Self-adaptive sidelobe anti-interference method for 84-channel antenna array

Abstract

The invention discloses an adaptive sidelobe anti-interference method for an 84-channel antenna array. The adaptive sidelobe anti-interference method comprises the following steps: constructing a simulation model of the 84-channel antenna array; baseband data are collected in the simulation model, and two-dimensional FFT data of the baseband data are calculated; calculating an angle fuzzy number Q; obtaining an interference signal detection threshold, and determining the number of anti-interference rejection points; the interference signal detection threshold is the maximum angle fuzzy number; the number of the anti-interference rejection points is greater than the product of the maximum interference number and the angle fuzzy number; constructing an interference subspace; and calculating a static weight coefficient according to the interference subspace to realize interference position null processing. According to the technical scheme, the influence of interference on signal detection can be reduced, and the improvement of the detection capability of the target signal and the suppression of the interference signal are realized; according to the method, the interference signal angle is detected in real time, and the adaptive weight is adjusted, so that the stability, reliability and adaptivity of beam forming are ensured.

Description

Technical Field

[0001] The present invention relates to the field of signal processing, and in particular to an adaptive sidelobe anti-interference method for an 84-channel antenna array. Background Art

[0002] As a new type of phased array antenna, digital array antenna is digitized at the array element level and uses digital beamforming (DBF) technology to generate receiving beams, which are used to improve the spatial resolution and anti-interference ability of signals. When applying array antennas, the processing of array signals is particularly important. The traditional DBF implementation method is non-adaptive DBF technology, also known as static beamforming. The data required for beamforming is independent, that is, it is independent of the input signal data. The weighting value of each array element is pre-set and fixed. Therefore, it is not suitable for rapidly changing environments or dynamically changing signal conditions, and may not even be able to effectively respond to signal changes, resulting in a decrease in the quality of the received signal. By controlling the side lobe level (SLL) of the antenna array beam pattern to reduce the impact of interference and environmental clutter on system performance, different weighting values ​​can achieve shaped beam synthesis, low side lobes, beam broadening, etc. For a given main-sidelobe ratio, the sum beam uses Taylor distribution to obtain weights, which controls the sidelobe by changing the sidelobe zero point position and has a good aperture utilization factor. The difference beam is obtained using Bayliss distribution.

[0003] In actual environments, commonly used weight coefficient schemes are easily affected by interference in complex interference environments. In particular, when performing signal detection and processing in strong interference environments, the signal detection performance is greatly reduced and cannot meet the anti-interference requirements. Therefore, a technical solution that can adaptively adjust the beam weights of an 84-channel antenna array is needed to improve the signal detection performance of the digital beamforming system. Summary of the invention

[0004] To achieve the above object, the present application provides an adaptive sidelobe anti-interference method for an 84-channel antenna array, comprising the following steps:

[0005] Build a simulation model of an 84-channel antenna array;

[0006] The simulation model collects baseband data and calculates two-dimensional FFT data of the baseband data, which is expressed as:

[0007] Among them, M x 、M y is the number of sub-arrays of baseband data in the x and y directions, K x , K yis the number of array elements in the x and y directions in each sub-array, dx and dy are the array element spacings in the x and y directions respectively, and l1 and l2 are the position indexes of the sub-array;

[0008] Calculate the angular blur amount Q;

[0009] Obtaining an interference signal detection threshold, and determining the number of anti-interference rejection points; the interference signal detection threshold is the maximum angular ambiguity number; the number of anti-interference rejection points is greater than the product of the maximum number of interference and the number of angular ambiguity;

[0010] Constructing an interference subspace; calculating a static weight coefficient according to the interference subspace to achieve interference position nulling processing.

[0011] Among them, the calculation method of the angle blur quantity Q is:

[0012]

[0013] Among them, K x , K y is the number of array elements in the x and y directions in each subarray, dx and dy are the array element spacings in the x and y directions respectively, q is the scaling factor in the x direction during FFT transformation, h is the scaling factor in the y direction during FFT transformation, and γ is the rotation coefficient around the origin of the coordinate center during FFT transformation.

