阅读说明：本技术 基于非均匀智能超表面阵列的无人机集群测向系统及方法 (Unmanned aerial vehicle cluster direction finding system and method based on non-uniform intelligent super-surface array ) 是由 陈鹏 杨子晗 于 2021-09-06 设计创作，主要内容包括：本发明公开了一种基于非均匀智能超表面阵列的无人机集群测向系统及方法,利用非均匀智能超表面阵列对来波进行多次测量,由全向接收天线接收反射信号。针对全向接收天线接收到的信号,构建多次测量后的反射信号模型,通过挖掘无人机集群目标的空域稀疏特性,采用稀疏重构算法,理论推导了未知到达角的求解表达式。方法通过构建一种新的原子范数,建立起稀疏重构问题,使用半正定规划问题求解相应的凸优化问题,实现对无人机集群信号到达角的稀疏估计。(The invention discloses an unmanned aerial vehicle cluster direction finding system and method based on a non-uniform intelligent super-surface array. Aiming at signals received by the omnidirectional receiving antenna, a reflected signal model after multiple measurements is constructed, and a solving expression of an unknown arrival angle is theoretically deduced by excavating the airspace sparse characteristic of an unmanned aerial vehicle cluster target and adopting a sparse reconstruction algorithm. According to the method, a new atomic norm is constructed, a sparse reconstruction problem is established, a semi-definite programming problem is used for solving a corresponding convex optimization problem, and sparse estimation of the arrival angle of the unmanned aerial vehicle cluster signal is achieved.)

1. An unmanned aerial vehicle cluster direction-finding system based on a non-uniform intelligent super-surface array is characterized by comprising a unit spacing non-uniform intelligent super-surface array and a single omnidirectional receiving antenna; the intelligent super surface is internally integrated with a variable capacitance diode, and the amplitude and the phase of an incident signal are changed by changing the bias voltage of the variable capacitance diode; the intelligent super-surface array has non-uniform space, and the position of the array element has deviation on the basis of the position of an ideal array element; the omnidirectional receiving antenna receives signals reflected to any direction by the non-uniform intelligent super-surface array.

2. The unmanned aerial vehicle cluster direction-finding method based on the non-uniform intelligent super-surface array is characterized by comprising the following steps of:

step 1, initializing unknown parameters including the number of sparse targets in a space domain, the array element number of the non-uniform intelligent super-surface array, the array element spacing, the number of times of measuring incident signals by the non-uniform intelligent super-surface array and the signal-to-noise ratio of an antenna;

step 2, constructing a signal receiving model of the omnidirectional receiving antenna;

the omni-directional receiving antenna receives a signal model represented as

r＝Bdiag{aα,p}Aθ,p s+w

Wherein r ═ r₀,r₁,...,r_M-1]^TReflected signals which are received by the omnidirectional receiving antenna and are subjected to M times of measurement by the non-uniform intelligent super surface array, wherein B is ═ B₀,b₁,...,b_N-1]In order to measure the matrix of the measurements,alpha is the angle of the non-uniform intelligent super-surface array reflected signal received by the omnidirectional receiving antenna,the position vector p of the array element in the non-uniform intelligent super-surface array is [ p ]₀,p₁,...,p_N-1]^TIn the unmanned plane cluster target coming direction θ ═ θ₀,θ₁,...,θ_K-1]^T，s＝[s₀,s₁,...,s_K-1]^TIs a signal source, w ═ w₀,w₁,...,w_M-1]^TIs additive white gaussian noise;

step 3, a sparse reconstruction expression based on a brand-new atomic norm is given;

the specific steps are firstly defining a new atomic norm as

Wherein | x | Y calculation_AIs the atomic norm of the vector x, a_nBased on the original subset A ═ e for vector x^jφa (theta, p), phi epsilon [0,2 pi) } coefficient of the n-th term after expansion, e^jφIn order to ensure that the coefficient of any vector after expansion based on an atomic set is a non-negative real number, and then a sparse reconstruction problem is constructed based on an atomic normr＝[r₀,r₁,...,r_M-1]^TThe reflected signal gamma after M times of measurement of the non-uniform intelligent super surface array is received by an omnidirectional receiving antenna>0, is a regularization parameter, which is the sparse and signal reconstruction error weight;

step 4, solving the sparse reconstruction problem is equivalent to solving a semi-definite programming problem;

and 5, constructing a polynomial by using the solved solution of the semi-definite planning problem, solving a solution corresponding to the maximum value of the polynomial, and finishing the estimation of the arrival angle of the relevant signal.

