A grating lobe clutter suppression method based on spatiotemporal adaptive processing

By constructing the covariance matrix of grating lobe clutter within a uniformly overlapping subarray and calculating the weight vector using the minimum variance distortion-free criterion, a null region is designed to align with the grating lobe clutter region. This solves the problem of grating lobe clutter suppression in radar systems and achieves efficient and reliable grating lobe clutter suppression and signal protection.

CN117741576BActive Publication Date: 2026-11-10XIDIAN UNIV
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202311640212.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-12-01
Publication Date
2026-11-10
Estimated Expiration
2043-12-01

AI Technical Summary

Technical Problem

Existing technologies struggle to obtain sufficient independent and identically distributed sample data under non-uniform and non-stationary conditions, leading to grating clutter in radar systems during STAP at the subarray level. This results in signal loss and deterioration of clutter suppression performance. Furthermore, existing grating clutter suppression methods are ineffective in suppressing grating clutter and involve large computational loads and complexities.

Method used

By estimating the clutter power in each grating region under the direction of the main lobe of the beam in a uniformly overlapping subarray structure, the discrete spatial steering vector is calculated, the grating clutter covariance matrix is ​​constructed, and the pattern weight vector is solved using the minimum variance distortion-free criterion. The null region is designed to align with the grating clutter region to form the pattern within the subarray to suppress grating clutter.

Benefits of technology

It effectively suppresses grating lobe clutter, improves the signal-to-noise ratio loss performance of subarray-level space-time adaptive output, simplifies the subarray partitioning method, and enhances the reliability and real-time performance of grating lobe clutter suppression.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117741576B_ABST
    Figure CN117741576B_ABST
Patent Text Reader

Abstract

This invention discloses a grating lobe clutter suppression method based on space-time adaptive processing, mainly addressing the problem that existing technologies, while suppressing grating lobe clutter through radar subarray-level space-time adaptive processing, also eliminate target signals coupled to the grating lobes. The method involves: constructing a uniformly overlapping subarray based on preset parameters; solving for the clutter power in each grating lobe region along the beam main lobe direction; calculating the actual discrete spatial steering vector of each grating lobe clutter region within the uniformly overlapping subarray; reconstructing and enhancing the grating lobe clutter covariance matrix based on this vector and clutter power; calculating the subarray weight vector of the enhanced grating lobe clutter covariance matrix; and using beamforming to obtain a null subarray radiation pattern to suppress grating lobe clutter. This invention improves the signal-to-clutter ratio (SNR) loss performance of the subarray-level space-time adaptive output when grating lobe clutter is present, ensures a simple and efficient subarray partitioning method, and enhances the reliability and real-time performance of grating lobe clutter suppression. It can be used for radar signal processing.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention belongs to the field of radar technology, and further relates to a grating lobe clutter suppression method that can be used in radar signal processing. Background Technology

[0002] Radars typically operate in a downward-looking mode, facing challenges such as high clutter intensity, wide distribution, and broadened clutter echoes. Space-time adaptive processing (STAP) technology effectively suppresses clutter through two-dimensional adaptive filtering in both the spatial and temporal domains. However, according to the Reed-Martit-Brennan RMB criterion, when using full-dimensional STAP, at least two degrees of freedom of training samples are needed to ensure an output signal-to-clutter-to-noise ratio loss within 3dB. However, obtaining sufficient independent and identically distributed sample data is difficult under non-uniform and non-stationary conditions. Furthermore, radars are typically equipped with active phased arrays, resulting in a large number of antennas, making element-level STAP inconvenient. Conventional processing involves first synthesizing array elements to form a certain number of subarrays, achieving spatial dimensionality reduction, and then performing subarray-level STAP. The simplest subarray partitioning method is uniform partitioning, but this method has significant drawbacks. Because the subarray spacing does not satisfy the spatial Nyquist sampling theorem, grating lobes are generated, causing spatial resolution ambiguity. When clutter is located in the grating lobe angle region of the subarray-level pattern, the subarray-level STAP will cause the loss of the desired signal, severely degrading the clutter suppression performance of the system. Therefore, in order to eliminate the impact of subarray-level grating lobe clutter suppression on target signal extraction, it is necessary to study effective grating lobe clutter suppression methods and improve the output signal-to-clutter ratio loss performance of subarray-level STAP when suppressing grating lobe clutter.

[0003] Patent application No. 202211449059.X discloses a "grating lobe interference suppression method based on subarray radiation pattern". Its implementation steps are as follows: First, generate a planar array signal including grating lobe angle interference signal and noise; second, calculate the weight of each subarray after uniformly dividing the planar array; third, design the radiation pattern of each subarray; fourth, suppress grating lobe interference. While this method can produce wide nulls in the subarray radiation pattern at the corresponding grating lobe angle, thus suppressing grating lobe interference, its shortcoming is that it only eliminates interference located in the grating lobe angle domain and does not fully consider the suppression of grating lobe clutter.

