Robust space-time adaptive processing method of improved FRACTA structure

Abstract

The invention discloses a robust space-time adaptive processing method for improving an FRACTA structure, and the method comprises the steps: carrying out the robust clutter suppression and target detection in allusion to a non-uniform clutter environment, carrying out the sample selection of echo data through employing a target guiding constraint and energy combined sample selection method, and obtaining initial sample data; performing covariance matrix estimation and adaptive weight calculation on the initial sample data to obtain an initial weight vector; performing primary clutter suppression on the echo data based on the initial weight vector to obtain a residual image; dividing the initial sample data into an initial target sample data set and a clutter sample data set based on the residual map; performing background covariance matrix and adaptive weight calculation on the clutter sample data set to obtain a secondary weight vector; performing clutter suppression on the initial target sample data set based on the secondary weight vector to obtain candidate target sample data; and determining effective target sample data based on the candidate target sample data and a preset signal-clutter-noise ratio threshold. According to the method, robust non-uniform clutter suppression can be realized.

Description

Technical Field

[0001] This invention belongs to the field of spaceborne radar clutter suppression technology, specifically relating to a robust space-time adaptive processing method for an improved FRACTA structure. Background Technology

[0002] Spaceborne radar, due to its wide observation range, has significant application value in fields such as Earth observation and military reconnaissance. However, the unique operating environment of spaceborne radar platforms presents severe challenges to clutter suppression and moving target detection. The wide illumination range and diverse surface cover of spaceborne radar result in a non-uniform and non-stationary distribution of ground clutter received by the radar in the space-time two-dimensional spectrum. Simultaneously, the high altitude and long operating range of the satellite platform cause clutter from different range cells to fold in both the time and space domains, significantly increasing the degrees of freedom of clutter and severely broadening the clutter spectrum. Space-time adaptive processing (STAP), a commonly used clutter suppression technique, typically assumes consistency with the statistical characteristics of clutter in the target cell, uses sample data from a reference cell to calculate the clutter covariance matrix, and then forms an optimal filter through adaptive algorithms such as inverting the sample covariance matrix. However, in practical applications of spaceborne radar systems, the aforementioned non-uniformity of clutter leads to a sharp decline in the performance of traditional STAP algorithms based on inverting the sample covariance matrix. Robust adaptive algorithms have been developed to improve the conditions for matrix inversion through methods such as diagonal loading. However, the selection of the loading amount often relies on experience and lacks the ability to specifically handle non-uniform scattering points, making it difficult to maintain the detection performance of moving targets while suppressing strong non-uniform clutter. Therefore, how to achieve effective clutter suppression and improve the reliability of moving target detection in non-uniform clutter environments is a technical problem that needs to be solved.

[0003] Currently, existing technologies suffer from clutter non-uniformity, leading to deterioration in clutter suppression performance, reduced accuracy in Doppler center and radial velocity estimation, and incomplete sample selection resulting in severe loss of target energy, thus affecting the accuracy of target detection and target parameter estimation. Summary of the Invention

[0004] To address the aforementioned problems in the prior art, this invention provides a robust space-time adaptive processing method for an improved FRACTA structure.

[0005] The technical problem to be solved by this invention is achieved through the following technical solution: This invention provides a robust space-time adaptive processing method for an improved FRACTA structure, comprising: Acquire echo data received by the target's spaceborne radar and complete data preprocessing; A combined target-oriented constraint and energy-based sample selection method is used to perform preliminary sample selection on the preprocessed echo data to obtain initial sample data; The initial weight vector is obtained by estimating the covariance matrix and calculating adaptive weights from the initial sample data. Primary clutter suppression is performed on the preprocessed echo data based on the initial weight vector to obtain a residual map; and the initial sample data is divided into an initial target sample dataset and a clutter sample dataset based on the residual map. The background covariance matrix and adaptive weights are calculated on the clutter sample dataset to obtain the secondary weight vector; Clutter suppression is performed on the initial target sample dataset based on the secondary weight vector to obtain candidate target sample data; Based on the candidate target sample data and the preset signal-to-noise ratio threshold, the effective target sample data is determined.

[0006] This invention provides a robust space-time adaptive processing method for an improved FRACTA structure. For non-uniform clutter, it combines robustly loaded dimension-reduced space-time adaptive processing (STAP) technology with the FRACTA processing framework and sample selection based on joint target guidance constraints and energy, to effectively suppress non-uniform clutter. This method effectively improves the performance degradation problem of traditional STAP technology in non-uniform clutter scenarios of spaceborne radar, increasing the detection probability of moving targets.

