A method for expanding a null of a distributed array reconfigurable super freedom degree

Abstract

The application discloses a kind of distributed array reconstruction super degree of freedom interference null broadening method, interpolation is carried out to virtual array element to distributed array, wherein transformation region contains all signals;Solving the transformation matrix of each region;According to the transformation matrix of solving, the covariance matrix of virtual array is constructed, and color noise is removed;Solving transformation error matrix, and further solving error covariance matrix;Covariance matrix and guide vector under error constraint are constructed;Spatial spectrum function is constructed, and spatial spectrum function is carried out space domain extension using window function;Add virtual interference spectrum to the spatial spectrum function after space domain extension;Reconstruction interference noise covariance matrix;According to MNV criterion, optimization weight is calculated.The application is suitable for distributed array, reduces inherent cost, increases array system degree of freedom by virtual array element interpolation, effectively broadens null while guaranteeing main lobe gain, solves the problem that interference source deviates from preset direction of arrival, improves array anti-interference performance.

Description

Technical Field

[0001] This invention belongs to the field of communication technology, specifically relating to a method for widening the interference null trap of distributed array reconfiguration. Background Technology

[0002] In array signal processing, adaptive beamforming effectively suppresses interference and enhances the desired signal by creating nulls in the direction of interference. However, in practical applications, interference often deviates from the preset null position due to factors such as interference angle disturbances, platform jitter, and channel errors. This leads to a sharp deterioration in the output signal-to-interference-plus-noise ratio (SNR), a significant decrease in suppression performance, and in severe cases, even system failure. This vulnerability is particularly pronounced in dynamic or high-precision applications such as airborne radar and communication anti-jamming, severely limiting the practical effectiveness of traditional adaptive beamforming technology.

[0003] Null widening techniques aim to broaden and flatten null regions near interference. However, traditional methods, such as covariance matrix tapering, require manual modification of the covariance matrix or the addition of constraints. This inevitably results in a loss of main lobe performance and an increase in side lobes, and fails to suppress interference with excessive degrees of freedom. In distributed arrays, this problem is even more pronounced: the array's inherent grating lobes introduce multiple high-gain side lobes, further increasing the complexity of null design; simultaneously, the limited number of array elements leads to insufficient system degrees of freedom, making it difficult to generate wide and deep nulls for multiple interferences while suppressing grating lobes and maintaining the main lobe, resulting in a trade-off between various performance aspects. Summary of the Invention

[0004] To overcome the shortcomings of existing technologies, this invention provides a method for reconstructing a distributed array with high degrees of freedom interference null widening. The method involves virtual element interpolation of the distributed array, where the transformed region contains all signals; solving the transformation matrix for each region; constructing the covariance matrix of the virtual array based on the solved transformation matrix and removing colored noise; solving the transformation error matrix and further solving the error covariance matrix; constructing the covariance matrix and steering vector under error constraints; constructing a spatial spectrum function and using a window function to spatially expand the spatial spectrum function; adding a virtual interference spectrum to the spatially expanded spatial spectrum function; reconstructing the interference plus noise covariance matrix; and calculating the optimization weights according to the MNV criterion. Compared with traditional null widening schemes, this invention is applicable to distributed arrays, requires fewer array elements for the same array aperture, reduces inherent costs, increases the degrees of freedom of the array system through virtual element interpolation, effectively widens nulls while maintaining main lobe gain, solves the problem of interference sources deviating from the preset direction of arrival, and improves the array's anti-interference performance.

[0005] The technical solution adopted by this invention to solve its technical problem is as follows: Step 1: Perform virtual element interpolation on the distributed array, assuming... For the actual element spacing, The actual number of array elements. This refers to the virtual element spacing; Step 2: Based on the direction of the interference, solve for the corresponding transformation matrix in each interference region; first, divide a certain observation area, assuming the signal falls within the region. Within, the area Divide evenly into:

[0006] in , for The left and right boundaries, Insertion step size; When multiple disturbances fall into different regions in space, it is necessary to divide the space into multiple observation sub-regions. The actual array manifold matrix and virtual array manifold matrix of each sub-region are represented as follows:

[0007]

[0008] and For actual array antennas and virtual array antennas respectively Directional guidance vector, Indicates the number of interferences.

