Unmanned aerial vehicle cluster intelligent beam forming method and system

Abstract

The invention relates to the technical field of intelligent signal processing, in particular to an unmanned aerial vehicle cluster intelligent beamforming method and system, and the method comprises the following steps: obtaining unmanned aerial vehicle node positions, antenna deviation and synchronization errors, analyzing the influence of quantization precision on beamforming, extracting low-precision waveform features, recognizing the amplitude-phase error compensation amount, and optimizing the anti-interference performance. Phase constraints are extracted and optimized, the consistency of the beam and the expected direction is compared, parameters are adjusted to obtain an adjustment instruction set, and a beam forming maintaining state label is output. According to the method, position information, pointing deviation and synchronization errors are analyzed through joint modeling, amplitude-phase feature extraction precision is improved, interference and echo separation capability is enhanced, control robustness is improved, discrete phase constraint and alternating direction optimization are introduced, control vector solution and convergence are accelerated, and direction errors are dynamically adjusted. A retention label is generated through convergence trend analysis, beam consistency and anti-interference performance are enhanced, and high-precision stable control is realized in a complex change scene.

Description

Technical Field

[0001] The present invention relates to the field of intelligent signal processing technology, and in particular to a method and system for intelligent beamforming of unmanned aerial vehicle (UAV) clusters. Background Art

[0002] The field of intelligent signal processing encompasses multiple aspects of signal perception, analysis, transformation, and control, forming a key technological foundation within modern communications, radar, navigation, and electronic countermeasure systems. The core of this field lies in the use of mathematical modeling and algorithmic reasoning to perform real-time processing and feature extraction of received multi-source signals, thereby enabling functions such as signal enhancement, interference suppression, beam steering, and target identification. Intelligent signal processing encompasses array signal processing, beamforming, adaptive filtering, blind signal separation, and other areas, systematically encompassing the entire process from signal acquisition and digitization to model building, parameter estimation, and multi-source collaborative computing. It offers high adaptability and scalability in multi-antenna systems, dynamic network environments, and complex electromagnetic environments.

[0003] Among them, the UAV cluster intelligent beamforming method refers to a method of centralized control and directional enhancement of the group signal transmission or reception direction based on the coordination mechanism between individual nodes in the communication or detection network constructed by multiple UAV platforms through specific array configuration control strategies and phase amplitude control means. The topic mainly focuses on the problems of beam pointing consistency and interference suppression in multi-node distributed antenna arrays, covering technical matters such as array virtual modeling based on node relative position information, setting beam target direction according to mission requirements, and adjusting the amplitude and phase of individual transmission signals in combination with the minimum mean square error method. Its methods include real-time acquisition of node status information to construct an array space model, calculating the amplitude and phase control vector based on the desired direction, and forming a cluster beam by independently performing signal control on each node.

[0004] Existing technologies primarily rely on inter-node coordination mechanisms to achieve array modeling and beam pointing control. However, in the face of inaccurate position information, antenna attitude deviation, or significant time synchronization errors, the modeling results are insensitive to feedback on beam performance, resulting in reduced beamforming accuracy. During amplitude and phase control, static parameter adjustment methods are mostly used, failing to consider the impact of errors on the control vector in a low-precision quantization environment. This causes the main lobe of the beam to deviate from the target direction, weakening the sidelobe control capability and insufficient interference suppression capability. During execution, there is a lack of a dynamic feedback path for the actual amplitude and phase offset of the waveform, resulting in an inability to correct the command output according to the real-time state, leading to a directional drift problem. In complex interference scenarios, existing methods are based on preset beam pointing planning and fail to dynamically analyze the impact of errors on the shaping path, resulting in poor pointing stability and reduced anti-interference capabilities. For example, in a multi-UAV array collaborative detection mission, synchronization errors at any node cause the group beam to shift in direction, which cannot be corrected in a timely manner using current methods, ultimately affecting target recognition accuracy and energy focusing efficiency. The existing process for evaluating beam stability mainly relies on pre-mission set indicators and lacks real-time judgment methods based on output trend changes. This makes beam performance degradation likely to occur during long-term operation, affecting the cluster's continuous detection capability and communication reliability in dynamic environments. Summary of the Invention