[0014] The calculation method of the interference signal detection threshold is:

[0015]

[0016] Where T is the interference signal detection threshold, α is the threshold factor, To estimate the noise power, L is the signal point set, L P is the point set to be eliminated, t is the sampling time, R is the false alarm probability, and X is the two-dimensional discrete FFT data of the single fast-beat sub-array received data.

[0017] Furthermore, constructing the interference subspace includes:

[0018] Obtain angle information of interference source;

[0019] Determine whether it is within the main lobe according to the angle information;

[0020] If it is within the main lobe, the interfering angle position is eliminated; otherwise, the interfering point is added to the interfering subspace.

[0021] Wherein, obtaining the angle information of the interference source includes:

[0022] Calculate the angle information of the interference source;

[0023] The peak decision threshold is calculated using the two-dimensional FFT data, and the two-dimensional FFT results are screened to retain the coordinate positions and numbers of points whose amplitude values ​​are greater than the decision threshold;

[0024] Determine whether the corresponding angle is in the visible area: For data in the visible area, retain the angle information, and if it is not in the visible area, discard the point information;

[0025] The angle information in the visible area is further screened using the local maximum characteristics of the peak value, that is, the angle information in the visible area is compared with the corresponding amplitude values ​​of the 8 points around it. If the amplitude of the point is greater than the amplitudes of the 8 points around it, it indicates that the point is the local maximum value and is retained, otherwise it is discarded.

[0026] Among them, the method of judging whether it is within the main lobe according to the angle information is: judging whether the calculated interference angle is within the range of ±5° of the main beam pointing, if the interference angle is within this range, it is interference within the main lobe.

[0027] Further, calculating the static weight coefficient according to the interference subspace includes:

[0028] Construct the interference-oriented vector subspace matrix C corresponding to R interferences, expressed as:

[0029]

[0030] Among them, a m (u,v) represents the subarray-level steering vector;

[0031] The interference-oriented vector subspace matrix C is column-orthogonalized to obtain an orthogonalized interference subspace C', which is expressed as:

[0032] C′=[β 1,1 ,β 1,2 ,β 1,3 ,...,β R,1 ,β R,2 ,β R,3 ];

[0033] Calculate the interference null weight vector as follows:

[0034]

[0035] Calculate the adaptive and differential three-beam weight coefficients as follows:

[0036]

[0037] Among them, W sp is the adaptive beam weight coefficient, W up is the adaptive elevation difference beam weight coefficient, Wvp is the adaptive azimuth difference beam weight coefficient, W s are the static and beam weight coefficients, W u is the static elevation difference beam weight coefficient, W v is the static azimuth difference beam weight coefficient.

[0038] According to the present invention, the weight distribution of the beam can be adjusted according to the change of interference and combined with the channel characteristics, so that the main beam is aimed at the target and the null is aimed at the interference, which greatly reduces the impact of interference on signal detection and realizes the improvement of the detection capability of the target signal and the suppression of the interference signal. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 is a step diagram of an adaptive sidelobe anti-interference method provided according to an embodiment of the present invention;

[0040] Figure 2 are a top view and a side view after two-dimensional FFT processing provided by an embodiment of the present invention;

[0041] Figure 3 are top and side views of a three-dimensional directional diagram of a difference beam provided according to an embodiment of the present invention;

[0042] Figure 4 4 is a cross-sectional view at a null position provided according to an embodiment of the present invention. DETAILED DESCRIPTION

[0043] The specific implementation of the present invention is described in detail below with reference to the accompanying drawings.

[0044] Figure 1 The present invention provides an adaptive sidelobe anti-interference method for an 84-channel antenna array, and the specific implementation process is as follows:

[0045] Step S110: constructing a simulation model of an 84-channel antenna array;

[0046] When implementing the solution provided by the present invention, a simulation model of an 84-channel antenna array is first constructed using HFS (High Frequency Structure Simulator) or other electromagnetic simulation software. The simulation model can be used to collect data and obtain signal parameters for subsequent design and analysis.