[0007] The present invention will now be described in further detail with reference to the accompanying drawings. Attached Figure Description

[0008] Figure 1 This is a flowchart illustrating a robust space-time adaptive processing method for an improved FRACTA structure provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of a robust space-time adaptive processing method for an improved FRACTA structure according to an embodiment of the present invention; Figures 3A to 3D This is a comparative schematic diagram of the clutter suppression results provided in the embodiments of the present invention; Figure 4A and Figure 4B This is a comparative schematic diagram of target detection and relocation results provided in an embodiment of the present invention; Figure 5 This is a schematic diagram comparing the energy before and after clutter suppression provided in an embodiment of the present invention; Figure 6A and Figure 6B This is a schematic diagram of the target distance slice and the target Doppler slice provided in the embodiments of the present invention. Detailed Implementation

[0009] The present invention will be further described in detail below with reference to specific embodiments, but the implementation of the present invention is not limited thereto.

[0010] This invention provides a robust space-time adaptive processing method for an improved FRACTA structure. See also... Figure 1 and Figure 2 The method includes the following steps: S10. Acquire the echo data received by the target's satellite-borne radar and complete the data preprocessing.

[0011] For example, the preprocessed echo data is echo data in the range-Doppler domain.

[0012] S20. The sample selection method of joint target-oriented constraint and energy is used to perform preliminary sample selection on the preprocessed echo data to obtain initial sample data.

[0013] Optionally, step S20 may specifically include: S201. Determine the spatial angle between each sample data in the echo data and the beam center.

[0014] For example, for coordinates on a distance-Doppler map... The data, in the first The value of each spatial channel is Then the pixel corresponding to 3D spatial data vector for:

[0015] Let the target's spatial steering vector be That is, the spatial steering vector corresponding to the beam center, then the angle between the spatial data vector of this coordinate and the spatial steering vector of the target. for:

[0016] Using the above calculation method, the spatial angle between each sample data contained in the echo data and the beam center can be obtained.

[0017] S202. Determine the sample data to be screened based on the spatial angle between each sample data and the beam center and the preset spatial angle threshold.

[0018] For example, for a contaminated target sample, its spatial direction of arrival differs from that of a real target sample. Since they are close, the selection of sample data to be screened is as follows:

[0019] Among them, the preset airspace angle threshold Generally, it can be taken By removing samples smaller than the preset spatial angle threshold, the sample data to be filtered can be obtained.

[0020] S203. Determine the initial sample data based on the sample data to be screened and the preset sample energy threshold.

[0021] For example, based on meeting the preset spatial angle threshold requirement, the sample data to be screened are sorted by energy level, and the data with the highest energy level is selected. Strong samples (i.e., those with a preset sample energy threshold) are used as initial sample data.

[0022] S30. Perform covariance matrix estimation and adaptive weight calculation on the initial sample data to obtain the initial weight vector.

[0023] Optionally, step S30 may specifically include: S301. Calculate the sample covariance matrix of the initial sample data.

[0024] Alternatively, the sample covariance matrix can be represented as:

[0025] in, Represents the sample covariance matrix. The total number of initial sample data. Indicates the first Initial sample data.

[0026] S302. Perform hierarchical loading on the sample covariance matrix to obtain a robustly estimated covariance matrix.

[0027] For example, to avoid estimation errors in the case of small samples, this embodiment performs robust loading on the sample covariance matrix to obtain a robustly estimated covariance matrix. In this embodiment, robust covariance estimation is achieved by using hierarchical loading of the PRI-Stagger covariance matrix.

[0028] Optionally, step S302 may specifically include: S3021. Perform eigenvalue decomposition on the sample covariance matrix to obtain multiple eigenvalues ​​and orthogonal eigenvectors; and arrange the multiple eigenvalues ​​in descending order to obtain the eigenvalue sequence.

[0029] For example, the sequence of eigenvalues ​​is represented as .

[0030] S3022. Based on the preset loading rules and feature value sequence, determine the estimated feature value corresponding to each feature value.

[0031] Optionally, the preset loading rules can be expressed as:

[0032] in, Indicates the first One estimated feature value, Indicates the first 1 eigenvalue, Represents a sequence of eigenvalues. This represents the total number of eigenvalues. This indicates taking the average. Indicates the number of spatial channels. This represents the robust scaling factor.