[0004] Patent application No. 202010885694.7 discloses a "subarray partitioning method based on space-time adaptive clutter suppression of a motion platform." The implementation steps are as follows: First, determine the planar array to be partitioned; second, encode the row and column partitioning cut points respectively, with each cut point corresponding to the boundary of the corresponding row or column, i.e., each cut point corresponds to a subarray partitioning scheme; third, construct an initial population using a differential evolution algorithm; fourth, design a fitness function using the minimum detectable velocity as the performance index for space-time adaptive clutter suppression; fifth, optimize the initial population using a differential evolution algorithm; sixth, optimize the result with the highest fitness function value in the final population as the optimal subarray partitioning scheme. While this method can extend existing subarray partitioning that only partitions columns to partition rows and columns simultaneously, it can reduce the influence of grating lobes and suppress grating lobe clutter to a certain extent. However, it has the following two shortcomings: First, the subarray division will lead to an increase in side lobes and limited grating clutter suppression performance; Second, the subarray design method based on the optimization algorithm is relatively complex, and as the number of array elements increases, the solution space of the optimization algorithm increases, resulting in an increase in the amount of computation and a slower convergence speed. Sometimes the optimal solution cannot be obtained, so the method is difficult to guarantee robustness in the case of large-scale arrays. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of the prior art by proposing a grating lobe clutter suppression method based on space-time adaptive processing, so as to improve the grating lobe clutter suppression performance, reduce the computational load, and enhance the reliability and real-time performance of clutter suppression.

[0006] The technical approach of this invention is as follows: by estimating the clutter power in each grating region under the direction of the main lobe of the beam in a uniformly overlapping subarray structure, and calculating the actual discrete spatial steering vector in each grating clutter region, a grating clutter covariance matrix within the subarray is constructed to reduce the computational load; by solving the radiation pattern weight vector within the receiving subarray based on the minimum variance distortion-free criterion, the radiation pattern null region within the subarray is aligned with the subarray-level grating clutter region to improve the grating clutter suppression performance.

[0007] Based on the above ideas, the implementation steps of the present invention include the following:

[0008] (1) Based on the pre-set total number of array elements N and number of subarrays N s Number of elements N in the subarray e Number of overlapping elements between subarrays N o Construct uniformly overlapping subarrays with fixed spacing between subarrays;

[0009] (2) Solve for the clutter power in each grating region in the direction of spatial ambiguity to the main lobe of the beam. p = 1, ..., U, where U represents the total number of grid lobe clutter regions;

[0010] (3) Calculate the actual discrete spatial steering vector a for each clutter region within the uniformly overlapping subarray. e (f e p,i ),in Let i represent the i-th discrete spatial frequency within the p-th grating lobe region, where i = 1, ..., I, and I is the total number of discrete spatial frequency points within the p-th grating lobe region.

[0011] (4) Construct the covariance matrix of grating lobe clutter within a uniformly overlapping subarray and enhance it:

[0012] (4a) Based on the above calculation results, calculate the covariance matrix R of the grating lobe clutter within the uniformly overlapping subarray:

[0013]

[0014] Wherein, the superscript H indicates the conjugate transpose operation;

[0015] (4b) Calculate the diagonally loaded grating clutter covariance matrix R based on the grating clutter covariance matrix R.

[0016]

[0017] Where η represents the diagonal loading factor, I N Represents an N×N dimensional identity matrix;

[0018] (5) Calculate the weight vector w within the uniformly overlapping subarray according to the minimum variance distortionless response criterion:

[0019]

[0020] in, The target spatial frequency within the subarray is represented as Spatial guidance vector at the location;

[0021] (6) Set the azimuth angle θ and elevation angle for the entire observation area respectively. Beam scanning is performed to obtain the scanning spatial steering vector within the uniformly overlapping subarray. As the spatial scanning angle changes, By searching for the grating lobe angle that matches the adaptive weight vector w, the subarray radiation pattern that can form a null in the grating lobe clutter region is obtained.

[0022]

[0023] (7) Grid lobe clutter suppression:

[0024] (7a) Construct the spatial dimension reduction transformation matrix T based on the weight vector w in the uniformly overlapping submatrix. sThe number of rows in the matrix is ​​equal to the total number of array elements, and the number of columns in the matrix is ​​equal to the number of uniformly overlapping subarrays into which the array is divided.

[0025] (7b) Calculate the clutter block full-dimensional receiving spatial steering vector a based on the total number of array elements preset in step (1). r Based on the preset radar platform operating speed v, operating wavelength λ, number of pulses K within one coherent processing time, and pulse repetition frequency f PRF The clutter block's azimuth and elevation angles are used to calculate the clutter block's time steering vector a. t ;

[0026] (7c) The dimension reduction transformation matrix T s With the full-dimensional receiving spatial steering vector a r Multiply, then multiply with the time-guided vector a t Perform the Kronecker product to obtain the subarray-level clutter data x after grating lobe clutter suppression. l l = 1, ..., L, where L represents the total number of distance gates.

