Distributed multi-curved-surface-array whole-airspace cooperative beam forming method and system

Abstract

The invention discloses a distributed multi-curved-surface-array whole-airspace cooperative beam forming method and system, and belongs to the field of aircraft measurement and control, and the method comprises the steps: S1, carrying out the first-stage weighting of an antenna array element in each curved surface array, and synthesizing a primary beam; and S2, the processing center carries out secondary weighting on the signals of the distributed curved surface array so as to synthesize an equivalent large-aperture phased array antenna and form an expected adaptive wave beam. The problem that a traditional measurement and control system antenna arraying technology is only suitable for a single-beam and single-target measurement and control scene is solved, and the problem that the traditional measurement and control system antenna arraying technology does not have an adaptive airspace interference suppression capability is solved.

Description

Technical Field

[0001] The present invention relates to the field of aircraft measurement and control, and more specifically, to a distributed multi-curved array full-airspace collaborative beamforming method and system. Background Art

[0002] The prior art related to this application is introduced as follows:

[0003] 1) Regarding aircraft measurement and control, "measurement and control" is the abbreviation of measurement and control. "Measurement" refers to the measurement and telemetry of the flight trajectory of the aircraft, and "control" refers to the remote control of the aircraft. "Measurement and control" is abbreviated as "TT&C" (Tracking, Telemetering and Command), where the first "T" refers to tracking and measurement, the second "T" refers to telemetry, and "C" means "command and remote control". Aircraft include aviation aircraft (aircraft with an altitude below 20km), near-space aircraft (aircraft with an altitude between 20 and 100km), and space aircraft (aircraft with an altitude above 100km). When the altitude exceeds 2×106km, it is also called deep space aircraft. These types of aircraft are all objects of measurement and control. Specific aircraft include: satellites, spacecraft, space stations, deep space probes, near-space aircraft, drones, airships, balloons, weapon platforms, etc. Measurement and control communication equipment includes radio, optical, infrared and other measurement and control communication equipment. The main equipment in the current measurement and control communication field is the radio continuous wave measurement and control communication system.

[0004] 2) Regarding antenna arrays, with the development of deep space exploration technology, the amount of exploration data required to be transmitted is increasing, and the distance of deep space exploration is getting farther and farther, resulting in great transmission path losses. Since the equivalent omnidirectional transmit power (EIRP) of the detector is very limited, this requires the ground station to use a large-aperture parabolic antenna to maximize the receiving area to improve the signal-to-noise ratio (SNR). Taking the Mars exploration mission as an example, according to the calculation of the parameters of the National Aeronautics and Space Administration (NASA) mission, an antenna with a diameter of 80m is required to transmit 1Mbps of data. However, the larger the antenna aperture, the narrower the beam, and the more difficult it is to capture the detector's downlink signal; at the same time, the large antenna structure and weight are huge, and it is very difficult to drive the servo. The 70m large antenna (weighing 7260 tons and 76 meters high) of the National Aeronautics and Space Administration (NASA) Deep Space Network (DSN) has reached the current engineering limit. Therefore, it is necessary to adopt antenna array technology to achieve a larger equivalent antenna aperture through multi-antenna signal synthesis.

[0005] 3) Regarding the new generation of curved array measurement and control systems, in the new generation of measurement and control systems, ultra-large-scale curved array measurement and control antennas are introduced. Due to its array configuration that covers the entire airspace and the multi-beam forming technology used, a single phased array antenna can track and control multiple targets in the entire airspace at the same time. In the future, for scenarios where multiple curved arrays are deployed in a distributed manner, a high-gain array antenna with a larger equivalent aperture can be formed through collaborative array signal processing, and it will have the following potential advantages: 1) According to the specific needs of the measurement and control tasks, the curved array resources can be uniformly and flexibly scheduled and managed to achieve higher resource utilization; 2) By adding an adaptive beamforming algorithm, it supports adaptive beamforming and has the ability to suppress airspace interference; 3) It supports flexible and mobile deployment of curved arrays and updates weights in real time; 4) It has partial anti-destruction capabilities and good system robustness.

[0006] 4) Regarding adaptive array antennas, adaptive array antennas form controllable antenna beam patterns by weighting antenna elements, and use adaptive filtering algorithms to dynamically optimize and adjust the weighting coefficients on each antenna element, so that the main lobe of the beam synthesized by the antenna array points in the desired direction, and at the same time, a beam null is generated in the direction of the interference source, effectively resisting suppression interference, and adaptively tracking the target user terminal, playing the role of spatial filtering. Anti-interference technology based on adaptive array antennas has been widely used in the fields of navigation system anti-interference receivers, various aircraft, etc.

[0007] 5) Regarding adaptive beamforming algorithms, as one of the key technologies for array signal processing, adaptive beamforming is widely used in radar, communication, medical treatment, sonar and other fields. Common adaptive beamforming algorithms mainly include the least mean square (LMS) algorithm, the sample matrix inversion (SMI) algorithm and the minimum variance distortionless response (MVDR) algorithm. The LMS algorithm cannot balance the contradiction between convergence speed and steady-state error; when the signal-to-noise ratio increases, the robustness of the SMI algorithm will decrease; the MVDR algorithm requires matrix inversion, which is computationally intensive and is not suitable for real-time updating of the optimal weight vector.

[0008] 6) About radial basis function neural network, radial basis function neural network (RBFNN) is a simple three-layer feedforward neural network, which has the characteristics of fast approximation, simple structure, fault tolerance and strong induction ability for any nonlinear function, and is widely used in signal processing and adaptive control. The radial basis function neural network consists of an input layer, a hidden layer and an output layer. Among them, the activation function of the hidden layer neurons is called the radial basis function. According to functional theory analysis, when the Gaussian function is used as the radial basis function, the function approximation ability of the network can be improved.

[0009] In some existing technical solutions, the antenna array of the traditional measurement and control system refers to the formation of an array of multiple parabolic antennas within a certain geographical area, synthesizing the signals received by each antenna from the same source to improve the received signal-to-noise ratio. The basic principle is to obtain and eliminate the time and phase differences between the signals of each antenna, and then coherently accumulate them to synthesize a high-gain narrow beam. Common algorithms for the antenna array of the traditional measurement and control system include Simple, Sumple, Eigen, etc. Under the condition that the noise of each antenna is uncorrelated, in theory, the SNR of the signal received by the N antenna array is N times that of a single antenna. If a small-sized, light-weight, wide-beam small-caliber antenna array is used, it will be more beneficial for signal capture and tracking. In addition, in terms of capability expansion, reliability, flexibility, multi-target communication and cost, the advantages of antenna arrays are more obvious than those of a single large antenna. The main disadvantages of the antenna array of the traditional measurement and control system are as follows: (1) The antenna array scheme and algorithm of the traditional measurement and control system are designed for parabolic measurement and control antennas and are only applicable to single-beam, single-target measurement and control scenarios. The curved array antennas in the new generation of measurement and control systems use multi-beam forming technology to support simultaneous measurement and control of multiple targets in the entire airspace. Therefore, the antenna array technology of the traditional measurement and control system cannot meet the requirements of the multi-curved array full-airspace collaborative beamforming of the new generation of measurement and control systems. (2) Since the antenna array algorithm of the traditional measurement and control system does not use an adaptive beamforming algorithm and only forms a high-gain beam for the target, it does not have the ability to adaptively suppress airspace interference.