[0009] Suppose there exists a fixed transformation relationship between the real array manifold matrix and the virtual array manifold matrix, such that:

[0010] Transformation matrix Guide the vector matrix of real and virtual array antennas and The following conditions must be met:

[0011] When the number of transformation points is greater than the actual number of array antenna elements and A is full rank, the virtual transformation matrix can be obtained from the above equation as follows:

[0012] Step 3: Based on the transformation matrix of each sub-region, further calculate the virtual array covariance matrix of the corresponding region as follows:

[0013] in, This represents the covariance matrix of the actual array received data. Represents the covariance matrix of the desired signal. Indicates noise power; After the virtual transformation, the original white noise of the array antenna is contaminated with colored noise. It is necessary to pre-whiten the colored noise and then synthesize the virtual covariance matrix of each sub-region using an arithmetic mean method. The resulting virtual covariance matrix after removing the colored noise is:

[0014] Step 4: Introduce the transformation error constraint method. First, define the transformation error matrix:

[0015] Within any transforming region, angle The transformation error at point is expressed as:

[0016] Known in the transformed region Inside, assuming the weights obtained after virtual interpolation are... The weighted transformation error output after virtual interpolation is:

[0017] The output power of the transformation error is then expressed as:

[0018] In the formula, , is defined as the transformation error covariance matrix; Step 5: To suppress the impact of transformation error on virtual antenna beamforming, and to minimize the output power of the transformation error while ensuring that the weights are orthogonal to the transformation error of the desired signal direction, the constraint optimization is expressed as:

[0019] Simultaneously considering the virtual interpolation transform beamforming algorithm and error constraint issues, the algorithm constraints are further expressed as follows:

[0020] make , Then the above equation simplifies to:

[0021] In the formula It is any constant between [0,1); Step 6: Construct the matrix Spatial spectral function:

[0022] To broaden the interference null, a window function is used to spatially extend the spatial spectral function:

[0023] in, For the extended spatial spectral function; For window functions, use the Hanning window:

[0024]

[0025] in The window width coefficient is set by... The value is then used to adjust the zero-depression width; To control the main lobe width, a virtual interference spatial spectrum is constructed by adding a virtual interference spectrum to the spatially broadened spatial spectral function. Let the main lobe constraint coefficient be represented, then the virtual interference direction is represented as:

[0026] Assumption To create a virtual interference spatial spectrum function, a virtual interference spectrum is added to the spatially broadened spatial spectrum function. The expression is as follows:

[0027] Step 7: Utilize Reconstructing the interference plus noise covariance matrix ;

[0028] In the formula The zero-deep-trap coefficient The larger the value, the greater the peak gain in the direction of interference, and the deeper the null trap.

[0029] Step 8: Calculate the optimization weights according to the MNV criterion, where Add noise covariance matrix to the reconstructed interference. The steering vector under the desired signal error constraint; .

[0030] Preferably, the virtual array element spacing It is half a wavelength.

[0031] Preferably, the zero-dimple depth coefficient ,and It is a positive number.

[0032] An electronic device includes: a processor and a memory; the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory to cause the electronic device to perform the above-described interference null widening method.

[0033] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described interference zero-trap broadening method.

[0034] A chip includes a processor for retrieving and running a computer program from a memory, causing a device on which the chip is mounted to perform the aforementioned interference zero-trap broadening method.

[0035] A computer program product includes a computer storage medium storing a computer program, the computer program including instructions executable by at least one processor, which, when executed by the at least one processor, implement the above-described interference zero-trap broadening method.