[0027] Compared with the prior art, the present invention has the following advantages:

[0028] First, this invention utilizes the power information of grating lobe clutter and the discrete spatial steering vector within the grating lobe angle region to construct a uniformly overlapping subarray grating lobe clutter covariance matrix. By using the reconstructed subarray grating lobe clutter covariance matrix and the minimum variance distortionless response criterion to calculate the weight vector, a subarray radiation pattern that can form a null in the grating lobe clutter region is designed. The subarray-level grating lobe clutter is placed in the null region of the subarray radiation pattern and thus eliminated. This overcomes the shortcomings of existing technologies that can only suppress grating lobe interference but cannot achieve grating lobe clutter suppression. This invention enables the suppression of grating lobe clutter while protecting the desired signal, improving the subarray-level space-time adaptive output signal-to-noise ratio loss performance when grating lobe clutter is present.

[0029] Secondly, this invention utilizes a simple uniform overlapping subarray design method to overcome the shortcomings of existing technologies that, when the number of array elements is large, require complex optimization algorithms for subarray partitioning, which leads to an increase in the solution space, increased computational load, and the inability to converge to the optimal solution. This invention enables the effective suppression of grating lobe clutter in cases with a large number of array elements while ensuring a simple and efficient subarray partitioning method, thereby improving the reliability and real-time performance of space-time adaptive grating lobe clutter suppression on the motion platform. Attached Figure Description

[0030] Figure 1 This is a flowchart illustrating the implementation of the present invention;

[0031] Figure 2 This is the radiation pattern within the uniformly overlapping subarray of the receiving end designed in this invention;

[0032] Figure 3This is a simulation diagram of the Capon power spectrum of subarray-level space-time two-dimensional clutter after grating lobe clutter suppression using the present invention;

[0033] Figure 4 This is a simulation diagram of the range Doppler output after grating lobe clutter suppression according to the present invention;

[0034] Figure 5 This is a simulation diagram of the signal-to-noise ratio loss of the output of the present invention as the Doppler frequency changes. Detailed Implementation

[0035] The embodiments of the present invention will now be described in further detail with reference to the accompanying drawings.

[0036] Reference Figure 1 The implementation steps for this example are as follows:

[0037] Step 1: Construct uniformly overlapping subarrays with fixed spacing between subarrays.

[0038] Set the total number of array elements N and the number of subarrays N. s Number of elements N in the subarray e Number of overlapping elements between subarrays N o ;

[0039] Based on the set parameters, a uniformly overlapping subarray with a fixed spacing between subarrays is obtained;

[0040] In this example, let N = 52, N s =10, N e =16, N o =12, meaning the radar uniform linear array has a total of 52 elements. At the receiving end, the uniform linear array is divided into 10 uniformly overlapping subarrays, each containing 16 elements, with 12 elements overlapping between subarrays.

[0041] The first subarray consists of array elements numbered 1 to 16;

[0042] The second subarray consists of array elements numbered 5 to 20;

[0043] The third subarray consists of array elements numbered 9 to 24;

[0044] The fourth subarray consists of array elements numbered 13 to 28;

[0045] The fifth subarray consists of array elements numbered 17 to 32;

[0046] The 6th subarray consists of array elements numbered 21 to 36;

[0047] The 7th subarray consists of array elements numbered 25 to 40;

[0048] The 8th subarray consists of array elements numbered 29 to 44;

[0049] The 9th subarray consists of array elements numbered 33 to 48;

[0050] The 10th subarray consists of array elements numbered 37 to 52.

[0051] Step 2: Based on the uniformly overlapping subarray model designed in Step 1, solve for the clutter power in each grating region along the main lobe direction where spatial ambiguity extends. p = 1, ..., U, where U represents the total number of grid lobe clutter regions.

[0052] In this example, a phased array radar array with uniformly overlapping subarrays at the receiver is mounted on an airborne platform. The radar array operates in front-side looking mode. The specific implementation of this step is as follows:

[0053] 2.1) Set the working geometric model conditions for the airborne radar array (front and side views):

[0054] A three-dimensional Cartesian coordinate system satisfying the right-hand rule is established with the radar array's projection on the ground as the origin. The airborne platform carrying the radar array flies along the Y-axis, and the radar array's axis also follows the Y-axis.

[0055] The azimuth angle is defined as the angle between the projection of the line connecting the ground scattering point and the geometric center of the radar array onto the XOY plane and the Y-axis, denoted as θ, where θ∈[0°,180°].

[0056] The elevation angle is defined as the angle between the line connecting the ground scattering point and the geometric center of the radar array and its projection onto the XOY plane, denoted as...