[0010] In other existing technical solutions, the commonly used adaptive beamforming algorithms are: Least Mean Square (LMS) algorithm, Sample Matrix Inversion (SMI) algorithm, Minimum Variance Distortionless Response (MVDR) algorithm. The characteristics of each algorithm are introduced as follows: (1) LMS adaptive beamforming algorithm. The LMS adaptive beamforming algorithm is a simple and commonly used adaptive beamforming algorithm. It is also an algorithm that uses the steepest gradient descent method and the minimum mean square error criterion in the theory of local Wiener filtering. The LMS algorithm is favored by people due to its simple algorithm structure, fewer parameters, faster convergence and easy implementation. (2) SMI adaptive beamforming algorithm. The SMI adaptive beamforming algorithm is an algorithm based on the minimum mean square error criterion. The principle of this algorithm is to directly calculate the inverse of the sampling covariance matrix. (3) MVDR adaptive beamforming algorithm. The MVDR adaptive beamforming algorithm constrains the power in the direction of the desired signal to 1 and minimizes the array output power. The MVDR adaptive beam can suppress interference and noise because the interference and noise power output by the array is minimized at this time. The main disadvantages of the current traditional adaptive beamforming technology are: (1) The LMS algorithm cannot balance the contradiction between convergence speed and steady-state error; (2) The robustness performance of the SMI algorithm will decrease when the signal-to-noise ratio increases; (3) The MVDR algorithm requires matrix inversion, which has high computational complexity and large amount of calculation, making it difficult to update the antenna array weight vector in real time; (4) For adaptive beamforming of large-scale antenna arrays, the above algorithms all have the problem of difficulty in implementation. Summary of the invention

[0011] The purpose of the present invention is to overcome the shortcomings of the prior art and provide a distributed multi-curved array full-airspace collaborative beamforming method and system. Aiming at the shortcomings of the antenna array technology of the traditional measurement and control system and facing the scenario of distributed deployment of multi-curved arrays in the new generation of measurement and control system, a distributed multi-curved array full-airspace beamforming scheme is proposed, which supports the needs of multi-beam forming and flexible resource scheduling, and solves the problem that the antenna array technology of the traditional measurement and control system can only be applied to single-beam, single-target measurement and control scenarios; at the same time, an adaptive beamforming algorithm is added to support real-time dynamic updating of beam weights, and solves the problem that the antenna array technology of the traditional measurement and control system does not have the ability to adaptively suppress airspace interference.

[0012] The object of the present invention is achieved through the following solutions:

[0013] A distributed multi-curved array full-space cooperative beamforming method includes multiple distributed curved arrays and a processing center. After the multiple distributed curved arrays are connected to the processing center respectively, parameters, control signals, receiving signals and transmitting signal data can be transmitted; wherein the processing center is used to control the multiple distributed curved arrays to complete their own beam synthesis and cooperative beam forming, and specifically includes the following steps:

[0014] S1, each curved array performs primary weighting on the antenna elements within it and synthesizes the primary beams respectively;

[0015] S2, the processing center performs secondary weighting on the signals of the distributed curved array to synthesize an equivalent large-aperture phased array antenna and form the desired adaptive beam.

[0016] Furthermore, in step S1, each curved array performs primary weighting on the antenna array elements inside thereof and synthesizes primary beams respectively, which specifically includes sub-steps: in the primary weighting stage, each curved array uses the CBF algorithm to generate the element weight vector inside thereof.

[0017] Furthermore, each curved array uses the CBF algorithm to generate its own internal array element weight vector, which specifically includes the following sub-steps:

[0018] S1-1, the processing center will expect the angle of the beam Send to each surface array;

[0019] S1-2, each curved array performs primary weighting on its internal antenna elements and synthesizes primary beams respectively; for the i-th curved array, P is activated according to the angle of the desired beam i antenna array elements, a three-dimensional rectangular coordinate system is established with the phase center of the curved array antenna as the origin, and the coordinates of the pth antenna array element are expressed as ξ p =[x p ,y p ,z p ] T , p represents the antenna element number, p=1,…,P i , x p represents the x-axis coordinate of the antenna array element p, y p represents the y-axis coordinate of the antenna element p, z p represents the z-axis coordinate of antenna element p, [·] T Represents the transpose of a matrix or vector; the i-th surface matrix is ​​based on P i The three-dimensional rectangular coordinates of the active antenna elements and the angle of the desired beam The antenna array weight vector w is calculated according to the following formula i ;

[0020]

[0021] in, g(·) is the value of the input The linear function of the i-th surface matrix uses the weight vector w i Synthesize the primary beam and output the received synthetic signal x i (k).

[0022] Furthermore, in step S2, the processing center performs secondary weighting on the signals of the distributed curved array, specifically including: in the secondary weighting stage, the processing center uses neural network fitting to generate weight vectors between the curved arrays, and can support real-time updating of the antenna array weight vectors.

[0023] Furthermore, in step S2, the processing center generates weight vectors between curved arrays by using neural network fitting, which can support real-time updating of antenna array weight vectors, and specifically includes the following sub-steps:

[0024] In the secondary weighting process, the i-th surface array transmits its own position coordinate information, delay τ i and the received signal x i (k) is sent to the processing center, the delay τ i Including fiber optic delay;

[0025] The processing center receives the parameters, data and expected beam directions of multiple curved arrays. The real vector is combined into a real number vector and input into the input layer of the neural network. After multiple layers of weighting and nonlinear processing, the neural network outputs the antenna array weight vector w in real time. NN ;

[0026] Reuse in processing center NN The signals of multiple curved arrays are weighted.

[0027] Furthermore, in step S2, the processing center performs secondary weighting on the signal of the distributed curved array, including an initialization stage, and specifically performs the following sub-steps in the initialization stage:

[0028] S2-3, the processing center receives its own position coordinate information and delay τ sent from the i-th curved array i , receive signal x i (k) and node status information;

[0029] S2-4, the processing center determines the number N of curved arrays that need to perform collaborative beamforming according to the measurement and control link budget requirement or antenna gain requirement;

[0030] S2-5, the processing center determines whether the current curved array is normal according to the parameters of each curved array, the received signal and the node status information, and selects N curved arrays that are working normally as the valid curved array set;

[0031] S2-6, after receiving the parameters and receiving signals of all normal N curved arrays, the processing center arbitrarily designates a spatial reference point as the coordinate origin and establishes a custom three-dimensional rectangular coordinate system. The processing center converts the position coordinates of each curved array into the coordinates of the custom three-dimensional rectangular coordinate system, expressed as Ω i =[x i ,y i ,z i ] T , i represents the surface array number, x i represents the x-axis coordinate of the surface array i, y i represents the y-axis coordinate of the surface array i, z i represents the z-axis coordinate of the surface array i, and the superscript T represents the transpose.

[0032] Further, in step S2-5, the step of judging whether the current curved array is normal and selecting N curved arrays that work normally as a valid curved array set specifically includes the following sub-steps:

[0033] If any of the following three conditions is met, it means that the surface array h has been destroyed or damaged, and the surface array h is discarded without processing; otherwise, the surface array h is added to the valid surface array set;

[0034] Case a), the processing center does not receive the data of the surface array h or the received data is incomplete;

[0035] Case b), the processing center receives abnormal parameters or signal data of the curved array h;

[0036] In case c), the processing center receives abnormal node status of the surface array h.

