阅读说明：本技术 一种三维超稀疏阵列静态方向图综合方法 (Three-dimensional ultra-sparse array static directional diagram synthesis method ) 是由 段克清 杨兴家 李想 王煜岗 于 2021-08-04 设计创作，主要内容包括：本发明公开了一种三维超稀疏阵列静态方向图综合方法,包括步骤如下：首先获取无人机阵元实时坐标,得到二维方向图；分别设置方位向方向图、俯仰向方向图的零零主瓣宽度,将二维方向图划分为若干个待优化区域与1个主瓣区域；为若干个待优化区域设置参考电平与初始干扰值,计算各个待优化区域优化迭代前的数据协方差矩阵以及各阵元初始权值；根据LCMV方法更新不同区域的干扰值,根据干扰值迭代计算数据协方差矩阵,迭代各阵元的权矢量；将权矢量作用到三维超稀疏阵列,得到优化后二维方向图,若各个待优化区域方向图达到所设置的参考电平或收敛条件,则停止迭代；否则,继续迭代；停止迭代后,得到最终的权矢量作用到目标阵列,即可得到满足性能的二维发射方向图。(The invention discloses a three-dimensional ultra-sparse array static directional diagram comprehensive method, which comprises the following steps of: firstly, acquiring real-time coordinates of an array element of an unmanned aerial vehicle to obtain a two-dimensional directional diagram; respectively setting zero main lobe widths of an azimuth directional diagram and a pitching directional diagram, and dividing a two-dimensional directional diagram into a plurality of areas to be optimized and 1 main lobe area; setting reference levels and initial interference values for a plurality of areas to be optimized, and calculating a data covariance matrix and initial weights of array elements before optimization iteration of each area to be optimized; updating interference values of different areas according to an LCMV method, iteratively calculating a data covariance matrix according to the interference values, and iterating weight vectors of each array element; applying the weight vector to the three-dimensional ultra-sparse array to obtain an optimized two-dimensional directional diagram, and stopping iteration if each directional diagram of the area to be optimized reaches the set reference level or convergence condition; otherwise, continuing iteration; and after the iteration is stopped, obtaining a final weight vector to be applied to the target array, and obtaining a two-dimensional emission directional diagram meeting the performance.)

1. A three-dimensional ultra-sparse array static directional diagram synthesis method is characterized by comprising the following steps: the method comprises the following steps:

s1: firstly, acquiring real-time coordinates of an array element of the unmanned aerial vehicle, thereby obtaining a two-dimensional directional diagram;

s2: respectively setting zero main lobe widths of an azimuth directional diagram and a pitching directional diagram, and dividing a two-dimensional directional diagram into a plurality of areas to be optimized and 1 main lobe area;

s3: respectively setting reference levels and initial interference values for a plurality of areas to be optimized, and calculating a data covariance matrix before optimization iteration of each area to be optimized and an initial weight of each array element;

s4: updating interference values of different areas according to an iterative formula of an LCMV method, iteratively calculating a data covariance matrix according to the interference values, and iterating weight vectors of each array element;

s5: weighting the weight vector obtained in the step S4 to each array element of the three-dimensional ultra-sparse array to obtain an optimized two-dimensional directional diagram, and stopping iteration if each directional diagram of the area to be optimized reaches the set reference level or convergence condition; otherwise, returning to step S4 to continue the iteration;

s6: and after the iteration is stopped, the final weight vector is obtained and weighted to the target array, and the two-dimensional emission directional diagram meeting the performance can be obtained.

