The seventh module is used to restore the clutter-suppressed space-time data to space-slow time data after frequency point traversal in the clutter region is completed.

6. A computer device, comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that, When the processor executes the computer program, it implements the method according to any one of claims 1 to 4.

Images