[0037] Furthermore, in step S2, the processing center performs secondary weighting on the signal of the distributed curved array, including a training phase, and the following sub-steps are specifically performed in the training phase:

[0038] S2-7, the processing center receives the three-dimensional rectangular coordinates Ω, delay τ, the receiving array signal vector x(k), and the angle of the desired beam from the N curved arrays. Calculate the antenna array weight vector w according to the adaptive beamforming algorithm opt ;

[0039]

[0040] Where Ω={Ω1,Ω2,…,Ω N} represents the three-dimensional rectangular coordinates of N curved arrays, Ω i =[x i ,y i ,z i ] T represents the three-dimensional rectangular coordinates of the i-th surface array; τ=[τ1,τ2,…,τ N ] T represents the delay vector from N curved arrays to the processing center, with a dimension of N×1, τ i represents the delay from the i-th curved array to the processing center, i=1,2,…,N, N represents the number of curved arrays; x(k)=[x1(k),x2(k),…,x N (k)] T represents the array signal vector obtained by the k-th sampling, with a dimension of N×1; represents the desired beam direction, θ0 is the horizontal azimuth angle, is the vertical pitch angle; f(·) is the angle with respect to the input Nonlinear function of

[0041] S2-8, the processing center will w opt As training samples, the neural network is trained so that the output of the neural network approaches w opt , and finally output the antenna array weight vector w NN ;

[0042] S2-9, the processing center calculates the training error vector e=w NN -w opt , and determine whether to end training; calculate the second-order norm of the error vector ‖e‖ 2 , Among them, e i represents the i-th element in the error vector, Σ(·) represents the summation function;

[0043] If the second-order norm of the error vector ||e|| 2 If it is greater than the preset threshold Threshold1, the training error vector e is fed back to the neural network to correct the parameters of the neural network, and the process jumps to step S2-7 to continue iterative training; otherwise, if the second-order norm of the error vector ||e|| 2 If it is less than or equal to the preset threshold Threshold1, the training ends.

[0044] Further, in step S2-8, the processing center w opt As training samples, the neural network is trained so that the output of the neural network approaches w opt , and finally output the antenna array weight vector w NN , specifically including the following sub-steps:

[0045] S2-8-1, the three-dimensional rectangular coordinates Ω, delay τ, receiving array signal vector x(k), and angle of the desired beam of N curved arrays are calculated. Enter the neural network;

[0046] S2-8-2, the neural network undergoes multi-layer weighting and nonlinear processing, and finally outputs the antenna array weight vector w NN , the dimension is N×1.

[0047] Further, in step S2, the processing center performs secondary weighting on the signal of the distributed curved array, including a normal working stage, and specifically performs the following sub-steps in the normal working stage:

[0048] S2-10, the processing center continuously transmits the three-dimensional rectangular coordinates Ω, delay τ, receiving array signal vector x(k), and angle of the desired beam of the N curved arrays. Input to the trained neural network, after multi-layer neural network weighting and nonlinear processing, update the output antenna array weight vector w′ NN , the dimension is N×1;

[0049] S2-11, the processing center uses the updated antenna array weight vector w′ NN The received array signal x(k) is weighted and summed to obtain the composite signal y(k):

[0050]

[0051] Thereby forming a desired adaptive receiving beam;

[0052] S2-12, the processing center uses the updated antenna array weight vector w′ NN The single-channel transmission signal s(k) is weighted and output to obtain the transmission array signal vector s(k):

[0053]

[0054] The N elements in s(k) are distributed to N curved arrays for transmission, thereby forming a desired adaptive transmission beam;

[0055] S2-13, during normal operation, if the processing center finds a new abnormal surface array h, the surface array h is removed from the valid surface array set, and then jumps to step S2-7 to retrain the neural network; otherwise, jumps to step S2-10 to continue updating parameters and iterating.

[0056] A distributed multi-curved array full-space collaborative beamforming system includes a computer device, a computer program is stored in the memory of the computer device, and when the computer program is loaded by the processor of the computer device, any of the methods described above is executed.

[0057] The beneficial effects of the present invention include:

[0058] (1) The present invention supports the unified and flexible scheduling and management capabilities of curved array resources: the processing center selects an appropriate number of curved arrays for collaborative beamforming according to the measurement and control link budget requirements or antenna gain requirements, and the processing center can uniformly schedule and manage the curved array resources according to the measurement and control task requirements, and generate multiple independent beams at the same time to communicate with multiple spacecraft in different areas of the sky, thereby achieving higher resource utilization.

[0059] (2) The present invention supports adaptive beamforming capability: Due to the addition of an adaptive beamforming algorithm, it supports adaptive beamforming and has the ability to suppress spatial interference.

[0060] (3) The present invention supports the real-time signal processing capability of distributed multi-curved array full-space collaborative beamforming: a two-level weighting method is used to respectively realize the beamforming of the curved array itself and the distributed collaborative beamforming between the curved arrays, thereby reducing the signal processing complexity of the collaborative beamforming.

[0061] (4) The present invention supports real-time updating of the antenna array weight vector: the processing center uses neural network fitting to generate the antenna array weight vector, which does not require complex operations such as matrix inversion, thereby reducing the computational complexity, reducing the amount of computation, and reducing the demand for computing resources, thereby supporting real-time updating of the antenna array weight vector.

[0062] (5) The present invention supports dynamic adaptation to real-time changes in array manifolds: the processing center uses the coordinate position and fiber delay of each curved array as the input of the neural network, and calculates and updates the antenna array weight vector in real time. It can dynamically adapt to scenarios where the array manifold changes in real time, such as scenarios where the position of the curved array changes dynamically.

[0063] (6) The present invention supports dynamic tracking and suppression of spatial interference: the processing center uses the receiving array signal vector as the input of the neural network, calculates and updates the antenna array weight vector in real time, can dynamically adapt to the change in the direction of the interference source in the received signal, and adjusts the zero point of the beam to align with the interference in real time, thereby realizing dynamic tracking and suppression of spatial interference.

[0064] (7) The present invention has partial anti-destruction capability and good system robustness: each curved array sends its own status information to the processing center, and the processing center selects N curved arrays that work normally as the effective curved array set. When a small number of curved arrays are destroyed, the processing center removes the invalid curved arrays from the effective curved array set, retrains the neural network and updates the antenna array weight vector in time, so as to continue to maintain the desired beam pattern, resulting in only slight performance loss and partial anti-destruction capability. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0066] Figure 1 It is a schematic diagram of the system composition of the present invention;

[0067] Figure 2 It is a schematic diagram of the curved array of the present invention;

[0068] Figure 3 It is a schematic diagram of the two-level weighted structure of the present invention;

[0069] Figure 4 This is the system workflow of the present invention. DETAILED DESCRIPTION

[0070] All features disclosed in all embodiments in this specification, or steps in all methods or processes implicitly disclosed, except for mutually exclusive features and / or steps, can be combined and / or expanded or replaced in any manner.