[0057] The radar array transmitter uses a full array to transmit signals, with static weights configured on the array, and the receiver uses a uniformly overlapping subarray mode to receive signals.

[0058] 2.2) Based on the condition for grating lobe generation, i.e., when the difference between the spatial frequency corresponding to any beam direction and the spatial frequency corresponding to the main beam direction is a non-zero integer, the set of azimuth angles corresponding to the centers of all grating lobe clutter regions is mathematically calculated:

[0059]

[0060] Among them, Ξ θ θ represents the set of azimuth angles of the center of the grating lobe region where the cosine value takes values ​​in the range [-1, 1]. g θ represents the azimuth angle of the grating lobe. g ∈[0°,180°], θ0 represents the main lobe azimuth angle, n represents a positive integer, λ represents the wavelength, and d represents the element spacing. Indicates the pitch angle of the main lobe, <>(-1,1) This indicates an operation that takes a value within the interval [-1, 1].

[0061] 2.3) Calculate the set of spatial frequency centers for all grating lobe clutter regions.

[0062]

[0063] Among them, f s Indicates the spatial frequency of the subarray level;

[0064] 2.4) Calculate the set of Doppler frequencies at the center of all grating lobe clutter regions.

[0065]

[0066] Among them, f t The value represents the Doppler frequency, v represents the airborne platform's operating speed, and f represents the speed of the Doppler frequency. PRF Indicates the repetition frequency of the transmitted pulse;

[0067] 2.5) Obtain the spatial ambiguity of the p-th grating lobe clutter region along the main lobe direction of the beam in a two-dimensional space-time plane, and discretize this grating lobe clutter region along the spatial dimension and Doppler dimension respectively:

[0068]

[0069]

[0070] Among them, f s p,i Let U be the spatial frequency point within the p-th grating lobe clutter region along the beam main lobe direction, where U is a uniform distribution function. The main lobe of the subarray points to the spatial frequency, Δf s f represents the spatial frequency broadening corresponding to a 3dB width of the spatial beam. t p,j Let j be the j-th Doppler frequency point within the p-th grating lobe clutter region along the main lobe direction of the beam, where i = 1, ..., I, where I represents the total number of discrete spatial frequency points, and j = 1, 2, ..., J, where J represents the total number of discrete Doppler frequency points. for The p-th element, △f t The Doppler filter has a 3dB width.

[0071] 2.6) Calculate the spatial frequency point f s p,i and Doppler frequency point f t p,j The corresponding subarray-level clutter spacetime steering vector a(f)s p,i ,f t p,j ):

[0072]

[0073] Among them, a s (f s p,i ) is the subarray-level spatial steering vector. Represents the Kronecker product, a t (f t p,j () represents the time-domain steering vector, and their specific expressions are as follows:

[0074]

[0075]

[0076] Where K represents the number of pulses within a coherent processing interval, and the superscript T indicates the transpose operation;

[0077] 2.7) Calculate the space-time steering vector a(f) of all subarray levels using the Capon spectral estimation method. s p,i ,f t p,j The clutter power corresponding to the p-th grid lobe clutter region is taken as the clutter power estimate of the p-th grid lobe clutter region, with the maximum power being the highest power.

[0078]

[0079] in, Indicates clutter power. This represents the subarray-level data covariance matrix obtained when static weights are used in both the transmitting and receiving subarrays.

[0080] Step 3: Calculate the actual discrete spatial steering vector 'a' for each clutter region within the uniformly overlapping subarray. e (f e p ,i ).

[0081] 3.1) Obtain the distribution of the p-th grating lobe clutter region in the space-time two-dimensional plane within the subarray, and discretize this grating lobe clutter region along the spatial dimension:

[0082]

[0083] in, Represents Ξ θ The p-th element,

[0084] 3.2) Based on the spatial frequency point, calculate the spatial steering vector a within the uniformly overlapping subarray corresponding to the i-th spatial frequency point in the p-th grating lobe clutter region. e (f e p,i ):

[0085]

[0086] Among them, f e p,i This represents the i-th discrete spatial frequency within the p-th grating lobe region.

[0087] Step 4: Construct the covariance matrix of grating lobe clutter within the uniformly overlapping subarray and enhance it.

[0088] 4.1) Clutter power estimates from each grid lobe clutter region The discrete spatial steering vector a in each grating lobe clutter region e (f e p,i The covariance matrix R of the grating lobe clutter within the reconstructed uniformly overlapping subarray is:

[0089]

[0090] Wherein, the superscript H indicates the conjugate transpose operation;

[0091] 4.2) To ensure that the covariance matrix R of the grating lobe clutter within the uniformly overlapping subarray is invertible, a diagonal loading technique can be used to enhance R, resulting in an enhanced grating lobe clutter covariance matrix.

[0092]

[0093] Where η represents the diagonal loading factor, I N This represents an N×N dimensional identity matrix.