2. The three-dimensional ultra-sparse array static pattern synthesis method according to claim 1, wherein: each unmanned aerial vehicle carries 8 multiplied by 6 array elements and phased array antennas with array element spacing half-wavelength, the phased array antennas of each unmanned aerial vehicle are combined into a channel through microwaves at a receiving end, therefore, the whole array is an ultra-sparse large-aperture array with 32 channels, the wavelength lambda is 0.6m, and a main beam points to

3. The three-dimensional ultra-sparse array static pattern synthesis method according to claim 1, wherein: the zero main lobe width of the azimuth directional diagram is set to be 20 degrees, the zero main lobe width of the elevation directional diagram is set to be 8 degrees, and the area to be optimized is divided into 3 areas which are respectively as follows: region 1 is azimuth sidelobe-pitch main lobe), region 2 is azimuth main lobe-pitch sidelobe, and region 3 is azimuth sidelobe-pitch sidelobe; the main lobe area is an azimuth main lobe-a pitching main lobe.

4. The three-dimensional ultra-sparse array of claim 3The state directional diagram synthesis method is characterized in that: the reference levels of the region 1, the region 2 and the region 3 are set to be P_r1＝-40dB，P_r2＝-30dB，P_r3＝-50dB。

5. The three-dimensional ultra-sparse array static pattern synthesis method according to claim 4, wherein: interference values xi of different regions_iThe iterative formula is as follows:

wherein ξ_iAn interference value representing a region i, i being 1,2, 3; k represents the number of iterations;the numerical value of the directional diagram of each area; p_riDenotes the reference level of the region i and δ is the iteration step.

6. The three-dimensional ultra-sparse array static pattern synthesis method according to claim 5, wherein: updating an iterative interference value: when the level of the angle directional diagram is lower than the reference level, the angle interference value is reduced; conversely, the angular disturbance value increases.

7. The three-dimensional ultra-sparse array static pattern synthesis method according to claim 5, wherein: constructing a signal manifold matrix of each region, and iteratively calculating a data covariance matrix by combining interference values of different regions;

wherein, the signal manifold matrix of each region is constructed as follows:

wherein i is 1,2,3, P_iIndicating the amount of interference, Q, in the direction of the region i_iRepresenting region i in planThe number of upward disturbances;representing a signal steering vector.

8. The three-dimensional ultra-sparse array static pattern synthesis method according to claim 7, wherein: k-th iteration data covariance matrix R_kThe calculation formula of (a) is as follows:

R_k＝[A₁ A₂ A₃]·diag[ξ_1k ξ_2k ξ_3k]·[A₁ A₂ A₃]^H+σI

where σ denotes a constant, and I denotes a (32 × 8 × 6) -dimensional identity matrix.

9. The method for synthesizing a three-dimensional ultra-sparse array static pattern according to claim 8, wherein: the weight vector w of the kth iteration song array element is represented as:

wherein when k is 0, make xi_i0When R is equal to 0, then R₀＝σI，w₀Expressed as:

10. the method for synthesizing a three-dimensional ultra-sparse array static pattern according to claim 9, wherein: setting a convergence condition as Δ < σ, where Δ is expressed as follows:

Δ＝|mean[F_sl(k)]-mean[F_sl(k-1)]|

wherein, F_slMean represents the mean value for the side lobe voltage values of the pattern.

Technical Field

The invention relates to the technical field of radar signal processing, in particular to a three-dimensional ultra-sparse array static directional diagram synthesis method.

Background

In the face of increasingly complex combat environments and combat tasks, accurate acquisition of war information and realization of zero casualties are urgent requirements for future weaponry. However, the airborne early warning radar of the single platform cannot go deep into an enemy to achieve stable coverage detection, and the radar has a large cross section and poor maneuverability, so that the airborne early warning radar is easy to destroy in war to cause casualties. At present, although the developed large unmanned early warning machine can reduce casualties, the cost is high and the maneuverability is extremely poor. In order to improve the defects of the early warning platform, researchers transfer the research gravity center to the unmanned aerial vehicle cluster cooperative detection.