[0071] The abbreviations in this application are explained as follows:

[0072] RBFNN: Radial Basis Function Neural Network, radial basis function neural network;

[0073] LMS: Least Mean Square, minimum mean square error (beamforming algorithm);

[0074] SMI: Sample Matrix Inversion, sampling matrix inversion (beamforming algorithm);

[0075] MVDR: Minimum Variance Distortionless Response, minimum variance distortion-free response (beamforming algorithm);

[0076] CBF: Conventional Beamforming.

[0077] In the overall technical concept of the present invention, the preferred embodiment provides a distributed multi-curved array full-space collaborative beamforming method. The system of this method includes N distributed curved arrays and a processing center. The N distributed curved arrays and the processing center are connected by optical fiber, and parameters, control signals, receiving and transmitting signal data can be transmitted. The processing center controls the N distributed curved arrays to complete collaborative beamforming. First, each curved array performs a primary weighting on the antenna array elements inside it, and synthesizes the primary beams respectively. Then, the processing center performs a secondary weighting on the signals of the distributed curved arrays through a neural network algorithm. In the secondary weighting process, the i-th curved array sends its own position coordinate information, optical fiber delay τ i , receive signal x i (k) The processing center receives the above parameters, data and expected beam directions of the N curved arrays. The real vector is combined into a real number vector and input into the input layer of the neural network. After multiple layers of weighting and nonlinear processing, the neural network outputs the antenna array weight vector w in real time. NN , the processing center uses w NN The signals of N curved arrays are weighted to synthesize an equivalent large-aperture phased array antenna and form the desired adaptive beam.

[0078] The overall technical solution of the present invention centrally schedules and manages the curved array resources through a processing center, and adopts a two-level weighted approach to respectively realize the beamforming of the curved array itself and the distributed collaborative beamforming between the curved arrays, thereby reducing the complexity of signal processing for large-scale distributed array antennas. At the same time, a neural network is adopted to realize the distributed collaborative beamforming between the curved arrays, which reduces the demand for computing resources compared with traditional algorithms, supports real-time updating of antenna array weight vectors, supports dynamic adaptation to scenarios where the array manifold changes in real time, supports dynamic tracking, suppresses interference in different directions, and has partial anti-destruction capabilities.

[0079] Furthermore, the system composition and working process of the present invention are described in detail as follows in conjunction with the accompanying drawings: Figure 1As shown, the system of the present invention includes N distributed curved arrays (a first curved array 101, a second curved array 102, a third curved array 103, and an Nth curved array 104) and a processing center 105. The N distributed curved arrays are connected to the processing center 105 through optical fibers, and parameters, control signals, and receiving and transmitting signal data can be transmitted. The processing center 105 controls the N distributed curved arrays to complete collaborative beamforming. Each curved array can independently receive and transmit signals, which is equivalent to an independent antenna unit. The processing center 105 realizes the weight calculation of the N curved arrays and completes the receiving, merging, and transmitting generation of the N curved array signals.

[0080] like Figure 2 As shown, the curved array i of the present invention is a traditional centralized array antenna, and its synthetic beam points to the target direction. Among them, θ0 is the horizontal azimuth, that is, the angle between the XY plane and the positive X axis, is the vertical pitch angle, that is, the angle with the XY plane.

[0081] like Figure 3 As shown, the present invention realizes the beamforming of the curved array itself and the distributed collaborative beamforming between the curved arrays respectively through a two-level weighting method. In the first-level weighting stage, each curved array uses the CBF algorithm to generate its own internal array element weight vector. In the second-level weighting stage, the processing center uses neural network fitting to generate the weight vector between the curved arrays, supporting real-time updating of the antenna array weight vector.

[0082] like Figure 4 As shown, the system workflow of the present invention is divided into two stages:

[0083] Step 1: First-level weighting stage: Each curved array performs first-level weighting on its internal antenna elements and synthesizes primary beams respectively.

[0084] Step 2, secondary weighting stage: The processing center performs secondary weighting on the synthetic signal of the distributed curved array through a neural network algorithm, thereby synthesizing an equivalent large-aperture phased array antenna and forming the desired adaptive beam.

[0085] Among them, the secondary weighted stage can be divided into three sub-stages:

[0086] (1) Initialization phase: complete initial parameter configuration, effective surface array set screening, and surface array coordinate conversion;

[0087] (2) Training phase: completing the training of the neural network;

[0088] (3) During the normal working phase, the trained neural network is used to generate the antenna array weight vector in real time, and the antenna array weight vector is used to generate the receiving beam and the transmitting beam.

[0089] Furthermore, the system workflow of the present invention is described in detail as follows:

[0090] 1. Primary weighting stage:

[0091] (1) The processing center sets the angle of the desired beam Sent to each surface array.

[0092] (2) Each curved array performs primary weighting on the antenna elements within it and synthesizes the primary beams respectively.

[0093] Specifically, for the i-th curved array, P is activated according to the angle of the desired beam i antenna array elements, a 3D rectangular coordinate system is established with the phase center of the curved array antenna as the origin. The coordinates of the pth antenna array element are expressed as ξ p =[x p ,y p ,z p ] T , where p represents the antenna element number, p = 1, ..., P i , x p represents the x-axis coordinate of the antenna array element p, y p represents the y-axis coordinate of the antenna element p, z p represents the z-axis coordinate of antenna element p, [·] T Represents the transpose of a matrix or vector.

[0094] The i-th surface matrix is ​​based on P i The 3D rectangular coordinates of the activated antenna elements Desired beam angle Calculate the antenna array weight vector w according to the traditional beamforming algorithm i .

[0095]

[0096] in, g(·) is the value of the input A linear function of .

[0097] The i-th surface matrix uses the weight vector w i The primary beam is synthesized and the received synthesized signal xi(k) is output.

[0098] 2. Secondary weighting stage:

[0099] Initialization phase:

[0100] (3) The i-th curved array transmits its own position coordinate information and fiber delay τ i , receive signal x i(k) and node status information are sent to the processing center.

[0101] (4) The processing center determines the number N of curved arrays that need to perform collaborative beamforming based on the measurement and control link budget requirements or antenna gain requirements.

[0102] (5) The processing center determines whether the current curved array is normal based on the parameters of each curved array, the received signal, and the node status information, and selects N curved arrays that are working normally as the valid curved array set.

[0103] The processing center determines the standard of the surface array h as an abnormal node:

[0104] a) The processing center did not receive the data of the curved array h or the received data was incomplete;

[0105] b) The processing center receives abnormal parameters or signal data of the curved array h, such as parameters exceeding the normal range or signal data being all zero;

[0106] c) The processing center receives abnormal node status of the curved array h, for example, the node status displays a radio frequency module fault code;

[0107] If any of the above three points is met, it means that the surface array h has been destroyed or damaged, so the surface array h is discarded without processing; otherwise, the surface array h is added to the valid surface array set.

[0108] (6) After receiving the parameters and reception signals of all the normal N curved arrays, the processing center arbitrarily designates a spatial reference point as the coordinate origin and establishes a custom 3D rectangular coordinate system. The processing center converts the position coordinates of each curved array into the custom 3D rectangular coordinate system coordinates, expressed as Ω i =[x i ,y i ,z i ] T , i represents the number of the surface array, x i represents the x-axis coordinate of the surface array i, y i represents the y-axis coordinate of the surface array i, z i represents the z-axis coordinate of the surface array i, and the superscript T represents the transpose.