[0094] Step 5, based on the grating lobe clutter covariance matrix Obtain the radiation pattern within the subarray with zero-depression regions.

[0095] 5.1) Based on the covariance matrix of grating lobe clutter Calculate the weight vector w within a uniformly overlapping submatrix using the minimum variance distortionless response criterion:

[0096]

[0097] in, Represents the target spatial frequency within the subarray Spatial guidance vector at the location;

[0098] 5.2) Set the azimuth θ and elevation angle for the entire observation area respectively. Beam scanning is performed to obtain the scanning spatial steering vector within the uniformly overlapping subarray.

[0099]

[0100] 5.3) As the spatial scanning angle changes, By searching for the grating lobe angle that matches the adaptive weight vector w, the subarray radiation pattern that can form a null in the grating lobe clutter region is obtained.

[0101]

[0102] Step 6, utilize the orientation pattern within the subarray Suppress grating clutter.

[0103] 6.1) Construct the spatial dimension reduction transformation matrix T based on the weight vector w within the uniformly overlapping submatrix. s :

[0104] The number of rows in this matrix equals the total number of array elements, and the number of columns equals the number of uniformly overlapping subarrays into which the array is divided. In an embodiment of this invention, the uniform linear array has 52 elements. During processing at the receiving end, the uniform linear array is divided into uniformly overlapping subarrays with a total of 10 subarrays. Each subarray contains 16 elements, and there is a 12-element overlap between each subarray. The meanings of the non-zero elements in each column of the matrix are as follows:

[0105] The non-zero elements in the first column, numbered from 1 to 16, represent the weight coefficients corresponding to the elements in the first subarray.

[0106] In the second column, the non-zero elements numbered from 5 to 20 represent the weight coefficients corresponding to the array elements in the second subarray.

[0107] In the third column, the non-zero elements numbered from 9 to 24 represent the weight coefficients corresponding to the array elements in the third subarray.

[0108] The non-zero elements in column 4, numbered from 13 to 28, represent the weight coefficients corresponding to the array elements in the 4th subarray.

[0109] The non-zero elements in column 5, numbered from 17 to 32, represent the weight coefficients corresponding to the elements in the 5th subarray.

[0110] The non-zero elements in column 6, numbered from 21 to 36, represent the weight coefficients corresponding to the elements in the 6th subarray.

[0111] The non-zero elements in column 7, numbered from 25 to 40, represent the weight coefficients corresponding to the elements in the 7th subarray.

[0112] The non-zero elements in column 8, numbered from 29 to 44, represent the weight coefficients corresponding to the elements in the 8th subarray.

[0113] In column 9, the non-zero elements numbered 33 to 48 represent the weight coefficients corresponding to the elements in the 9th subarray.

[0114] In column 10, the non-zero elements numbered 37 to 52 represent the weight coefficients corresponding to the elements in the 10th subarray.

[0115] Its formula is expressed as follows:

[0116]

[0117] Among them, t i For T s The i-th column, The dimension is represented as 1×(i-1)(N) e -N o A vector of all zeros. The dimension is 1×N e A vector of all 1s The dimension is 1×(NN) e -(i-1)(N e -N o A vector consisting entirely of zeros;

[0118] 6.2) Based on the total number of array elements N preset in step 1, obtain the clutter block full-dimensional receiving spatial steering vector a. r :

[0119]

[0120] 6.3) Based on the preset radar platform operating speed v, operating wavelength λ, number of pulses K within one coherent processing time, and pulse repetition frequency f PRF From the clutter block azimuth and elevation angles, the clutter block time steering vector a is obtained. t :

[0121]

[0122] Where the subscript lq represents the q-th clutter block on the l-th range ring, l = 1, ..., L, where L represents the total number of range gates, and q = 1, ..., N c N c Indicates the total number of clutter blocks. θ represents the pitch angle of the clutter block. lq Indicates the azimuth angle of the clutter block;

[0123] 6.4) Transform the dimension reduction matrix T s With the full-dimensional receiving spatial steering vector a r Multiply, then multiply with the time-guided vector at Perform the Kronecker product to obtain the subarray-level clutter data x after grating lobe clutter suppression. l :

[0124]

[0125] Where ρ lq This represents the complex coefficient of the q-th clutter block on the l-th range ring containing static weighted transmit pattern modulation;

[0126] The above-mentioned dimensionality reduction transformation matrix T s With the full-dimensional spatial receiving steering vector a r Multiplication for dimensionality reduction essentially involves using the weight vector w to perform beamforming within a uniformly overlapping subarray, resulting in a subarray radiation pattern capable of forming nulls in the grating lobe clutter region. By using the subarray internal radiation pattern The zero-depression alignment of the grating lobe clutter region ultimately achieves grating lobe clutter suppression.