3. The unmanned aerial vehicle cluster direction finding method based on the non-uniform intelligent super-surface array as claimed in claim 2, wherein in step 4, a sparse reconstruction problem based on atomic norm is equivalent to a semi-definite programming problem, and the problem is solved to obtain an optimal value

Tr G＝γ²/u

Wherein G, q and u are optimization variables introduced, G is a Hermitian matrix with N rows and N columns, q is an N-ary column vector, and u is a scalar quantity, defining C ═ Bdiag { a (α, p) }, symbol (·)^HPerforming conjugate transpose operation on the matrix, wherein the transpose matrix T passes through a relational expressionIs obtained in whichAnd expressing the position vector of the uniformly distributed intelligent super-surface array elements.

4. The method as claimed in claim 3, wherein in step 5, a polynomial is constructed using the solved semi-definite planning problemAnd solving a solution corresponding to the maximum value of the arrival angle of the signal to complete the estimation of the arrival angle of the signal.

Technical Field

The invention belongs to the field of array signal processing, and particularly relates to an unmanned aerial vehicle cluster direction-finding system and method based on a non-uniform intelligent super-surface array.

Background

Angle of arrival (DOA) estimation is an important component in array signal processing. The antenna array usually adopts large-scale radio frequency channels, and the cost is high.

In recent years, intelligent super-surfaces have been widely used in the fields of wireless communication, signal processing and radar. The intelligent super surface intelligently reconstructs the propagation environment of wireless electromagnetic waves by integrating a large number of low-cost passive reflecting elements on one plane and controlling the amplitude/phase of a reflecting unit to adjust the reflecting direction of an incident signal. The intelligent super surface can be arranged on the surface of any building in practical application, and the array element spacing is random.

Early DOA estimation methods, which are simple and effective but limited by the physical aperture of the array, i.e. the presence of a "rayleigh limit", often derive an estimate of the spatial spectrum by fourier transformation. In order to break through the constraint of Rayleigh limit, scholars at home and abroad propose a plurality of super-resolution algorithms, including a subspace decomposition algorithm and a spatial filtering algorithm, which can realize high-resolution spatial spectrum estimation. However, the subspace decomposition-based algorithm only utilizes subspace class information of signals and noise, and does not use more signal characteristic information. The unmanned aerial vehicle cluster has sparse characteristics in an airspace, so the arrival angle estimation of signals can be realized by means of a sparse reconstruction algorithm. The sparse reconstruction method is to discretize a space domain into grids, and assume that a target exists on the grids, and the finer the grid division is, the higher the estimation accuracy is. However, in actual application, there is a problem of grid deviation, which deteriorates positioning performance.

Therefore, under the condition of reducing the system cost, the space domain sparse characteristic of a signal is mined, and the arrival angle estimation is carried out by adopting the problem of meshless sparse reconstruction based on the atomic norm.

Disclosure of Invention

The invention aims to provide an unmanned aerial vehicle cluster direction finding system and method based on a non-uniform intelligent super-surface array, and aims to solve the technical problems that the airspace sparse characteristic of signals is mined and the arrival angle is estimated by adopting the problem of meshless sparse reconstruction based on an atomic norm under the condition that the system cost is reduced.

In order to solve the technical problems, the specific technical scheme of the invention is as follows:

an unmanned aerial vehicle cluster direction finding system based on a non-uniform intelligent super-surface array comprises a unit spacing non-uniform intelligent super-surface array and a single omnidirectional receiving antenna; the intelligent super surface is internally integrated with a variable capacitance diode, and the amplitude and the phase of an incident signal are changed by changing the bias voltage of the variable capacitance diode; the intelligent super-surface array has non-uniform space, and the position of the array element has deviation on the basis of the position of an ideal array element; the omnidirectional receiving antenna receives signals reflected to any direction by the non-uniform intelligent super-surface array.

An unmanned aerial vehicle cluster direction finding method based on a non-uniform intelligent super-surface array comprises the following steps:

step 1, initializing unknown parameters including the number of sparse targets in a space domain, the array element number of the non-uniform intelligent super-surface array, the array element spacing, the number of times of measuring incident signals by the non-uniform intelligent super-surface array and the signal-to-noise ratio of an antenna;

step 2, constructing a signal receiving model of the omnidirectional receiving antenna;

the omni-directional receiving antenna receives a signal model represented as

r＝B diag{a α,p}A θ,p s+w

Wherein r ═ r₀,r₁,...,r_M-1]^TReflected signals which are received by the omnidirectional receiving antenna and are subjected to M times of measurement by the non-uniform intelligent super surface array, wherein B is ═ B₀,b₁,...,b_N-1]In order to measure the matrix of the measurements,alpha is the angle of the non-uniform intelligent super-surface array reflected signal received by the omnidirectional receiving antenna,the position vector p of the array element in the non-uniform intelligent super-surface array is [ p ]₀,p₁,...,p_N-1]^TIn the unmanned plane cluster target coming direction θ ═ θ₀,θ₁,...,θ_K-1]^T，s＝[s₀,s₁,…,s_K-1]^TIs a signal source, w ═ w₀,w₁,...,w_M-1]^TIs additive white gaussian noise;