[0109] Training phase:

[0110] (7) The processing center receives the 3D rectangular coordinates Ω of the N curved arrays, the fiber delay τ, the received array signal vector x(k), and the angle of the desired beam. Calculate the antenna array weight vector w according to the adaptive beamforming algorithm opt .

[0111]

[0112] Where Ω={Ω1,Ω2,…,Ω N} represents the 3D rectangular coordinates of N curved arrays, Ω i =[x i ,y i ,z i ] T represents the 3D rectangular coordinates of the i-th surface matrix; τ=[τ1,τ2,…,τ N ] T represents the fiber delay vector from N curved arrays to the processing center, with a dimension of N×1, τ i represents the fiber delay from the i-th curved array to the processing center, i = 1, 2, ..., N, N represents the number of curved arrays; x(k) = [x1(k), x2(k), ..., x N (k)] T represents the array signal vector obtained by the k-th sampling, with a dimension of N×1; represents the desired beam direction, θ0 is the horizontal azimuth, that is, the angle between the X-axis and the X-axis in the XY plane, is the vertical pitch angle, that is, the angle with the XY plane; f(·) is the angle with respect to the input The nonlinear function of .

[0113] (8) The processing center will w opt As training samples, the neural network is trained so that the output of the neural network approaches w opt .

[0114] (8-1) The 3D rectangular coordinates Ω of the N curved arrays, the fiber delay τ, the receiving array signal vector x(k), and the angle of the desired beam are Enter neural networks.

[0115] (8-2) The neural network undergoes multi-layer weighting and nonlinear processing, and finally outputs the antenna array weight vector w NN , the dimension is N×1.

[0116] (9) The processing center calculates the training error vector e = w NN -w opt , and determine whether to end the training.

[0117] Calculate the 2nd-order norm of the error vector ||e|| 2 , Among them, e i represents the i-th element in the error vector, and Σ(·) represents the summation function.

[0118] If the second-order norm of the error vector ||e|| 2If it is greater than the preset threshold Threshold1, the training error vector e is fed back to the neural network to correct the parameters of the neural network, and the process jumps to step (7) to continue iterative training. Otherwise, if the second-order norm of the error vector ‖e 2 If it is less than or equal to the preset threshold Threshold1, the training ends.

[0119] Normal working stage:

[0120] (10) The processing center continuously transmits the 3D rectangular coordinates Ω, fiber delay τ, receiving array signal vector x(k), and the angle of the desired beam to the N curved arrays. Input to the trained neural network, after multi-layer neural network weighting and nonlinear processing, update the output antenna array weight vector w′ NN , the dimension is N×1.

[0121] (11) The processing center uses the updated antenna array weight vector w′ NN The received array signal x(k) is weighted and summed to obtain the composite signal y(k):

[0122]

[0123] Thereby forming the desired adaptive receiving beam.

[0124] (12) The processing center uses the updated antenna array weight vector w′ NN The single-channel transmission signal s(k) is weighted and output to obtain the transmission array signal vector s(k):

[0125]

[0126] The N elements in s(k) are distributed to N curved arrays for transmission, thereby forming a desired adaptive transmission beam.

[0127] (13) During normal operation, if the processing center finds a new abnormal surface array h, the surface array h is removed from the valid surface array set, and then jumps to step (7) to retrain the neural network; otherwise, jumps to step (10) to continue updating parameters and iterating.

[0128] The technical solution of the embodiment of the present invention has the following advantages:

[0129] 1) In the above technical solution of the embodiment of the present invention, the beamforming of the curved array itself and the distributed collaborative beamforming between the curved arrays are respectively realized by a two-level weighting method.

[0130] 2) In the above technical solution of the embodiment of the present invention, in the first-level weighting stage, each curved array uses the CBF algorithm to generate its own internal array element weight vector.

[0131] 3) In the above technical solution of the embodiment of the present invention, in the secondary weighting stage, the processing center uses neural network fitting to generate weight vectors between curved arrays, and supports real-time updating of antenna array weight vectors.

[0132] 4) In the above technical solution of the embodiment of the present invention, each curved array sends its own position coordinate information, fiber delay τ i , receive signal x i (k) and node status information are sent to the processing center.

[0133] 5) In the above technical solution of the embodiment of the present invention, the processing center determines the number N of curved arrays that need to perform collaborative beamforming according to the measurement and control link budget requirement or the antenna gain requirement.

[0134] 6) In the above technical solution of the embodiment of the present invention, the processing center selects N curved arrays that work normally as the effective curved array set. When some curved arrays are destroyed, the processing center removes the invalid curved arrays from the effective curved array set, retrains the neural network and updates the antenna array weight vector in time, so as to continue to maintain the desired beam pattern and have partial anti-destruction capability.

[0135] 7) In the above technical solution of the embodiment of the present invention, the standard for the processing center to judge that the surface array h is an abnormal node is:

[0136] a) In the above technical solution of the embodiment of the present invention, the processing center does not receive the data of the curved array h or the received data is incomplete;

[0137] b) In the above technical solution of the embodiment of the present invention, the processing center receives abnormal parameters or signal data of the curved array h, for example, the parameters are out of the normal range and the signal data are all zero;

[0138] c) In the above technical solution of the embodiment of the present invention, the processing center receives an abnormal node status of the curved array h, for example, the node status displays a radio frequency module fault code;

[0139] If any of the above three points is met, it means that the surface array h has been destroyed or damaged.

[0140] 8) In the above technical solution of the embodiment of the present invention, the processing center uses neural network fitting to generate the antenna array weight vector, which does not require complex operations such as matrix inversion, reduces the calculation complexity, reduces the amount of calculation, and reduces the demand for computing resources, thereby supporting real-time updating of the antenna array weight vector.

[0141] 9) In the above technical solution of the embodiment of the present invention, the processing center uses the coordinate position and fiber delay of each curved array as the input of the neural network, calculates and updates the antenna array weight vector in real time, and can dynamically adapt to the scenario where the array manifold changes in real time, such as the scenario where the position of the curved array changes dynamically.

[0142] 10) In the above technical solution of the embodiment of the present invention, the processing center uses the receiving array signal vector as the input of the neural network, calculates and updates the antenna array weight vector in real time, can dynamically adapt to the change of the direction of the interference source in the received signal, and adjusts the zero point of the beam to align with the interference in real time, thereby realizing dynamic tracking and suppressing spatial interference.

[0143] In other preferred embodiments of the present invention, the following embodiment 1 is provided:

[0144] The application system architecture or scenario of the first embodiment is that multiple curved arrays jointly form the desired beam, and the network elements involved are curved arrays and processing centers. In the first-level weighting stage, the CBF algorithm is used to generate the antenna array weight vector, and in the second-level weighting stage, the radial basis function neural network is used to approximate the MVDR algorithm to generate the antenna array weight vector. The system workflow of this embodiment is specifically described as follows:

[0145] 1. Primary weighting stage:

[0146] (1) The processing center sets the angle of the desired beam Sent to each surface array.

[0147] (2) Each curved array performs primary weighting on the antenna elements within it and synthesizes the primary beams respectively.