[0033] Specifically, This indicates the result of the mean calculation.

[0034] S3023. Based on the estimated eigenvalues ​​and orthogonal eigenvectors, calculate the covariance matrix of the robust estimate.

[0035] Alternatively, the covariance matrix of the robust estimate can be expressed as:

[0036] in, The covariance matrix represents the robust estimation. This represents orthogonal eigenvectors.

[0037] S303. Based on the minimum variance criterion of linear constraints, adaptive weight calculation is performed on the covariance matrix of the robust estimate to obtain the initial weight vector.

[0038] Alternatively, the initial weight vector can be expressed as:

[0039] in, Represents the initial weight vector. This indicates target spacetime orientation.

[0040] S40. Perform primary clutter suppression on the preprocessed echo data based on the initial weight vector to obtain a residual map; and divide the initial sample data into an initial target sample dataset and a clutter sample dataset based on the residual map.

[0041] Optionally, step S40 may specifically include: S401. Perform inner product processing on the echo data based on the initial weight vector to obtain the residual map.

[0042] S402. Based on the preset target energy threshold and the energy of each initial sample data in the residual map, the initial sample data is divided into the initial target sample dataset and the clutter sample dataset.

[0043] The initial target sample dataset includes initial sample data with energy greater than a preset target energy threshold, while the clutter sample dataset includes initial sample data with energy less than or equal to a preset target energy threshold.

[0044] For example, a preset target energy threshold is set. ,by To distinguish clutter sample datasets Compared with the initial target sample dataset Energy greater than The initial sample data is incorporated into the initial target sample dataset. Energy less than or equal to The initial sample data was incorporated into the clutter sample dataset. .

[0045] S50. Calculate the background covariance matrix and adaptive weights for the clutter sample dataset to obtain the secondary weight vector.

[0046] For example, the background covariance matrix is ​​calculated on the clutter sample dataset to obtain the background covariance matrix. :

[0047] in, This represents the total number of samples in the clutter sample dataset. Indicating the clutter sample dataset, the first... Initial sample data.

[0048] Secondary weight vector , is represented as:

[0049] S60. Based on the secondary weight vector, clutter suppression is performed on the initial target sample dataset to obtain candidate target sample data.

[0050] For example, for the initial target sample dataset For each initial target sample data, use a secondary weight vector. Clutter suppression is performed to obtain candidate target sample data. .

[0051] S70. Based on the candidate target sample data and the preset signal-to-noise ratio threshold, determine the effective target sample data.

[0052] Optionally, step S70 may specifically include: S701. Based on the initial weight vector and clutter sample dataset, determine the background clutter plus noise energy.

[0053] For example, the energy of background clutter plus noise is represented as .

[0054] S702. Determine the energy of the candidate target based on the secondary weight vector and the candidate target sample data.

[0055] For example, the candidate target energy is represented as .

[0056] S703. Determine the candidate target signal-to-clutter-to-noise ratio corresponding to the candidate target sample data based on the background clutter plus noise energy and the target energy.

[0057] For example, the signal-to-clutter-to-noise ratio (SCNR) of a candidate target is expressed as:

[0058] S704. Candidate target sample data with a signal-to-noise ratio (SNR) greater than a preset SNR threshold are identified as valid target sample data.

[0059] For example, the candidate target signal-to-noise ratio (SCNR) is greater than a preset SCNR threshold. The candidate target sample data were determined as valid target sample data.

[0060] This invention presents a robust space-time adaptive processing method with an improved FRACTA structure. By combining the FRACTA framework with the robustly loaded, dimension-reduced space-time adaptive processing STAP method, it effectively achieves clutter suppression and reliable moving target detection under non-uniform, small-sample conditions. The FRACTA structure employed significantly improves the reliability of sample selection in non-uniform environments, effectively suppresses the interference of strong scattering points on clutter characteristic estimation, and avoids target energy contamination of sample data, thereby improving detection accuracy.

[0061] Furthermore, addressing the challenge of inaccurate covariance matrix estimation under small sample conditions, the robust loading technique employed in this invention effectively improves the stability of covariance matrix estimation, overcomes the decline in clutter suppression performance caused by insufficient samples, and ensures effective clutter suppression in complex environments. Moreover, the method of this invention is simple to implement, has low computational complexity, is easy to implement in engineering, and can meet the actual needs of spaceborne radar systems for real-time signal processing.