[0047] Step S120: collecting baseband data in the simulation model, and calculating two-dimensional FFT data of the baseband data:

[0048] In this step, the number of sampling points is 256 points, and the number of FFT points is 64-point two-dimensional FFT. In the array antenna beamforming, there is a Fourier transform relationship between the weight coefficient vector and the 84-channel array factor. By performing a two-dimensional FFT transform on the single snapshot data of the 84-channel received signal, the spatial spectrum in the two-dimensional angle domain can be obtained.

[0049] Specifically, let M x and M y are the number of sub-arrays in the x and y directions respectively, M = M x ×M y , K x and K y The number of array elements in the x and y directions in each sub-array is K = K x ×K y ; dx and dy are the array element spacings in the x and y directions respectively; then the beam output pointing in the (u, v) direction is:

[0050]

[0051] Furthermore, define is the output m of the sub-matrix x ,m y Signal, l1 and l2 are the position indexes of the subarray, X is the two-dimensional discrete FFT data of the single-fast subarray receiving data, then the two-dimensional discrete Fourier transform of the single-fast subarray receiving data is expressed as follows:

[0052]

[0053] From the above formula, it can be determined that there is a direct correspondence between the output of the subarray digital beamforming and the discrete Fourier transform of the single-quick subarray data vector x(t), where (l1, l2) and (u p ,v p ) is expressed as:

[0054]

[0055] Thus, the calculation method of the two-dimensional FFT data of the baseband data can be obtained, which is expressed as:

[0056]

[0057] Among them, M x 、M y is the number of sub-arrays of baseband data in the x and y directions, K x , K y is the number of array elements in the x and y directions in each subarray, dx and dy are the array element spacings in the x and y directions respectively, l1 and l2 are the position indexes of the subarray, then K x d x , Ky d y The subarray spacing is the subarray spacing in the x and y directions respectively.

[0058] Based on the above formula, by calculating the u of the two-dimensional FFT p 、v p The different angle information of the entire array can be obtained by using u and v, where u and v are equivalent to the index position of each angle information. When the FFT output is detected, the corresponding relationship between the FFT output result and the interference angle can be obtained through formula (4): when the FFT related channel has a large signal output, it is set that there is interference in the direction corresponding to the beam pointing, and the angle information of the interference can be calculated by combining the corresponding relationship.

[0059] Step S130: performing grating lobe ambiguity interference angle analysis and calculating the angle ambiguity quantity Q;

[0060] When calculating the interference angle of the 84-channel array receiving data, the interference detection and angle estimation are performed at the channel level. Due to the increase in channel spacing, grating lobes need to be introduced in the interference angle calculation, but the grating lobes have a certain impact on the accuracy of the interference angle calculation: for example, if the interference pitch angle direction is far away from the main lobe of the beam, two peaks will be formed in the visible area, resulting in angle ambiguity.

[0061] In this step, a fast constant false alarm processing method is used to detect, group and judge interference in order to estimate the interference angle in the case of angle ambiguity.

[0062] In the specific implementation, firstly, formula (4) is further rewritten as follows:

[0063]

[0064] Wherein, λ is the signal wavelength, g1 and g2 are FFT conversion coefficients, and both are integers.

[0065] It can be seen from Formula 5 that when the sub-array spacing K x d x >λ2, that is, |λ / (K x d x )|<2, multiple peaks satisfying the above formula may appear in the visible spatial interval [-1,1], and the peaks all correspond to the same target, that is, angular blur occurs, where the blur spacing |λ / (K x d x )| is related to the subarray spacing.

[0066] From formulas (4) and (5), the calculation method of the angle blur quantity Q can be obtained as follows:

[0067]

[0068] Among them, q is the scaling factor in the x direction during FFT transformation, h is the scaling factor in the y direction during FFT transformation, and γ is the rotation coefficient about the origin of the coordinate center during FFT transformation.

[0069] Step S131: obtaining the interference signal detection threshold and determining the number of anti-interference rejection points;

[0070] In subsequent processing, it is necessary to further estimate the noise power and remove the part with greater interference in the two-dimensional FFT data output by the sub-array.