[0148] For the i-th curved array, activate P according to the angle of the desired beam. i antenna array elements, a 3D rectangular coordinate system is established with the phase center of the curved array antenna as the origin. The coordinates of the pth antenna array element are expressed as ξ p =[x p ,y p ,z p ] T , where p represents the antenna element number, p = 1, ..., P i , x p represents the x-axis coordinate of the antenna array element p, y p represents the y-axis coordinate of the antenna element p, z p represents the z-axis coordinate of antenna element p, [·] T Represents the transpose of a matrix or vector.

[0149] The i-th surface matrix is ​​based on P i The 3D rectangular coordinates of the activated antenna elements Desired beam angle Calculate the antenna array weight vector w according to the CBF beamforming algorithm i :

[0150]

[0151] in, Indicates that the antenna array of the i-th curved array is Array manifold vector of direction, dimension P i ×1.

[0152]

[0153] represents the desired beam direction, θ0 is the horizontal azimuth, that is, the angle between the X-axis and the X-axis in the XY plane, is the vertical pitch angle, that is, the angle with the XY plane; Unit vector representing the desired beam direction:

[0154]

[0155] exp(·) represents an exponential function with the natural constant e as the base, j represents the imaginary part of the complex number, π represents the circumference of a circle, f represents the signal frequency, c represents the speed of light, ξ p =[x p ,y p ,z p ] T represents the 3D rectangular coordinates of the pth antenna array element, p = 1, ..., P i , P i represents the number of active antenna elements of the i-th curved array, sin represents the sine function, cos represents the cosine function, [·] T Represents the transpose of a matrix or vector.

[0156] The i-th surface matrix uses the weight vector w i Synthesize the primary beam. For example, the output signal of the synthesized receive beam is:

[0157]

[0158] Among them, x i (k) represents the synthetic received signal obtained by sampling the kth time of the i-th curved array, It represents the array signal vector obtained by sampling the kth time of the i-th curved array, with a dimension of P i ×1; [·] H Represents the conjugate transpose of a matrix or vector.

[0159] 2. Secondary weighting stage:

[0160] Initialization phase:

[0161] (3) The i-th curved array transmits its own position coordinate information and fiber delay τ i , receive signal x i (k) and node status information are sent to the processing center.

[0162] (4) The processing center determines the number N of curved arrays that need to perform collaborative beamforming based on the measurement and control link budget requirements or antenna gain requirements.

[0163] (5) The processing center determines whether the current curved array is normal based on the parameters of each curved array, the received signal, and the node status information, and selects N curved arrays that are working normally as the valid curved array set.

[0164] The processing center determines the standard of the surface array h as an abnormal node:

[0165] a) The processing center did not receive the data of the curved array h or the received data was incomplete;

[0166] b) The processing center receives abnormal parameters or signal data of the curved array h, such as parameters exceeding the normal range or signal data being all zero;

[0167] c) The processing center receives abnormal node status of the curved array h, for example, the node status displays a radio frequency module fault code;

[0168] If any of the above three points is met, it means that the surface array h has been destroyed or damaged, so the surface array h is discarded without processing; otherwise, the surface array h is added to the valid surface array set.

[0169] (6) After the processing center receives the parameters and reception signals of all the normal N curved arrays, it designates the processing center as the coordinate origin and establishes a custom 3D rectangular coordinate system. The processing center converts the position coordinates of each curved array into the custom 3D rectangular coordinate system coordinates, expressed as Ω i =[x i ,y i ,z i ] T , i represents the number of the surface array, x i represents the x-axis coordinate of the surface array i, y i represents the y-axis coordinate of the surface array i, z i represents the z-axis coordinate of the surface array i, and the superscript T represents the transpose.

[0170] Training phase:

[0171] (7) The processing center receives the 3D rectangular coordinates Ω of the N curved arrays, the fiber delay τ, the received array signal vector x(k), and the angle of the desired beam. According to the traditional MVDR adaptive beamforming algorithm, the antenna array weight vector w is calculated. MVDR .

[0172] The weight vector of the MVDR beamformer is:

[0173]

[0174] Among them, R x represents the covariance matrix of the received array signal vector, R x =E(x(k)x H (k)), with a dimension of N×N; x(k) represents the array signal vector obtained by the k-th sampling, with a dimension of N×1; [·] H represents the conjugate transpose of a matrix or vector, and E(·) represents the mathematical expectation; Represents the matrix R x Seek the inverse; Indicates that the antenna array is Array of direction manifold vectors, of dimension N×1.

[0175]

[0176] represents the desired beam direction, θ0 is the horizontal azimuth, that is, the angle between the X-axis and the X-axis in the XY plane, is the vertical pitch angle, that is, the angle with the XY plane; Unit vector representing the desired beam direction:

[0177]

[0178] exp(·) represents the exponential function with the natural constant e as the base, j represents the imaginary part of the complex number, π represents the circumference of a circle, f represents the signal frequency, c represents the speed of light, Ω i =[x i ,y i ,z i ] T represents the 3D rectangular coordinates of the i-th surface array, τ i represents the fiber delay from the i-th curved array to the processing center, i = 1, 2, ..., N, N represents the number of curved arrays, sin represents the sine function, cos represents the cosine function, [·] T Represents the transpose of a matrix or vector.

[0179] (8) The processing center will w MVDR As training samples, we train the RBF neural network so that the output of the neural network approaches w MVDR .

[0180] The 3D rectangular coordinates Ω of the N curved arrays, the fiber delay τ, the receiving array signal vector x(k), and the angle of the desired beam Input the neural network, the neural network undergoes multi-layer weighting and nonlinear processing, and finally outputs the antenna array weight vector w NN .

[0181] (8-1) Input layer processing:

[0182] Input parameters and signals Combined into a large real number vector as the input layer of the RBF neural network:

[0183]

[0184] Among them, Input represents the vector of the neural network input layer, with a dimension of Q×1; Re(·) means taking the real part, Im(·) means taking the imaginary part, Q=3N+N+2N+2=6N+2.

[0185] (8-2) Hidden layer processing:

[0186]

[0187] in, Represents the output value of the mth neuron node in the hidden layer, m = 1, 2, ..., M, M is the number of neuron nodes in the hidden layer;

[0188] W (1) represents the connection weight matrix from the input layer to the hidden layer, The dimension is Q×M;

[0189] Represents the Gaussian function center of the mth neuron node, with a dimension of Q×1;

[0190] b (1) represents the bias vector from the input layer to the hidden layer, The dimension is M×1;

[0191] a (1) represents the neuron node vector of the hidden layer, The dimension is M×1;

[0192] Ψ(·) represents the radial basis function. In the embodiment, the Gaussian basis function is used, that is, Ψ(x)=exp(-x 2 );

[0193] ||x,y|| represents the distance between vectors x and y. In this embodiment, the cosine distance is used, that is,

[0194]

[0195] (8-3) Output layer processing:

[0196] a (2) =W (2) a (1) +b (2) ;

[0197] Among them, a (2) Represents the node vector of the output layer. Its training result is the antenna weight vector with a dimension of 2N×1, where N represents the number of curved arrays. Since the antenna weight vector is a complex number, its real and imaginary parts are represented separately.