With the development of unmanned aerial vehicle technology, unmanned aerial vehicles have gradually turned from a single platform to a group cooperation manner. Compared with a traditional large single-base unmanned aerial vehicle platform, unmanned aerial vehicle cluster cooperative detection has many obvious advantages. Firstly, the larger antenna aperture and the larger spatial degree of freedom can realize better spatial resolution and clutter suppression performance. Secondly, the unmanned aerial vehicle cluster can greatly improve the endurance of the early warning detection system, and even if one or more unmanned aerial vehicles are destroyed by enemies, the better overall detection performance of the cluster can be still maintained. Finally, the small unmanned aerial vehicle has a smaller radar cross-sectional area and higher maneuverability, can pass through a hidden environment, and effectively detects the front edge of a high threat area in a deep mode, so that the detection range is expanded. In addition, the cost of the small unmanned aerial vehicle is far lower than that of a large single-base unmanned aerial vehicle platform.

In actual combat, in order to meet combat requirements and unmanned aerial vehicle safety spacing, unmanned aerial vehicle clusters are often distributed randomly or according to a certain formation of a three-dimensional ultra-sparse array (array element spacing is greater than half wavelength). The main beam distributed in the three-dimensional ultra-sparse array is narrow and has high spatial resolution, but the sidelobe level is high, the grating lobe effect is strong, strong interference signals are easily brought, and false alarms and target angle ambiguity are caused. This will result in a significant degradation of the drone cluster cooperative probing performance. To alleviate this phenomenon, we need to integrate the patterns of the clustered system to obtain an array pattern of low sidelobe weak grating lobe effect.

The currently proposed pattern synthesis method based on the Linear Constrained Minimum Variance (LCMV) criterion calculates each array element weight by injecting virtual interference from different angles, which is only directed to one-dimensional or two-dimensional distributed arrays, and can only optimize the azimuth pattern and the pitch pattern sidelobes to the same level. However, in actual combat, the required performance of the azimuth and elevation patterns tends to be different. Therefore, if the azimuth directional diagram and the elevation directional diagram can be respectively optimized to different performances, the actual combat requirement can be more close, and the combat success rate is greatly increased.

Disclosure of Invention

In order to solve the problems of the prior art, the invention provides a three-dimensional ultra-sparse array static directional diagram synthesis method, which is more close to the actual combat requirement, and has fewer iteration times and higher iteration speed than the conventional LCMV directional diagram synthesis method.

In order to achieve the purpose of the invention, the technical scheme is as follows:

a three-dimensional ultra-sparse array static directional diagram synthesis method comprises the following steps:

s1: firstly, acquiring real-time coordinates of an array element of the unmanned aerial vehicle, thereby obtaining a two-dimensional directional diagram;

s2: respectively setting zero main lobe widths of an azimuth directional diagram and a pitching directional diagram, and dividing a two-dimensional directional diagram into a plurality of areas to be optimized and 1 main lobe area;

s3: respectively setting reference levels and initial interference values for a plurality of areas to be optimized, and calculating a data covariance matrix before optimization iteration of each area to be optimized and an initial weight of each array element;

s4: updating interference values of different areas according to an iterative formula of an LCMV method, iteratively calculating a data covariance matrix according to the interference values, and iterating weight vectors of each array element;

s5: weighting the weight vector obtained in the step S4 to each array element of the three-dimensional ultra-sparse array to obtain an optimized two-dimensional directional diagram, and stopping iteration if each directional diagram of the area to be optimized reaches the set reference level or convergence condition; otherwise, returning to step S4 to continue the iteration;

s6: and after the iteration is stopped, the final weight vector is obtained and weighted to the target array, and the two-dimensional emission directional diagram meeting the performance can be obtained.

Preferably, each unmanned aerial vehicle carries a phased array antenna with 8 × 6 array elements and an array element spacing of half wavelength, and the phased array antennas of each unmanned aerial vehicle are combined into one channel by microwave at a receiving end, so that the whole array is a 32-channel ultra-sparse large-aperture array, the wavelength λ is 0.6m, and the main beam points to

Further, the zero main lobe width of the azimuth directional diagram is set to be 20 degrees, the zero main lobe width of the pitch directional diagram is set to be 8 degrees, and the area to be optimized is divided into 3 areas, which are respectively as follows: region 1 is azimuth sidelobe-pitch main lobe), region 2 is azimuth main lobe-pitch sidelobe, and region 3 is azimuth sidelobe-pitch sidelobe; the main lobe area is an azimuth main lobe-a pitching main lobe.