step 3, a sparse reconstruction expression based on a brand-new atomic norm is given;

the specific steps are firstly defining a new atomic norm as

Wherein | x | Y calculation_AIs the atomic norm of the vector x, a_nBased on the original subset A ═ e for vector x^jφa (theta, p), phi epsilon [0,2 pi) } coefficient of the n-th term after expansion, e^jφIn order to ensure that the coefficient of any vector after expansion based on an atomic set is a non-negative real number, and then a sparse reconstruction problem is constructed based on an atomic normr＝[r₀,r₁,...,r_M-1]^TThe reflected signal gamma after M times of measurement of the non-uniform intelligent super surface array is received by an omnidirectional receiving antenna>0, is a regularization parameter, which is the sparse and signal reconstruction error weight;

step 4, solving the sparse reconstruction problem is equivalent to solving a semi-definite programming problem;

and 5, constructing a polynomial by using the solved solution of the semi-definite planning problem, solving a solution corresponding to the maximum value of the polynomial, and finishing the estimation of the arrival angle of the relevant signal.

Further, in step 4, a sparse reconstruction problem based on the atomic norm is equivalent to a semi-definite programming problem, and an optimal value can be obtained by solving the problem

Tr G＝γ²/u

Wherein G, q and u are optimization variables introduced, G is a Hermitian matrix with N rows and N columns, q is an N-ary column vector, and u is a scalar quantity, defining C ═ Bdiag { a (α, p) }, symbol (·)^HPerforming conjugate transpose operation on the matrix, wherein the transpose matrix T passes through a relational expressionIs obtained in whichAnd expressing the position vector of the uniformly distributed intelligent super-surface array elements.

Further, in step 5, a polynomial is constructed using the solved solution of the semi-definite programming problemAnd solving a solution corresponding to the maximum value of the arrival angle of the signal to complete the estimation of the arrival angle of the signal.

The unmanned aerial vehicle cluster direction finding system and method based on the non-uniform intelligent super-surface array have the following advantages:

1. the unmanned aerial vehicle cluster direction finding method based on the combination of the non-uniform intelligent super-surface array and the single omnidirectional receiving antenna greatly reduces the cost compared with the unmanned aerial vehicle cluster direction finding method based on the phased array antenna.

2. According to the method, the estimation problem of the arrival angle is modeled into a sparse reconstruction problem, so that the sparse characteristic of a signal can be sufficiently excavated, and the estimation performance of the arrival angle is improved;

3. the complexity of calculation is reduced, and the calculation complexity of the algorithm is greatly reduced by using a semi-definite programming algorithm under the condition of ensuring the performance of the algorithm;

4. through the arrival angle estimation process based on the atomic norm, the problem that the traditional estimation algorithm is poor in estimation precision based on unmanned aerial vehicle cluster signals is effectively solved, and the arrival angle estimation precision is effectively improved.

Drawings

FIG. 1 is a schematic diagram of the direction finding operation of an UAV cluster by the combination of a non-uniform intelligent super-surface array and an omnidirectional receiving antenna system according to the present invention;

FIG. 2 is a diagram illustrating the performance of angle of arrival estimation under different SNR conditions according to the present invention;

FIG. 3 is a diagram illustrating the performance of angle-of-arrival estimation under different intelligent super-surface position deviations according to the present invention.

Detailed Description

In order to better understand the purpose, structure and function of the present invention, the following describes an unmanned aerial vehicle cluster direction finding system and method based on a non-uniform intelligent super surface array in further detail with reference to the accompanying drawings.

As shown in fig. 1, the unmanned aerial vehicle cluster direction-finding system based on the non-uniform intelligent super-surface array comprises two parts, namely an intelligent super-surface array with non-uniform cell spacing and a single omnidirectional receiving antenna; the intelligent super surface is internally integrated with a low-cost variable capacitance diode, and the amplitude and the phase of an incident signal can be changed by changing the bias voltage of the variable capacitance diode; compared with a uniform linear array, the intelligent super-surface array has non-uniform space, and the positions of the array elements have deviation on the basis of the positions of ideal array elements; the omnidirectional receiving antenna can receive signals reflected to any direction by the non-uniform intelligent super-surface array.

The invention relates to an unmanned aerial vehicle cluster direction finding method based on a non-uniform intelligent super-surface array, which comprises the following steps:

step 1, initializing unknown parameters including the number of sparse targets in an airspace, the array element number of the non-uniform intelligent super-surface array, the array element position, the number of times of measuring incident signals by the non-uniform intelligent super-surface array and the signal-to-noise ratio of an antenna.