[0198] W (2) Represents the connection weight matrix from the hidden layer to the output layer, with a dimension of 2N×M;

[0199] b (2) Represents the bias vector from the hidden layer to the output layer, with a dimension of 2N×1.

[0200] (8-4) Output antenna weight vector

[0201] Finally, a (2) Convert to a complex vector and get the antenna weight vector:

[0202]

[0203] Among them, w NN represents the antenna weight vector, the dimension is N×1, and j represents the imaginary part of the complex number.

[0204] (9) The processing center calculates the training error vector e = w NN -w MVDR , and determine whether to end the training.

[0205] Calculate the 2nd-order norm of the error vector ||e|| 2 , Among them, e i represents the i-th element in the error vector, and ∑(·) represents the summation function.

[0206] If the second-order norm of the error vector ||e|| 2 If it is greater than the preset threshold Threshold1, the training error vector e is fed back to the neural network to correct the parameters of the neural network, and the process jumps to step (7) to continue iterative training; otherwise, if the second-order norm of the error vector ||e|| 2 If it is less than or equal to the preset threshold Threshold1, the training ends.

[0207] Normal working stage:

[0208] (10) The processing center continuously transmits the 3D rectangular coordinates Ω, fiber delay τ, receiving array signal vector x(k), and the angle of the desired beam to the N curved arrays. Input to the trained neural network, after multi-layer neural network weighting and nonlinear processing, update the output antenna array weight vector w′ NN , the dimension is N×1.

[0209] (11) The processing center uses the updated antenna array weight vector w′ NN The received array signal x(k) is weighted and summed to obtain the composite signal y(k):

[0210]

[0211] Thereby forming the desired adaptive receiving beam.

[0212] (12) The processing center uses the updated antenna array weight vector w′ NN The single-channel transmission signal s(k) is weighted and output to obtain the transmission array signal vector s(k):

[0213]

[0214] The N elements in s(k) are distributed to N curved arrays for transmission, thereby forming a desired adaptive transmission beam.

[0215] (13) During normal operation, if the processing center finds a new abnormal surface array h, the surface array h is removed from the valid surface array set, and then jumps to step (7) to retrain the neural network; otherwise, jumps to step (10) to continue updating parameters and iterating.

[0216] Embodiment 1 implements the beamforming of the curved array itself and the distributed collaborative beamforming between the curved arrays through a two-stage weighting method, thereby reducing the complexity of signal processing implementation of large-scale distributed array antennas. At the same time, a radial basis function neural network is used to approximate the MVDR algorithm to generate antenna array weight vectors. Compared with the traditional MVDR algorithm, the computational complexity and amount are reduced, and real-time updating of the antenna array weight vector is supported. It supports dynamic adaptation to the scene of real-time changes in the array manifold, supports dynamic tracking, suppression of interference in different directions, and has partial anti-destruction capabilities.

[0217] The technical solution of the first embodiment has the following advantages:

[0218] 1) In the technical solution of the first embodiment, the beamforming of the curved array itself and the distributed collaborative beamforming between the curved arrays are respectively implemented in a two-stage weighted manner, thereby reducing the complexity of signal processing implementation of large-scale distributed array antennas.

[0219] 2) In the technical solution of the first embodiment, in the first-level weighting stage, each curved array uses the CBF algorithm to generate its own internal array element weight vector, which reduces the implementation complexity of the internal array element weight calculation of the curved array.

[0220] 3) In the technical solution of the first embodiment of the present invention, in the secondary weighted stage, the processing center uses neural network fitting to generate the antenna array weight vector, which does not require complex operations such as matrix inversion, thereby reducing the computational complexity, reducing the amount of computation, and reducing the demand for computing resources, thereby supporting real-time updating of the antenna array weight vector.

[0221] 4) In the technical solution of the first embodiment of the present invention, the processing center uses the coordinate position and fiber delay of each curved array as the input of the neural network, calculates and updates the antenna array weight vector in real time, and can dynamically adapt to the scenario where the array manifold changes in real time, such as the scenario where the position of the curved array changes dynamically.

[0222] 5) In the technical solution of the first embodiment of the present invention, the processing center uses the receiving array signal vector as the input of the neural network, calculates and updates the antenna array weight vector in real time, can dynamically adapt to the change in the direction of the interference source in the received signal, and adjusts the zero point of the beam to align with the interference in real time, thereby realizing dynamic tracking and interference suppression.

[0223] 6) In the technical solution of the first embodiment of the present invention, each curved array sends its own status information to the processing center, and the processing center selects N curved arrays that are working normally as the effective curved array set. When some curved arrays are destroyed, the processing center removes the invalid curved arrays from the effective curved array set, retrains the neural network and updates the antenna array weight vector in time to continue to maintain the desired beam pattern and have partial anti-destruction capability.

[0224] The units involved in the embodiments of the present invention may be implemented by software or hardware, and the units described may also be arranged in a processor. The names of these units do not, in some cases, limit the units themselves.

[0225] According to one aspect of an embodiment of the present invention, a computer program product or a computer program is provided, the computer program product or the computer program includes a computer instruction, and the computer instruction is stored in a computer-readable storage medium. A processor of a computer device reads the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the computer device executes the method provided in the above various optional implementations.

[0226] As another aspect, an embodiment of the present invention further provides a computer-readable medium, which may be included in the electronic device described in the above embodiment; or may exist independently without being assembled into the electronic device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by an electronic device, the electronic device implements the method described in the above embodiment.

Claims

1. A distributed multi-curved array full-space collaborative beamforming method, characterized in that: The invention comprises a plurality of distributed curved arrays and a processing center. After the plurality of distributed curved arrays are respectively connected to the processing center, parameters, control signals, receiving signals and transmitting signal data can be transmitted. The processing center is used to control the plurality of distributed curved arrays to complete their own beam synthesis and collaborative beam forming, and specifically comprises the following steps: S1, each curved array performs primary weighting on the antenna elements within it and synthesizes the primary beams respectively; S2, the processing center performs secondary weighting on the signals of the distributed curved array to synthesize an equivalent large-aperture phased array antenna and form the desired adaptive beam.

2. The distributed multi-curved array full-space collaborative beamforming method according to claim 1, characterized in that: In step S1, each curved array performs primary weighting on the antenna array elements inside each of them, and synthesizes primary beams respectively, which specifically includes sub-steps: in the primary weighting stage, each curved array uses the CBF algorithm to generate the element weight vector inside each of them.

3. The distributed multi-curved array full-space collaborative beamforming method according to claim 2, characterized in that: Each curved array uses the CBF algorithm to generate its own internal array element weight vector, which specifically includes the following sub-steps: S1-1, the processing center will expect the angle of the beam Send to each surface array; S1-2, each curved array performs primary weighting on its internal antenna elements and synthesizes primary beams respectively; for the i-th curved array, P is activated according to the angle of the desired beam i antenna array elements, a three-dimensional rectangular coordinate system is established with the phase center of the curved array antenna as the origin, and the coordinates of the pth antenna array element are expressed as ξ p =[x p ,y p ,z p ] T , p represents the antenna element number, p=1,…,P i , x p represents the x-axis coordinate of the antenna array element p, y p represents the y-axis coordinate of the antenna element p, z p represents the z-axis coordinate of antenna element p, [·] T Represents the transpose of a matrix or vector; the i-th surface matrix is ​​based on P i The three-dimensional rectangular coordinates of the active antenna elements and the angle of the desired beam The antenna array weight vector w is calculated according to the following formula i ; in, g(·) is the value of the input The linear function of the i-th surface matrix uses the weight vector w i Synthesize the primary beam and output the received synthetic signal x i (k).