Still further, the reference levels of the region 1, the region 2 and the region 3 are set to be P respectively_r1＝-40dB，P_r2＝-30dB，P_r3＝-50dB。

Still further, the interference values ξ of the different regions_iThe iterative formula is as follows:

wherein ξ_iAn interference value representing a region i, i being 1,2, 3; k represents the number of iterations;the numerical value of the directional diagram of each area; p_riDenotes the reference level of the region i and δ is the iteration step.

Still further, the iterative interference value is updated: when the level of the angle directional diagram is lower than the reference level, the angle interference value is reduced; conversely, the angular disturbance value increases.

Further, constructing a signal manifold matrix of each region, and combining interference values of different regions to iteratively calculate a data covariance matrix;

wherein, the signal manifold matrix of each region is constructed as follows:

wherein i is 1,2,3, P_iIndicating the amount of interference, Q, in the direction of the region i_iRepresenting the amount of interference in the pitch direction of the area i;representing a signal steering vector.

Still further, the kth iteration data covariance matrix R_kThe calculation formula of (a) is as follows:

R_k＝[A₁ A₂ A₃]·diag[ξ_1kξ_2kξ_3k]·[A₁ A₂ A₃]^H+σI

where σ denotes a constant, and I denotes a (32 × 8 × 6) -dimensional identity matrix.

Still further, the weight vector w of the kth iteration song array element is represented as:

wherein when k is 0, make xi_i0When R is equal to 0, then R₀＝σI，w₀Expressed as:

still further, the convergence condition is set to Δ < σ, where Δ is expressed as follows:

Δ＝|mean[F_sl(k)]-mean[F_sl(k-1)]|

wherein, F_slMean represents the mean value for the side lobe voltage values of the pattern.

The invention has the following beneficial effects:

according to the method, a two-dimensional directional diagram of the three-dimensional ultra-sparse array is divided into a plurality of areas to be optimized and 1 main lobe area for partition processing, then different reference levels are respectively set for the areas to be optimized, and interference is injected into different areas, so that the directional diagram levels of different areas continuously approach the set levels until convergence. The method can respectively optimize the azimuth direction and the pitching direction diagram to different reference levels, thereby being closer to the actual combat requirement. Meanwhile, through the partition operation, the method has higher iteration speed and better convergence effect than the traditional LCMV method. Therefore, the method is suitable for unmanned aerial vehicle cluster cooperative detection.

Drawings

FIG. 1 is a flow chart of the steps of the method described in this example 1.

Detailed Description

The invention is described in detail below with reference to the drawings and the detailed description.

Example 1

Suppose 32 drones are at random 240 × 240 × 240m³Three-dimensional space, flight safety interval is more than 10m, unmanned aerial vehicle cluster height is 5000m, each unmanned aerial vehicle carries 8 x 6 array elements and array element interval half-wavelength's phased array antenna, each unmanned aerial vehicle phased array antenna microwave synthesis a passageway at the receiving end, consequently whole array is 32 super sparse large aperture array of passageway, wavelength lambda is 0.6m, the directional main beam is to pointing toThe following detailed steps of the overall invention are described in conjunction with fig. 1 and the examples:

as shown in fig. 1, a method for synthesizing a three-dimensional ultra-sparse array static pattern includes the following steps:

s1: firstly, acquiring real-time coordinates of an unmanned aerial vehicle array element by adopting a three-dimensional ultra-sparse array airborne radar to obtain a two-dimensional directional diagram, wherein the three-dimensional ultra-sparse array airborne radar adopts a phased array antenna;

s2: respectively setting zero main lobe widths of an azimuth directional diagram and a pitching directional diagram, and dividing a two-dimensional directional diagram into a plurality of areas to be optimized and 1 main lobe area;