[0071] Taking into account the angle ambiguity caused by the sub-array spacing, in this step, the interference signal detection threshold is determined according to the angle ambiguity number calculated by the ambiguity analysis of the interference angle in step S130, and the rejection points in the sub-array are determined at the same time. Specifically, the interference signal detection threshold is the maximum angle ambiguity number; the number of anti-interference rejection points is greater than the product of the maximum interference number and the angle ambiguity number.

[0072] First, define the point set to be eliminated as L P , the remaining data is averaged to obtain the noise power estimate after beamforming

[0073]

[0074] Next, according to the false alarm probability P fa Determine the constant false alarm detection threshold factor α:

[0075]

[0076] Finally, the threshold factor α and the estimated noise power The interference detection threshold T is calculated, namely:

[0077]

[0078] Where T is the interference signal detection threshold, α is the threshold factor, To estimate the noise power, L is the signal point set, L P is the point set to be eliminated, t is the sampling time, R is the false alarm probability, and X is the two-dimensional discrete FFT data of the single fast-beat sub-array received data.

[0079] Step S140: construct an interference subspace; calculate a static weight coefficient according to the interference subspace to implement interference position nulling processing.

[0080] The implementation process of constructing the interference subspace includes the following:

[0081] 1) Obtain the angle information of the interference source; in this step, perform peak search, process the two-dimensional FFT data, and calculate the angle information of the interference source; further use the two-dimensional FFT data (64×64) to calculate the peak decision threshold, and perform the first screening of the two-dimensional FFT results to retain the coordinate position and number of points whose amplitude values ​​are greater than the decision threshold;

[0082] Then determine whether the corresponding angle is in the visible area: if the data is in the visible area, keep the angle information; if it is not in the visible area, discard the point information;

[0083] Furthermore, the angle information in the visible area is further screened using the local maximum characteristics of the peak value, that is, the angle information in the visible area is compared with the corresponding amplitude values ​​of the 8 points around it. If the amplitude of the point is greater than the amplitudes of the 8 points around it, it indicates that the point is the local maximum value and is retained, otherwise it is discarded.

[0084] 2) Determine whether the interference is within the main lobe based on the angle information:

[0085] Since it is necessary to form a null for the interference angle position, if the interference is located in the main lobe, it will affect the performance of beamforming. Therefore, it is necessary to determine whether the interference is in the main lobe: the judgment criterion is: determine whether the calculated interference angle is within the ±5° range of the main beam pointing. If the interference angle estimate is within this range, it is considered to be interference in the main lobe and is eliminated without subsequent null formation processing.

[0086] If it is within the main lobe, the interfering angle position is eliminated.

[0087] After eliminating the angular positions within the main lobe, the interference subspace is further constructed.

[0088] In order to improve the stability of interference suppression and avoid interference angle deviation caused by angle estimation error, platform jitter and target rapid maneuvering, in this step, the interference angle is appropriately widened when generating the interference null of the directional pattern.

[0089] Specifically, the calculation of the static weight coefficient based on the interference subspace includes:

[0090] 1) Construct the interference-oriented vector subspace matrix C corresponding to R interferences, expressed as:

[0091]

[0092] Among them, a m (u,v) represents the subarray-level steering vector;

[0093] 2) Construct the interference orthogonal subspace projection matrix from C It can be expressed as:

[0094]

[0095] Where I is the identity matrix.

[0096] 3) Further, Schmidt orthogonalization is used to eliminate the matrix inversion operation:

[0097] When the above operation is implemented in a DSP (digital signal processor) such as TMS320C6678, due to the existence of matrix inversion, especially in complex matrices, it is very time-consuming to implement the inversion operation, which has a great impact on the timeliness of the method. Therefore, the Schmidt orthogonalization operation idea is used in this method to eliminate the large amount of calculation caused by the inversion operation in the above formula. The process of Schmidt orthogonalization is as follows: For a set of vectors α1, α2, α3, …, α n , we need to calculate a set of orthogonalized vectors β1,β2,β3,…,β n , the calculation steps are as follows:

[0098]

[0099] Formula (12) is used to orthogonalize the interference-oriented vector subspace matrix C to obtain the orthogonalized interference subspace C', which is expressed as:

[0100] C′=[β 1,1 ,β 1,2 ,β 1,3 ,...,β R,1 ,β R,2 ,β R,3 ] (13)

[0101] Since β in formula (13) 1,1 ,β 1,2 ,β 1,3 .... are standard orthogonal bases that intersect each other, then:

[0102] C′ H C′=I (14)

[0103] According to the orthogonal characteristics of the standard orthogonal basis, the orthogonal subspace projection matrix is ​​disturbed at this time can be rewritten as:

[0104]

[0105] 3) Implement static weight coefficient calculation: In this step, the static and beam weight coefficients use Taylor weight coefficients, and the static elevation difference beam weight coefficient and the static azimuth difference beam weight coefficient use Baylis weight coefficients.

[0106] Specifically, the low sidelobe static weight vector pointing to the direction (u0, v0) is multiplied by The interference null sink weight vector can be obtained, namely:

[0107]

[0108] Calculate the adaptive and differential three-beam weight coefficients using the following method:

[0109]

[0110] Among them, W sp is the adaptive beam weight coefficient, W up is the adaptive elevation difference beam weight coefficient, W vp is the adaptive azimuth difference beam weight coefficient, W s are the static and beam weight coefficients, W u is the static elevation difference beam weight coefficient, W v is the static azimuth difference beam weight coefficient.

[0111] Circular aperture Taylor distribution and Baylis distribution are applicable to circular aperture phased array radars, such as aircraft radars, missile radars, etc. Circular aperture antenna array is a planar array with circular edges. The antenna units in the array can be arranged in a rectangular grid or a triangular grid. The circular aperture is an inseparable distribution.

[0112] The present invention provides a simulation example with an operating frequency of 10 GHz:

[0113] Assume that the beam is pointed to (θ0, φ0) = (0°, 0°), the interference direction is (θ1, φ1) = (-25°, 0°), the interference-to-noise ratio is 5 dB, and there is no desired signal.

[0114] The two-dimensional FFT processing results are as follows Figure 2 As shown in part (a), the area within the red circle is the visible area u 2 +v 2 ≤1. It can be seen that there are many peaks outside the visible area due to the repetition of the signal space period, but these peaks are eliminated in step S140; at this time, there is only one interference peak in the visible area, and the interference direction is estimated to be (-25.88°, 0.81°) according to the simulation tool;

[0115] Furthermore, the orthogonal projection technique of triangular null widening is used to obtain the top view and side view of the three-dimensional direction diagram of the sum and difference beams, as shown in the following figure: Figure 3As shown, (a) and (b) are the three-dimensional top view and side view of the beam velocity when the beam is pointing to (θ0, φ0) = (0°, 0°); (c) and (d) are the three-dimensional top view and side view of the elevation difference velocity when the interference direction is (θ1, φ1) = (-25°, 0°); (e) and (f) are the three-dimensional top view and side view of the azimuth difference beam. It can be seen from the figure that the present invention can effectively detect interference and generate a null at the interference location;

[0116] Figure 4 The cross-sectional diagram at the null position is provided. Part (a) is the cross-sectional diagram of the sum beam at θ1 = -25°.

[0117] Part (b) is the sum beam in (c) is the cross-sectional view of the elevation difference beam at θ1 = -25°, and (d) is the cross-sectional view of the elevation difference beam at θ1 = -25°. A cross-sectional view of the Figure 4 It can be seen that the sum and difference beams of the anti-interference method algorithm provided by the present invention can generate widened nulls near the interference position, the null width is greater than 3°, and the null depth is better than -25dB, which effectively improves the stability of signal detection.

[0118] The adaptive sidelobe anti-interference method provided by the present invention can adjust the weight distribution of the beam according to the change of interference and the channel characteristics, so that the main beam is aligned with the target and the null is aligned with the interference, which greatly reduces the influence of interference on signal detection and improves the detection capability of the target signal and suppresses the interference signal; the method realizes real-time detection of the interference signal angle and adjustment of the adaptive weight to ensure the stability, reliability and adaptability of the beam forming.

[0119] The above disclosures are only several specific embodiments of the present invention; however, the present invention is not limited thereto, and any changes that can be conceived by those skilled in the art should fall within the protection scope of the present invention.