[0036] The beneficial effects of this invention are as follows: Compared with traditional null widening schemes, this invention is applicable to distributed arrays. With the same array aperture, fewer array elements are required, reducing inherent costs. Virtual array element interpolation increases the degrees of freedom of the array system. While ensuring the main lobe gain, it effectively widens the null, solves the problem of interference sources deviating from the preset direction of arrival, and improves the array's anti-interference performance. Attached Figure Description

[0037] Figure 1 This is a schematic diagram of the virtual array element interpolation structure; Figure 2 To broaden the radiation pattern to resist null traps caused by super-degrees of freedom interference; Figure 3 A comparison chart of main lobe performance; Figure 4 A comparison chart of the input-output signal-to-interference-plus-noise ratio before and after correction; Figure 5 This is a flowchart illustrating the implementation of the present invention. Detailed Implementation

[0038] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0039] To effectively eliminate the performance degradation caused by multiple interferences deviating from the null position in distributed arrays, and to adaptively broaden the null while ensuring the main lobe performance, thereby improving the array's anti-interference capability, this invention proposes a null broadening method based on virtual interpolation transformation constraints and spatial spectrum function reconstruction. This method constructs virtual array elements to increase the degrees of freedom of the distributed array system while eliminating grating lobes, and combines spatial spectrum function reconstruction technology to maintain the main lobe gain, achieving adaptive null broadening under hyper-degree-of-freedom interference.

[0040] This invention proposes a method for broadening null traps of super-degrees of freedom based on virtual interpolation transformation constraints and spatial spectrum function reconstruction. By constructing virtual array elements and combining spatial spectrum function reconstruction technology, the method adaptively broadens null traps while eliminating distributed array grating lobes and maintaining the main lobe gain.

[0041] The technical solution of this invention is as follows: Step 1: Perform virtual element interpolation on the distributed array, with the array structure as follows: Figure 1 As shown. Among them. For the actual element spacing, The actual number of array elements. The virtual element spacing is typically half a wavelength.

[0042] Step 2: Solve for the corresponding transformation matrix in each interference region based on the direction of the interference. First, divide the observation area into regions, assuming the signal falls within the region. Within, the area Divide evenly into:

[0043] in , for The left and right boundaries, This is the insertion step size.

[0044] When multiple disturbances fall into different regions in space, it is necessary to divide the space into multiple observation sub-regions. The actual array manifold matrix and the virtual array manifold matrix of each sub-region can be represented as:

[0045]

[0046] and For actual array antennas and virtual array antennas respectively Directional guidance vector, Indicates the number of interferences.

[0047] Suppose there exists a fixed transformation relationship between the real array manifold matrix and the virtual array manifold matrix, such that:

[0048] Transformation matrix Guide the vector matrix of real and virtual array antennas and The following conditions must be met:

[0049] When the number of transformation points is greater than the actual number of array antenna elements and A is full rank, the virtual transformation matrix can be obtained from the above equation as follows:

[0050] Step 3: Based on the transformation matrix of each sub-region, further calculate the virtual array covariance matrix of the corresponding region as follows:

[0051] As can be seen from the above equation, after the virtual transformation, the original white noise of the array antenna is contaminated with colored noise. It is necessary to pre-whiten the colored noise and then synthesize the virtual covariance matrix of each sub-region using the arithmetic mean method. The resulting virtual covariance matrix after removing the colored noise is:

[0052] Step 4: To avoid the large errors caused by the aggregation of covariance data from multiple small regions, and to effectively suppress interference from excessive degrees of freedom, a transformation error constraint method is introduced below. First, the transformation error matrix is ​​defined:

[0053] Within any transforming region, angle The transformation error at point can be expressed as:

[0054] Known in the transformed region Inside, assuming the weights obtained after virtual interpolation are... The weighted transformation error output after virtual interpolation is:

[0055] The output power of the transformation error can then be expressed as:

[0056] In the formula, , is defined as the transformation error covariance matrix.

[0057] Step 5: To suppress the impact of transformation error on virtual antenna beamforming, and to minimize the output power of the transformation error while ensuring that the weights are orthogonal to the transformation error of the desired signal direction, the constraint optimization can be expressed as:

[0058] Considering both the virtual interpolation transform beamforming algorithm and the aforementioned error constraint problem, the algorithm constraint can be further expressed as:

[0059] make , Then the above formula can be simplified to:

[0060] In the formula It is any constant between [0,1).