4. The distributed multi-curved array full-space collaborative beamforming method according to claim 1, characterized in that: In step S2, the processing center performs secondary weighting on the signals of the distributed curved array, specifically including: in the secondary weighting stage, the processing center uses neural network fitting to generate weight vectors between the curved arrays, and can support real-time updating of the antenna array weight vectors.

5. The distributed multi-curved array full-space collaborative beamforming method according to claim 4, characterized in that: In step S2, the processing center generates weight vectors between curved arrays by using neural network fitting, which can support real-time updating of antenna array weight vectors, and specifically includes the following sub-steps: In the secondary weighting process, the i-th surface array transmits its own position coordinate information, delay τ i and the received signal x i (k) is sent to the processing center, the delay τ i Including fiber optic delay; The processing center receives the parameters, data and expected beam directions of multiple curved arrays. The real vector is combined into a real number vector and input into the input layer of the neural network. After multiple layers of weighting and nonlinear processing, the neural network outputs the antenna array weight vector w in real time. NN ; Reuse in processing center NN The signals of multiple curved arrays are weighted.

6. The distributed multi-curved array full-space collaborative beamforming method according to claim 1, characterized in that: In step S2, the processing center performs secondary weighting on the signal of the distributed curved array, including an initialization phase, in which the following sub-steps are specifically performed: S2-3, the processing center receives its own position coordinate information and delay τ sent by the i-th curved array i , receive signal x i (k) and node status information; S2-4, the processing center determines the number N of curved arrays that need to perform collaborative beamforming according to the measurement and control link budget requirement or antenna gain requirement; S2-5, the processing center determines whether the current curved array is normal according to the parameters of each curved array, the received signal and the node status information, and selects N curved arrays that are working normally as the valid curved array set; S2-6, after receiving the parameters and receiving signals of all normal N curved arrays, the processing center arbitrarily designates a spatial reference point as the coordinate origin and establishes a custom three-dimensional rectangular coordinate system. The processing center converts the position coordinates of each curved array into the coordinates of the custom three-dimensional rectangular coordinate system, expressed as Ω i =[x i ,y i ,z i ] T , i represents the surface array number, x i represents the x-axis coordinate of the surface array i, y i represents the y-axis coordinate of the surface array i, z i represents the z-axis coordinate of the surface array i, and the superscript T represents the transpose.

7. The distributed multi-curved array full-space collaborative beamforming method according to claim 6, characterized in that: In step S2-5, the process of determining whether the current curved array is normal and selecting N curved arrays that work normally as a valid curved array set specifically includes the following sub-steps: If any of the following three conditions is met, it means that the surface array h has been destroyed or damaged, and the surface array h is discarded without processing; otherwise, the surface array h is added to the valid surface array set; Case a), the processing center does not receive the data of the surface array h or the received data is incomplete; Case b), the processing center receives abnormal parameters or signal data of the curved array h; In case c), the processing center receives abnormal node status of the surface array h.

8. The distributed multi-curved array full-space collaborative beamforming method according to claim 1, characterized in that: In step S2, the processing center performs secondary weighting on the signals of the distributed curved array, including a training phase, and specifically performs the following sub-steps in the training phase: S2-7, the processing center receives the three-dimensional rectangular coordinates Ω, delay τ, the receiving array signal vector x(k), and the angle of the desired beam from the N curved arrays. Calculate the antenna array weight vector w according to the adaptive beamforming algorithm opt ; Where Ω={Ω1,Ω2,…,Ω N } represents the three-dimensional rectangular coordinates of N curved arrays, Ω i =[x i ,y i ,z i ] T represents the three-dimensional rectangular coordinates of the i-th surface array; τ=[τ1,τ2,…,τ N ] T represents the delay vector from N curved arrays to the processing center, with a dimension of N×1, τ i represents the delay from the i-th curved array to the processing center, i=1,2,…,N, N represents the number of curved arrays; x(k)=[x1(k),x2(k),…,x N (k)] T represents the array signal vector obtained by the k-th sampling, with a dimension of N×1; represents the desired beam direction, θ0 is the horizontal azimuth angle, is the vertical pitch angle; f(·) is the vertical pitch angle with respect to the input Nonlinear function of S2-8, the processing center will w opt As training samples, the neural network is trained so that the output of the neural network approaches w opt , and finally output the antenna array weight vector w NN ; S2-9, the processing center calculates the training error vector e=w NN -w opt , and determine whether to end training; calculate the second-order norm of the error vector ||e|| 2 , Among them, e i represents the i-th element in the error vector, ∑(·) represents the summation function; If the second-order norm of the error vector ‖e|| 2 If it is greater than the preset threshold Threshold1, the training error vector e is fed back to the neural network to correct the parameters of the neural network, and the process jumps to step S2-7 to continue iterative training; otherwise, if the second-order norm of the error vector ||e|| 2 If it is less than or equal to the preset threshold Threshold1, the training ends.

9. The distributed multi-curved array full-space collaborative beamforming method according to claim 8, characterized in that: In step S2-8, the processing center w opt As training samples, the neural network is trained so that the output of the neural network approaches w opt , and finally output the antenna array weight vector w NN , specifically including the following sub-steps: S2-8-1, the three-dimensional rectangular coordinates Ω, delay τ, receiving array signal vector x(k), and angle of the desired beam of N curved arrays are calculated. Enter the neural network; S2-8-2, the neural network undergoes multi-layer weighting and nonlinear processing, and finally outputs the antenna array weight vector w NN , the dimension is N×1.

10. The distributed multi-curved array full-space collaborative beamforming method according to claim 8, characterized in that: In step S2, the processing center performs secondary weighting on the signal of the distributed curved array, including a normal working stage, and specifically performs the following sub-steps in the normal working stage: S2-10, the processing center continuously transmits the three-dimensional rectangular coordinates Ω, delay τ, receiving array signal vector x(k), and angle of the desired beam of the N curved arrays. Input to the trained neural network, after multi-layer neural network weighting and nonlinear processing, update the output antenna array weight vector w′ NN , the dimension is N×1; S2-11, the processing center uses the updated antenna array weight vector w′ NN The received array signal x(k) is weighted and summed to obtain the composite signal y(k): Thereby forming a desired adaptive receiving beam; S2-12, the processing center uses the updated antenna array weight vector w′ NN The single-channel transmission signal s(k) is weighted and output to obtain the transmission array signal vector s(k): The N elements in s(k) are distributed to N curved arrays for transmission, thereby forming a desired adaptive transmission beam; S2-13, during normal operation, if the processing center finds a new abnormal surface array h, the surface array h is removed from the valid surface array set, and then jumps to step S2-7 to retrain the neural network; otherwise, jumps to step S2-10 to continue updating parameters and iterating.

11. A distributed multi-curved array full-space collaborative beamforming system, comprising a computer device, characterized in that: A computer program is stored in the memory of the computer device, and when the computer program is loaded by the processor of the computer device, the method according to any one of claims 1 to 10 is executed.

Images