Claims

1. An adaptive sidelobe anti-interference method for an 84-channel antenna array, characterized in that: The following steps are involved: Build a simulation model of an 84-channel antenna array; The simulation model collects baseband data and calculates two-dimensional FFT data of the baseband data, which is expressed as: Among them, M x 、M y is the number of sub-arrays of baseband data in the x and y directions, K x , K t is the number of array elements in the x and y directions in each sub-array, dx and dy are the array element spacings in the x and y directions respectively, and l1 and l2 are the position indexes of the sub-array; Calculate the angular blur amount Q; Obtaining an interference signal detection threshold, and determining the number of anti-interference rejection points; the interference signal detection threshold is the maximum angular ambiguity number; the number of anti-interference rejection points is greater than the product of the maximum number of interference and the number of angular ambiguity; Constructing an interference subspace; calculating a static weight coefficient according to the interference subspace to achieve interference position nulling processing.

2. The adaptive sidelobe anti-interference method according to claim 1, characterized in that: The calculation method of the angle blur quantity Q is: Among them, K x , K y is the number of array elements in the x and y directions in each subarray, dx and dy are the array element spacings in the x and y directions respectively, q is the scaling factor in the x direction during FFT transformation, h is the scaling factor in the y direction during FFT transformation, and γ is the rotation coefficient around the origin of the coordinate center during FFT transformation.

3. The adaptive sidelobe anti-interference method according to claim 1, characterized in that: The calculation method of the interference signal detection threshold is: Where T is the interference signal detection threshold, α is the threshold factor, To estimate the noise power, L is the signal point set, L P is the point set to be eliminated, t is the sampling time, R is the false alarm probability, and X is the two-dimensional discrete FFT data of the single fast-beat sub-array received data.

4. The adaptive sidelobe anti-interference method according to claim 1, characterized in that: The constructing of the interference subspace comprises: Obtain angle information of interference source; Determine whether it is within the main lobe according to the angle information; If it is within the main lobe, the interfering angle position is eliminated; otherwise, the interfering point is added to the interfering subspace.

5. The adaptive sidelobe anti-interference method according to claim 4, characterized in that: The obtaining of the angle information of the interference source comprises: Calculate the angle information of the interference source; The peak decision threshold is calculated using the two-dimensional FFT data, and the two-dimensional FFT results are screened to retain the coordinate positions and numbers of points whose amplitude values ​​are greater than the decision threshold; Determine whether the corresponding angle is in the visible area: For data in the visible area, retain the angle information, and if it is not in the visible area, discard the point information; The angle information in the visible area is further screened using the local maximum characteristics of the peak value, that is, the angle information in the visible area is compared with the corresponding amplitude values ​​of the 8 points around it. If the amplitude of the point is greater than the amplitudes of the 8 points around it, it indicates that the point is the local maximum value and is retained, otherwise it is discarded.

6. The adaptive sidelobe anti-interference method according to claim 4, characterized in that: The method for judging whether it is within the main lobe according to the angle information is: judging whether the calculated interference angle is within the range of ±5° of the main beam pointing, if the interference angle is within this range, it is interference within the main lobe.

7. The adaptive sidelobe anti-interference method according to claim 1, characterized in that: Calculating the static weight coefficient according to the interference subspace includes: Construct the interference-oriented vector subspace matrix C corresponding to R interferences, expressed as: Among them, a m (u,v) represents the subarray-level steering vector; The interference-oriented vector subspace matrix C is column-orthogonalized to obtain an orthogonalized interference subspace C', which is expressed as: C′＝[β 1,1 ,β 1,2 ,β 1,3 ,...,β R,1 ,β R,2 ,β R,3 ]； Calculate the interference null weight vector as follows: Calculate the adaptive and differential three-beam weight coefficients as follows: Among them, W sp is the adaptive beam weight coefficient, W up is the adaptive elevation difference beam weight coefficient, W vp is the adaptive azimuth difference beam weight coefficient, W s are the static and beam weight coefficients, W u is the static elevation difference beam weight coefficient, W v is the static azimuth difference beam weight coefficient.

Images