[0061] Step 6: First construct the matrix Spatial spectral function:

[0062] To broaden the interference null, a window function is used to spatially extend the spatial spectral function:

[0063] in, For the extended spatial spectral function, For window functions. Its characteristics mainly depend on the window function. Here we use the Hanning window.

[0064]

[0065]

[0066] in The window width factor can be set by... The value is then used to adjust the width of the zero trap.

[0067] To control the main lobe width, a virtual interference spatial spectrum can be constructed by adding a virtual interference spectrum to the spatial spectrum function after spatial expansion. Let the main lobe constraint coefficient be represented, then the virtual interference direction can be expressed as:

[0068] Assumption To create a virtual interference spatial spectrum function, a virtual interference spectrum is added to the spatially broadened spatial spectrum function. The expression is as follows:

[0069] Step 7: Utilize Reconstructing the interference plus noise covariance matrix ;

[0070] In the formula The zero-depression coefficient ( ,and (for positive integers) The larger the value, the greater the peak gain in the direction of interference, and the deeper the null trap.

[0071] Step 8: Calculate the optimization weights according to the MNV criterion, where Add noise covariance matrix to the reconstructed interference. The steering vector under the desired signal error constraint;

[0072] Example: Basic experimental setup: The array structure used is as follows: Figure 1 As shown. The actual element spacing is shown in the figure. Actual number of array elements Virtual array element spacing To ensure the same array aperture, the number of virtual array elements is... This study investigates two array structures: a virtual element interpolation array (containing 4 real elements and 9 virtual elements) and a practical extended array (13 real elements). It is assumed that the array model receives five signals during testing, including one target signal and four interference signals, with the target signal originating from... The interference signals are from -60°, -20°, 25°, and 50° respectively. The input signal-to-noise ratio (SNR) is 0dB, and the interference-to-noise ratio (INR) is set to 40dB. The main lobe constraint coefficient is set. Width of each transformation region Zero-deep coefficient Window width coefficient .

[0073] Experiment 1: A comparison of the radiation patterns of two arrays undergoing null-spot broadening. Figure 2 Null broadening patterns for two arrays using traditional methods and the algorithm presented in this paper are given. As can be seen from the figure, the virtual array using traditional methods cannot generate nulls and broaden under hyper-degree-of-freedom disturbances. The method in this paper can effectively solve this problem and constrain the main lobe width and main lobe offset angle to enhance the main lobe performance, but it will lead to an increase in the sidelobe level. Among them, the virtual array has better performance in terms of sidelobe gain than the actual extended array, but its null depth performance is relatively poor.

[0074] Experiment 2: The main lobe performance of the two arrays after null-spot widening was studied. This simulation used 20 Monte Carlo experiments. Figure 3 The paper presents a comparison of the main lobe performance of two arrays after null widening. As can be seen from the figure, compared with the traditional algorithm, the proposed algorithm has a narrower main lobe width and a smaller main lobe offset angle, indicating that its main lobe performance is significantly improved. However, the main lobe constraint performance of the actual extended array is slightly better than that of the virtual array element interpolation array.

[0075] Experiment 3: This study investigates the relationship between the input SNR and output SINR of two arrays after null stretching. The simulation used 20 Monte Carlo experiments, and the average value of the results was taken. Figure 4 The input-output signal-to-interference-plus-noise ratio (SINR) of the two arrays after null-spot widening is presented. It can be seen from the figure that compared with the traditional algorithm, the proposed algorithm improves the output SINR performance under both array structures. However, the overall output SINR performance of the actual extended array is better than that of the virtual array element interpolation array.