[0127] The effects of this invention will be further illustrated below with simulation experiments:

[0128] 1. Simulation experimental conditions:

[0129] The hardware platform for the simulation experiment of this invention is: an Intel(R) Core(TM) i7-10700 CPU with a main frequency of 2.90GHz and 16GB of memory.

[0130] The simulation experiment software platform of this invention is: Windows 10 operating system and MATLAB R2020b.

[0131] Parameter settings for the simulation experiment of this invention:

[0132] A uniform linear radar array with a total number of elements N=52 and an element spacing d=0.125m is adopted.

[0133] The transmitter uses a full array for transmission and is configured with static weights.

[0134] During receiver processing, the array is divided into uniformly overlapping subarrays, with N subarrays. s =10, N is the number of elements in the subarray. e =16, the number of overlapping elements between subarrays N o =12;

[0135] The airborne platform has an altitude of H = 10 km, a speed of v = 125 m / s, an operating carrier frequency of f0 = 1.2 GHz, a wavelength of λ = 0.25 m, a bandwidth of B = 5 MHz, an accumulated pulse number K = 16 within one CPI, and a pulse repetition frequency f. PRF =2kHz;

[0136] The point target has an azimuth angle θ0 = 82.79° and an elevation angle of 0°. Its slant range with the radar is R = 115 km, signal-to-noise ratio (SNR) = 3 dB, and clutter ratio (CNR) = 60 dB.

[0137] 2. Simulation content and result analysis:

[0138] Simulation 1: Using the above simulation conditions, 721 discrete angle points are selected within the azimuth range [0°, 180°]. Beamforming is performed on each discrete angle point using this invention, resulting in 721 pattern amplitude response values. Each amplitude response value is then divided by its maximum value to obtain the normalized pattern amplitude response. These amplitude response values ​​are then concatenated to obtain the pattern curve within the uniformly overlapping subarray at the receiver, as shown below. Figure 2 As shown.

[0139] Depend on Figure 2 As can be seen from the simulation results, the present invention forms null regions in the three grating lobe angle regions centered at azimuth angles of 51°, 112°, and 151° in the radiation pattern within the subarray. The null depth can reach -60dB, indicating that the present invention can reduce the clutter signal power located in the grating lobe angle region.

[0140] Simulation 2: Using the above simulation conditions, 181 discrete points were selected in both the spatial frequency domain [-0.5, 0.5] and the normalized Doppler frequency domain [-0.5, 0.5]. The present invention was used to perform two-dimensional clutter Capon spectrum estimation for each spatial frequency point and normalized Doppler frequency point to obtain clutter power values. These values ​​were then concatenated to obtain a simulation diagram of the subarray-level space-time two-dimensional clutter Capon power spectrum after grating clutter suppression, as shown below. Figure 3 As shown.

[0141] Depend on Figure 3 As can be seen from the simulation results, the normalized Doppler frequencies and spatial frequencies at the center of the three grating lobe clutter regions in the direction of the main lobe of the beam, which are spatially blurred to the beam main lobe, are represented by two-dimensional coordinates as [-0.44, 0.25], [-0.19, 0.25], and [0.31, 0.25], respectively. The two-dimensional Capon power spectrum of clutter in the space-time of these three grating lobe regions is completely eliminated, indicating that the present invention can effectively suppress clutter signals located at the grating lobes of the subarray level pattern within a uniformly overlapping subarray.

[0142] Simulation 3: Using the above simulation conditions, within 2000 range gates, the data for each range gate is discretized into 64 frequency points within the normalized Doppler frequency range [-0.5, 0.5]. The present invention is used to perform matched filtering on each range gate and the normalized Doppler frequency point to obtain the output amplitude. These amplitudes are then concatenated to obtain the range-Doppler output, as shown below. Figure 4 As shown.

[0143] Depend on Figure 4 As shown in the simulation results, the two-dimensional coordinates [X,Y] within the box represent the target's Doppler frequency of 0.3095 and range gate number of 1334, respectively. Although the target's Doppler channel is located in the grating lobe clutter region with a normalized Doppler center of 0.31, the grating lobe clutter is suppressed in advance within the uniformly overlapping subarray by the present invention. Subsequently, during subarray-level STAP processing, the target located in the grating lobe clutter region can be preserved. This indicates that the present invention can effectively achieve clutter suppression and greatly improve the moving target detection performance.

[0144] Simulation 4: Using the simulation conditions described above, 721 frequency points were selected within the normalized Doppler frequency range [-0.5, 0.5]. The output signal-to-noise ratio (SNR) loss for each frequency point was calculated using both traditional and the method of this invention. The output SNR loss is defined as the ratio of the output SNR to the ideal clutter-free output SNR, used to measure the clutter suppression capability of the clutter suppression algorithm. Connecting these discrete output SNR loss values ​​yields the curve of output SNR loss versus Doppler frequency, as shown below. Figure 5 As shown in the figure, the dashed line represents the output clutter ratio loss as a function of normalized Doppler frequency after processing by the conventional method, while the solid line represents the output clutter ratio loss as a function of normalized Doppler frequency after processing by the method of this invention.