s3: respectively setting reference levels and initial interference values for a plurality of areas to be optimized, and calculating a data covariance matrix before optimization iteration of each area to be optimized and an initial weight of each array element;

s4: updating interference values of different areas according to an iterative formula of an LCMV method, iteratively calculating a data covariance matrix according to the interference values, and iterating weight vectors of each array element;

s5: weighting the weight vector obtained in the step S4 to each array element of the three-dimensional ultra-sparse array to obtain an optimized two-dimensional directional diagram, and stopping iteration if each directional diagram of the area to be optimized reaches the set reference level or convergence condition; otherwise, returning to step S4 to continue the iteration;

s6: after stopping iteration, obtaining the final weight vector to weight to the target array, namely F ═ w_optA |, wherein w_optAnd (3) obtaining a two-dimensional emission directional diagram F meeting the performance by taking the optimal weight vector as A and taking the optimal weight vector as A, wherein A is a signal flow matrix of the array.

In a specific embodiment, the null main lobe width of the azimuth directional diagram is set to be 20 °, the null main lobe width of the pitch directional diagram is set to be 8 °, and the region to be optimized is divided into 3 regions, which are respectively as follows: region 1 is azimuth sidelobe-pitch main lobe), region 2 is azimuth main lobe-pitch sidelobe, and region 3 is azimuth sidelobe-pitch sidelobe; the main lobe area is an azimuth main lobe-a pitching main lobe. The azimuth and elevation main lobe widths are zero main lobe widths and are related to the set side lobe levels.

The present embodiment sets the reference level of region 1 to be P_r1-40 dB; setting region 2 reference level to P_r2-30 dB; reference level of region 3 is P_r3-50 dB. The initial interference value of the area to be optimized is set to 1.

In a specific embodiment, the interference values ξ for the different regions_iThe iterative formula is as follows:

wherein ξ_iAn interference value representing a region i, i being 1,2, 3; k represents the number of iterations;the numerical value of the directional diagram of each area; p_riDenotes the reference level of the region i and δ is the iteration step.

The present embodiment may update the iterative interference value by using an LCMV iterative formula: when the level of the angle directional diagram is lower than the reference level, the angle interference value should be reduced; conversely, the angular disturbance value should be increased.

In a specific embodiment, a signal manifold matrix of each region is constructed, and a data covariance matrix is calculated by combining interference values of different regions in an iterative manner;

wherein, the signal manifold matrix of each region is constructed as follows:

wherein i is 1,2,3, P_iIndicating the amount of interference, Q, in the direction of the region i_iRepresenting the amount of interference in the pitch direction of the area i;representing a signal steering vector.

Wherein, the k-th iteration data covariance matrix R_kThe calculation formula of (a) is as follows:

R_k＝[A₁ A₂ A₃]·diag[ξ_1kξ_2kξ_3k]·[A₁ A₂ A₃]^H+σI

where σ denotes a constant, and I denotes a (32 × 8 × 6) -dimensional identity matrix; diag () represents a function whose role is to extract the diagonal elements of the matrix and create a diagonal matrix.

In a specific embodiment, the weight vector w of the song array element of the kth iteration is represented as:

wherein when k is 0, make xi_i0When R is equal to 0, then R₀σ I denotes an initialization data covariance matrix, w₀The initialization weight vector is represented as:

in a specific embodiment, the convergence condition is set according to the pattern characteristics, such as when the average level of the side lobe pattern varies by less than a small constant, the iteration is stopped. Therefore, the present embodiment stops iteration and sets the convergence condition to Δ < σ, where Δ is expressed as follows:

Δ＝|mean[F_sl(k)]-mean[F_sl(k-1)]|

wherein, F_slMean represents the mean value for the side lobe voltage values of the pattern.

The weight vector obtained by the embodiment is applied to the three-dimensional ultra-sparse array, so that the side lobe levels in different areas can be reduced to different values, and the grating lobe effect is reduced, and the requirement of actual combat is met.

It should be understood that the above-described embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the claims of the present invention.