[0062] The following simulation experiment further illustrates the robust space-time adaptive processing method for the improved FRACTA structure provided by this invention.

[0063] 1. Simulation conditions The simulation experiment in this embodiment is a simulation experiment of the processing method of the present invention based on the spaceborne radar echo data. The data parameters are configured as follows: orbital altitude 700km, working wavelength 0.03m, pulse repetition rate 12000Hz, transmission bandwidth 1MHz, target signal-to-noise ratio of 40dB, and clutter noise ratio of 50dB.

[0064] 2. Simulation Result Analysis Figure 3A , Figure 3B , Figure 3C and Figure 3D This is a clutter suppression result diagram provided in an embodiment of the present invention, wherein, Figure 3A The distance-Doppler image after suppression using existing technology. Figure 3B This is a distance-Doppler 3D image after suppression using existing technology. Figure 3C This is the distance-Doppler image after suppression by the method of the present invention. Figure 3D This is a three-dimensional distance-Doppler image after suppression using the method of this invention. From... Figure 3C , Figure 3D and Figure 3A , Figure 3B The comparison shows that the clutter suppression of the method of the present invention is more thorough, the suppressed clutter and noise background energy in the output result is more stable, the target energy is higher, which is more conducive to target detection. The simulation results show that the results of the present invention are significantly better than those of the prior art.

[0065] Figure 4A and Figure 4B This is a diagram showing the target detection and relocation results provided in an embodiment of the present invention. Figure 4A The results of target detection and relocation after suppression using existing technologies are as follows. Figure 4B This shows the target detection and relocation results after suppression according to the present invention. (Comparison) Figure 4A and Figure 4B It can be seen that the target is more obvious in the residual image after clutter suppression in this invention (the target has a relatively higher energy).

[0066] Figure 5 This is a comparison diagram of energy before and after clutter suppression provided in an embodiment of the present invention. Figure 5 The paper presents a comparison between the energy before clutter suppression and the energy after suppression using existing techniques and the method of this invention. Compared with existing techniques, the output of this invention is more robust and more conducive to target detection and reducing false alarm rate.

[0067] Figure 6A and Figure 6B This is a target slice image provided in an embodiment of the present invention. Figure 6A For target distance slices, Figure 6B Target Doppler slice image. Figure 6A and Figure 6B The target relative energy (SCNR) is represented by the target range slice and Doppler slice in the residual image. The figure shows two-dimensional slices of the target detected by the prior art and the method of the present invention (blue represents the prior art and red represents the method of the present invention). It can be clearly seen that the processing of the present invention results in less loss of target energy.

[0068] Simulation results demonstrate that the method described in this invention can effectively suppress clutter and reliably detect moving targets under non-uniform, small-sample conditions. Compared with existing technologies, it effectively suppresses the interference of strong scattering points in non-uniform clutter backgrounds on clutter performance, while avoiding target energy contamination of sample data leading to target cancellation. The robust loading technique employed under small-sample conditions effectively improves the robustness of covariance matrix estimation, ensuring the detection capability of spaceborne radar in complex environments.

[0069] It should be noted that the terms "first," "second," etc., are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Rather, they are merely examples of apparatuses and methods consistent with some aspects of the invention.

[0070] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Furthermore, those skilled in the art can combine and integrate the different embodiments or examples described in this specification.

[0071] Although the invention has been described herein in conjunction with various embodiments, those skilled in the art will understand and implement other variations of the disclosed embodiments by reviewing the accompanying drawings and the disclosure in carrying out the claimed invention. In the description of the invention, the word "comprising" does not exclude other components or steps, "a" or "an" does not exclude a plurality, and "a plurality" means two or more, unless otherwise explicitly specified. Furthermore, while different embodiments may describe certain measures, this does not mean that these measures cannot be combined to produce good results.

[0072] The above description, in conjunction with specific preferred embodiments, provides a further detailed explanation of the present invention. It should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various simple deductions or substitutions can be made without departing from the concept of the present invention, and all such modifications and substitutions should be considered within the scope of protection of the present invention.