[0145] The traditional method refers to a radar array transmitter that uses full array transmission with static weights, and a receiver that uses the same uniformly overlapping subarray division method as the simulation parameters, but also configures static weights within the subarrays.

[0146] Depend on Figure 5 As can be seen from the simulation results, in the three grating lobe clutter regions centered at normalized Doppler frequencies of -0.44, -0.19, and 0.31, the output signal-to-noise ratio (SNR) loss performance of the method of the present invention is better than that of the traditional method. The greatest performance improvement is at the normalized Doppler frequency of 0.313889, with an output SNR loss performance improvement of nearly 19 dB. Moreover, the output SNR loss notch is relatively narrow, indicating that the present invention can effectively suppress grating lobe clutter while also detecting slow-moving targets well.

[0147] The simulation results above verify the correctness, effectiveness, and reliability of the present invention.

[0148] The above description is merely a specific example of the present invention and does not constitute any limitation on the present invention. The reference numerals for each step are only for clear description of the embodiments of the present invention and for ease of understanding, and their order is not limited. Obviously, those skilled in the art, after understanding the content and principles of the present invention, may make various modifications and changes in form and detail without departing from the principles and structure of the present invention, but these modifications and changes based on the concept of the present invention are still within the scope of protection of the claims of the present invention.

Claims

1. A grating lobe clutter suppression method based on spatiotemporal adaptive processing, characterized in that, Grating lobe clutter cancellation is achieved by aligning the nulls in the subarray pattern with the corresponding grating lobe angle regions of the subarray level pattern, including the following steps: (1) Based on the pre-set total number of array elements N and number of subarrays N s Number of elements N in the subarray e Number of overlapping elements between subarrays N o Construct uniformly overlapping subarrays with fixed spacing between subarrays; (2) Solve for the clutter power in each grating region in the direction of spatial ambiguity to the main lobe of the beam. U represents the total number of grid lobe clutter regions; (3) Calculate the actual discrete spatial steering vector a for each clutter region within the uniformly overlapping subarray. e (f e p,i ), where f e p,i Let i represent the i-th discrete spatial frequency within the p-th grating lobe region, where i = 1, ..., I, and I is the total number of discrete spatial frequency points within the p-th grating lobe region. (4) Construct the covariance matrix of grating lobe clutter within a uniformly overlapping subarray and enhance it: (4a) Based on the above calculation results, calculate the covariance matrix R of the grating lobe clutter within the uniformly overlapping subarray: Wherein, the superscript H indicates the conjugate transpose operation; (4b) Calculate the diagonally loaded grating clutter covariance matrix R based on the grating clutter covariance matrix R. Where η represents the diagonal loading factor, I N Represents an N×N dimensional identity matrix; (5) Calculate the weight vector w within the uniformly overlapping subarray according to the minimum variance distortionless response criterion: in, The target spatial frequency within the subarray is represented as Spatial guidance vector at the location; (6) Set the azimuth angle θ and elevation angle for the entire observation area respectively. Beam scanning is performed to obtain the scanning spatial steering vector within the uniformly overlapping subarray. As the spatial scanning angle changes, By searching for the grating lobe angle that matches the adaptive weight vector w, the subarray radiation pattern that can form a null in the grating lobe clutter region is obtained. (7) Grid lobe clutter suppression: (7a) Construct the spatial dimension reduction transformation matrix T based on the weight vector w in the uniformly overlapping submatrix. s The number of rows in this matrix is ​​equal to the total number of array elements, and the number of columns in this matrix is ​​equal to the number of uniformly overlapping subarrays into which the array is divided. (7b) Based on the total number of array elements N preset in step (1), calculate the clutter block full-dimensional receiving spatial steering vector a. r Based on the preset radar platform operating speed v, operating wavelength λ, number of pulses K within one coherent processing time, and pulse repetition frequency f PRF The clutter block's azimuth and elevation angles are used to calculate the clutter block's time steering vector a. t ; (7c) The dimension reduction transformation matrix T s With the full-dimensional receiving spatial steering vector a r Multiply, then multiply with the time-guided vector a t Perform the Kronecker product to obtain the subarray-level clutter data x after grating lobe clutter suppression. l l = 1, ..., L, where L represents the total number of distance gates.