And 2, constructing a signal receiving model of the omnidirectional receiving antenna.

During the m measurement, the reflection signal of the nth array element of the non-uniform intelligent super-surface array is represented as follows:wherein A is_n,mExpressed as the reflection amplitude phi of the nth intelligent super-surface array element during the mth measurement_n,mExpressed as the reflection phase phi of the nth intelligent super-surface array element in the mth measurement_nIs 0 degree or 180 degrees, K is the number of signal sources, s_kWhich represents the source of the k-th signal,indicating the location of the nth intelligent super-surface element, lambda being the wavelength,the difference between the actual placement position of the nth array element and the uniform array is referred, and the arrival angle of the kth target is theta_k。

The reflection signal received by a single omnidirectional receiving antenna at an angle α is represented as:wherein w_mExpressed as gaussian additive white noise.

After all m measurements, all signals received by a single omni-directional receive antenna are represented as:

wherein w ═ w₀,w₁,...,w_M-1]^TIs additive white Gaussian noise, s ═ s₀,s₁,...,s_K-1]^TFor signal source, unmanned aerial vehicle cluster target coming direction theta ═ theta₀,θ₁,...,θ_K-1]^TThe measurement matrix B ═ B₀,b₁,...,b_N-1]Measuring the n-th column vector in the matrix BWherein A is_n,mExpressed as the reflection amplitude phi of the nth intelligent super-surface array element during the mth measurement_n,mExpressed as the m-th measurementReflection phase of n intelligent super surface array elements phi_nIs 0 ° or 180 °.

In addition, the steering matrix is defined as:the steering vector is defined as:wherein e represents a natural constant, j represents an imaginary number unit, lambda is a wavelength, and an array element position vector p ═ p in the non-uniform intelligent super-surface array₀,p₁,...,p_N-1]^T。

Step 3, estimating the arrival angle of the unmanned aerial vehicle cluster target according to the received signal r by using a new method based on the atomic norm, and firstly defining a new atomic norm:wherein | x | Y calculation_AIs the atomic norm of the vector x, a_nBased on the original subset A ═ e for vector x^jφa (theta, p), phi epsilon [0,2 pi) } coefficient of the n-th term after expansion, e^jφIn order to ensure that the coefficient of any vector after expansion based on an atomic set is a non-negative real number, and then a sparse reconstruction problem is constructed based on an atomic normr＝[r₀,r₁,...,r_M-1]^TThe reflected signal gamma after M times of measurement of the non-uniform intelligent super surface array is received by an omnidirectional receiving antenna>0 is a regularization parameter, which is the sparse and signal reconstruction error weight.

The sparse reconstruction problem can be re-described as the semi-definite programming (SDP) problem as follows:

Tr G＝γ²/u

wherein G, q and u are optimization variables introduced, G is a Hermitian matrix with N rows and N columns, q is an N-ary column vector, and u is a scalar quantity, defining C ═ Bdiag { a (α, p) }, symbol (·)^HPerforming conjugate transpose operation on the matrix, wherein the transpose matrix T passes through a relational expressionIs obtained in whichAnd expressing the position vector of the uniformly distributed intelligent super-surface array elements.

Solving the SDP problem by using a MATLAB convex optimization tool box (CVX) to obtain an optimal result

Finally solving the polynomialSolving the angle of arrival of the drone cluster.

The present invention will be further described with reference to examples.

Watch 1

For the system used in the present invention, we compare the current main arrival angle estimation methods (deployed methods) including fast fourier transform algorithm (FFT method), conventional atomic norm minimization algorithm (ANM method), and orthogonal matching pursuit algorithm (OMP method), cramer Circle (CRB) with the simulation parameters in table one.

Fig. 2 shows the arrival angle estimation performance under different signal-to-noise ratios, and it can be seen from the graph that, for the correlated signals, when the signal-to-noise ratio (SNR) is greater than 15dB, the arrival angle estimation algorithm provided by the present invention is significantly superior to the existing algorithm.

The arrival angle estimation performance under different intelligent super-surface array element spacing standard deviations is shown in FIG. 3, and it can be seen from the graph that the performance of the algorithm provided by the invention is not affected by the positions of the intelligent super-surface array elements, and the algorithm is suitable for intelligent super-surface arrays with any spacing.

It is to be understood that the present invention has been described with reference to certain embodiments, and that various changes in the features and embodiments, or equivalent substitutions may be made therein by those skilled in the art without departing from the spirit and scope of the invention. In addition, many modifications may be made to adapt a particular situation or material to the teachings of the invention without departing from the essential scope thereof. Therefore, it is intended that the invention not be limited to the particular embodiment disclosed, but that the invention will include all embodiments falling within the scope of the appended claims.