Claims

1. A method for broadening the zero-traps of super-degree-of-freedom interference during distributed array reconfiguration, characterized in that, Includes the following steps: Step 1: Perform virtual element interpolation on the distributed array, assuming... For the actual element spacing, The actual number of array elements. This refers to the virtual element spacing; Step 2: Based on the direction of the interference, solve for the corresponding transformation matrix in each interference region; first, divide a certain observation area, assuming the signal falls within the region. Within, the area Divide evenly into: in , for The left and right boundaries, Insertion step size; When multiple disturbances fall into different regions in space, it is necessary to divide the space into multiple observation sub-regions. The actual array manifold matrix and virtual array manifold matrix of each sub-region are represented as follows: and For actual array antennas and virtual array antennas respectively Directional guidance vector, Indicates the number of interferences; Suppose there exists a fixed transformation relationship between the real array manifold matrix and the virtual array manifold matrix, such that: Transformation matrix Guide the vector matrix of real and virtual array antennas and The following conditions must be met: When the number of transformation points is greater than the actual number of array antenna elements and A is full rank, the virtual transformation matrix can be obtained from the above equation as follows: Step 3: Based on the transformation matrix of each sub-region, further calculate the virtual array covariance matrix of the corresponding region as follows: in, This represents the covariance matrix of the actual array received data. Represents the covariance matrix of the desired signal. Indicates noise power; After the virtual transformation, the original white noise of the array antenna is contaminated with colored noise. It is necessary to pre-whiten the colored noise and then synthesize the virtual covariance matrix of each sub-region using an arithmetic mean method. The resulting virtual covariance matrix after removing the colored noise is: Step 4: Introduce the transformation error constraint method. First, define the transformation error matrix: Within any transforming region, angle The transformation error at point is expressed as: Known in the transformed region Inside, assuming the weights obtained after virtual interpolation are... The weighted transformation error output after virtual interpolation is: The output power of the transformation error is then expressed as: In the formula, , is defined as the transformation error covariance matrix; Step 5: To suppress the impact of transformation error on virtual antenna beamforming, and to minimize the output power of the transformation error while ensuring that the weights are orthogonal to the transformation error of the desired signal direction, the constraint optimization is expressed as: Simultaneously considering the virtual interpolation transform beamforming algorithm and error constraint issues, the algorithm constraints are further expressed as follows: make , Then the above equation simplifies to: In the formula It is any constant between [0,1); Step 6: Construct the matrix Spatial spectral function: To broaden the interference null, a window function is used to spatially extend the spatial spectral function: in, For the extended spatial spectral function; For window functions, use the Hanning window: in The window width coefficient is set by... The value is then used to adjust the zero-depression width; To control the main lobe width, a virtual interference spatial spectrum is constructed by adding a virtual interference spectrum to the spatially broadened spatial spectral function. Let the main lobe constraint coefficient be represented, then the virtual interference direction is represented as: Assumption To create a virtual interference spatial spectrum function, a virtual interference spectrum is added to the spatially broadened spatial spectrum function. The expression is as follows: Step 7: Utilize Reconstructing the interference plus noise covariance matrix ; In the formula The zero-deep-trap coefficient The larger the value, the greater the peak gain in the direction of interference, and the deeper the null trap; Step 8: Calculate the optimization weights according to the MNV criterion, where Add noise covariance matrix to the reconstructed interference. The steering vector under the desired signal error constraint; 。 2. The method for broadening the zero-traps of super-degree-of-freedom interference during distributed array reconfiguration according to claim 1, characterized in that, The virtual array element spacing It is half a wavelength.

3. The method for broadening the zero-traps of super-degree-of-freedom interference during distributed array reconfiguration according to claim 1, characterized in that, The zero-deep coefficient ,and It is a positive number.

4. An electronic device, characterized in that, include: Processor and memory; The memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory to cause the electronic device to perform the method as described in any one of claims 1 to 3.

5. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 3.

6. A chip, characterized in that, include: A processor for retrieving and running a computer program from memory, causing a device on which the chip is mounted to perform the method as described in any one of claims 1 to 3.

7. A computer program product, characterized in that, The computer program product includes a computer storage medium storing a computer program, the computer program including instructions executable by at least one processor, which, when executed by the at least one processor, implement the method as described in any one of claims 1 to 3.

Images