2. The method according to claim 1, characterized in that, In step (2), the spatial ambiguity is solved to the clutter power in each grating region along the main lobe direction of the beam. The implementation is as follows: (2a) Calculate the set of azimuth angles corresponding to the centers of all grating clutter regions according to the following formula: Among them, Ξ θ Let θ be the set of azimuth angles of the center of the grating lobe region where the cosine value takes values ​​in the range [-1, 1]. g θ is the azimuth angle of the grating lobe. g ∈[0°,180°], θ0 is the azimuth angle of the main lobe, n is a positive integer, and d is the element spacing. Main lobe pitch angle, <> (-1,1) This indicates an operation that takes a value within the interval [-1, 1]. (2b) Calculate the set of spatial frequency centers for all grating lobe clutter regions according to the following formula. Among them, f s Indicates the spatial frequency of the subarray level; (2c) Calculate the set of Doppler frequencies at the center of all grating lobe clutter regions according to the following formula. Among them, f t Indicates the normalized Doppler frequency; (2d) According to the following formula, the spatially blurred distribution of the p-th grating clutter region in the beam main lobe direction in the space-time two-dimensional plane is obtained, and the grating clutter region is discretized along the spatial dimension and the Doppler dimension respectively: Among them, f s p,i Let U be the spatial frequency point within the p-th grating lobe clutter region along the beam main lobe direction, where U is a uniform distribution function. The main lobe of the subarray points to the spatial frequency, Δf s f represents the spatial frequency broadening corresponding to a 3dB width of the spatial beam. t p,j Let j be the j-th Doppler frequency point within the p-th grating lobe clutter region along the main lobe direction of the beam, where j = 1, 2, ..., J, and J is the total number of discrete Doppler frequency points. for The p-th element in △f t The Doppler filter has a 3dB width. (2e) Calculate the spatial frequency point f according to the following formula. s p,i and Doppler frequency point f t p,j The corresponding subarray-level clutter spacetime steering vector a(f) s p,i ,f t p,j ): Among them, a s (f s p,i ) is the subarray-level spatial steering vector. This represents the Kronecker product operation, a t (f t p,j ) represents the time-domain steering vector; (2f) Calculate the spacetime steering vector a(f) of all subarray levels using the Capon spectral estimation method. s p,i ,f t p,j The clutter power corresponding to the p-th grid lobe clutter region is taken as the clutter power estimate of the p-th grid lobe clutter region, with the maximum power being the highest power. in, Indicates clutter power. This represents the subarray-level data covariance matrix obtained when static weights are used in both the transmitting and receiving subarrays.

3. The method according to claim 2, characterized in that, In step (2e), the subarray-level clutter spatial steering vector a s (f s p,i ) and time-domain steering vector a t (f t p,j ), respectively, represent the following: The superscript T indicates the transpose operation.

4. The method according to claim 1, characterized in that, In step (3), the actual discrete spatial steering vector a of each grating lobe clutter region within the uniformly overlapping subarray is calculated. e (f e p,i The implementation is as follows: (3a) Obtain the distribution of the p-th grating lobe clutter region in the space-time two-dimensional plane according to the following formula, and discretize the grating lobe clutter region along the spatial dimension: Among them, f e p,i Let be the i-th spatial frequency point in the p-th grating lobe clutter region within the subarray, where U represents a uniform distribution and d represents the element spacing. The elevation angle of the main lobe within the subarray. For Ξ θ The p-th element, △f s This represents the spatial frequency broadening corresponding to a 3dB width of the spatial beam. (3b) Based on the spatial frequency point, calculate the spatial steering vector a in the uniformly overlapping subarray corresponding to the i-th spatial frequency point within the p-th grating lobe clutter region. e (f e p,i ): The superscript T indicates the transpose operation.

5. The method according to claim 1, characterized in that, Step (6) Scanning spatial guidance vector within uniformly overlapping subarrays It is expressed as follows: Where d represents the element spacing. This represents the Kronecker product, and the superscript T indicates the transpose operation.

6. The method according to claim 1, characterized in that, In step (7a), the spatial dimension reduction transformation matrix T is constructed based on the weight vector w within the uniformly overlapping subarray. s , means as follows: Among them, t i For T s The i-th column is a vector of all zeros (0) and all ones (1). The superscript T indicates the transpose operation.

7. The method according to claim 1, characterized in that, Step (7b) calculates the clutter block full-dimensional receiving spatial steering vector a r clutter block time steering vector a t They are represented as follows: Where the subscript lq represents the q-th clutter block on the l-th range ring, q = 1, ..., N c N c Indicates the total number of clutter blocks. θ represents the pitch angle of the clutter block. lq The azimuth angle of the clutter block is indicated by 'd', the element spacing is indicated by 'd', and the superscript T indicates the transpose operation.

8. The method according to claim 1, characterized in that, In step (7c), the subarray-level clutter data after grating lobe clutter suppression x l , means as follows: Where, N c ρ represents the total number of clutter blocks. lq This represents the complex coefficient of the q-th clutter block on the l-th range ring containing statically weighted transmit pattern modulation.

Citation Information

Patent Citations

  • Subarray partitioning method based on motion platform spatiotemporal adaptive clutter suppression

    CN111896930B

  • Main lobe interference suppression method for projection matrix preprocessing

    CN115327490A

  • Grating lobe interference suppression method based on subarray directional diagram

    CN115712089A