Claims

1. A robust space-time adaptive processing method for an improved FRACTA structure, characterized in that, include: Acquire echo data received by the target's spaceborne radar and complete data preprocessing; A combined target-oriented constraint and energy-based sample selection method is used to perform preliminary sample selection on the preprocessed echo data to obtain initial sample data; The initial sample data is subjected to covariance matrix estimation and adaptive weight calculation to obtain the initial weight vector; Primary clutter suppression is performed on the preprocessed echo data based on the initial weight vector to obtain a residual map; and the initial sample data is divided into an initial target sample dataset and a clutter sample dataset based on the residual map. The background covariance matrix and adaptive weights are calculated on the clutter sample dataset to obtain the secondary weight vector; Based on the secondary weight vector, secondary clutter suppression is performed on the initial target sample dataset to obtain candidate target sample data; Based on the candidate target sample data and the preset signal-to-noise ratio threshold, the effective target sample data is determined.

2. The robust space-time adaptive processing method for the improved FRACTA structure according to claim 1, characterized in that, The method of selecting samples from the echo data using a combined target-oriented constraint and energy approach yields initial sample data, including: Determine the spatial angle between each sample data in the echo data and the beam center; The sample data to be screened is determined based on the spatial angle between each sample data and the beam center and the preset spatial angle threshold. The initial sample data is determined based on the sample data to be screened and the preset sample energy threshold.

3. The robust space-time adaptive processing method for the improved FRACTA structure according to claim 1, characterized in that, The step of estimating the covariance matrix and calculating the adaptive weights on the initial sample data to obtain the initial weight vector includes: Calculate the sample covariance matrix of the initial sample data; The sample covariance matrix is ​​subjected to hierarchical loading to obtain a robustly estimated covariance matrix; The covariance matrix of the robust estimate is adaptively weighted according to the minimum variance criterion of linear constraints to obtain the initial weight vector.

4. A robust space-time adaptive processing method for an improved FRACTA structure according to claim 3, characterized in that, The step of performing hierarchical loading on the sample covariance matrix to obtain a robustly estimated covariance matrix includes: The sample covariance matrix is ​​subjected to eigenvalue decomposition to obtain multiple eigenvalues ​​and orthogonal eigenvectors; and the multiple eigenvalues ​​are arranged in descending order from largest to smallest to obtain an eigenvalue sequence. Based on the preset loading rules and the feature value sequence, the estimated feature value corresponding to each feature value is determined; Based on the estimated eigenvalues ​​and the orthogonal eigenvectors, the covariance matrix of the robust estimate is calculated.

5. A robust space-time adaptive processing method for an improved FRACTA structure according to claim 4, characterized in that, The sample covariance matrix is ​​expressed as: in, Let the sample covariance matrix be represented. The total number of initial sample data. Indicates the first Initial sample data; The preset loading rule is expressed as follows: in, Indicates the first One estimated feature value, Indicates the first 1 eigenvalue, Represents a sequence of eigenvalues. This represents the total number of eigenvalues. This indicates taking the average. Indicates the number of spatial channels. Indicates a robust scaling factor; The covariance matrix of the robust estimate is expressed as: in, The covariance matrix represents the robust estimation. Represents orthogonal eigenvectors; The initial weight vector is expressed as: in, This represents the initial weight vector. This indicates target spacetime orientation.

6. A robust space-time adaptive processing method for an improved FRACTA structure according to claim 1, characterized in that, The step of performing primary clutter suppression on the echo data based on the initial weight vector to obtain a residual map includes: The echo data is processed by inner product based on the initial weight vector to obtain the residual map.

7. A robust space-time adaptive processing method for an improved FRACTA structure according to claim 6, characterized in that, The process of dividing the initial sample data into an initial target sample dataset and a clutter sample dataset based on the residual map includes: Based on the preset target energy threshold and the energy of each initial sample data in the residual graph, the initial sample data is divided into an initial target sample dataset and a clutter sample dataset. The initial target sample dataset includes initial sample data with energy greater than the preset target energy threshold, and the clutter sample dataset includes initial sample data with energy less than or equal to the preset target energy threshold.

8. A robust space-time adaptive processing method for an improved FRACTA structure according to claim 1, characterized in that, The step of determining valid target sample data based on the candidate target sample data and a preset signal-to-noise ratio threshold includes: Based on the initial weight vector and the clutter sample dataset, the background clutter plus noise energy is determined; Based on the secondary weight vector and the candidate target sample data, the energy of the candidate target is determined; Based on the background clutter plus noise energy and the target energy, the candidate target signal-to-clutter-to-noise ratio corresponding to the candidate target sample data is determined; Candidate target sample data whose signal-to-noise ratio (SNR) is greater than the preset SNR threshold are determined as valid target sample data.

Images