[0005] The purpose of the present invention is to solve the shortcomings of the existing technology and propose a UAV cluster intelligent beamforming method and system.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a UAV cluster intelligent beamforming method, comprising the following steps:

[0007] S1: Obtain the corresponding position information, antenna pointing angle deviation, and time synchronization error of the nodes in the UAV cluster array. Analyze the impact of quantization accuracy on beamforming through joint modeling, extract the amplitude and phase characteristics of the low-precision quantization waveform, and combine the interference signal model and target echo model to generate the basic beamforming parameters.

[0008] S2: Based on the basic beamforming parameters, the amplitude and phase error compensation amount of the node transmission waveform is identified, and the anti-interference performance optimization direction of the low-precision quantized waveform is screened to obtain the beamforming improvement path;

[0009] S3: Based on the beamforming improvement path, the discrete phase constraint sequence parallel optimization theory is extracted, the amplitude and phase error distribution characteristics in the waveform design are analyzed, and the robust beamforming control instructions are generated by combining the alternating direction penalty optimization theory.

[0010] S4: calling the robust beamforming control instruction, comparing the consistency between the real-time beamforming output and the desired beam direction, and if there is a deviation, adjusting the beamforming parameters and the interference suppression feedback to obtain a beamforming adjustment instruction set.

[0011] As a further solution of the present invention, the basic beamforming parameters include position error range, time synchronization error range, antenna pointing deviation range, interference signal strength, and target echo strength; the beamforming improvement path includes waveform amplitude and phase error compensation amount, anti-interference performance optimization direction, and beam pointing consistency; the robust beamforming control instructions include amplitude and phase error distribution characteristics, discrete phase constraints, and alternating direction penalty method optimization results; the beamforming adjustment instruction set includes beamforming parameter adjustment amount, interference suppression feedback correction amount, and beam direction consistency correction amount.

[0012] As a further solution of the present invention, the step of obtaining the basic beamforming parameters is specifically as follows:

[0013] S111: Obtain the corresponding position information, antenna pointing angle deviation, and time synchronization error of the nodes in the UAV cluster array. Analyze the impact of quantization accuracy on beamforming through joint modeling, extract the amplitude and phase characteristics of the low-precision quantization waveform, and combine the interference signal model and target echo model to generate the basic beamforming parameters.

[0014] S112: Based on the basic beamforming parameters, a UAV cluster array antenna configuration model is constructed, the amplitude and phase error distribution characteristics of the differentiated quantization precision waveform under non-ideal conditions are analyzed, the variation pattern of the key beamforming parameters is identified, and an initial beamforming parameter set is generated;

[0015] S113: Based on the initial beamforming parameter group, combined with error analysis and the coupling relationship between multi-dimensional amplitude and phase errors of the low-precision quantized waveform, the beamforming error compensation requirement is evaluated to generate basic beamforming parameters.

[0016] As a further solution of the present invention, the step of acquiring the beamforming improved path is specifically as follows:

[0017] S211: Based on the basic beamforming parameters, extract the transmission waveforms of the UAV nodes in the cluster, calculate the amplitude and phase errors, and combine the quantitative benchmark identification deviation range to obtain the node amplitude and phase error mapping value group;

[0018] S212: Calling the node amplitude and phase error mapping value group, monitoring the interference response of the node with the largest deviation value, calculating the directional gain change and sidelobe energy distribution of the waveform under the interference environment, screening the stable waveform direction nodes by the change amplitude, and obtaining the interference steady-state direction waveform set;

[0019] S213: Based on the interference steady-state directional waveform set, waveform samples with consistent directional angle retention and frequency domain coupling performance are extracted, and the directional response interference suppression index is calculated by combining the gain distribution difference and frequency domain energy expansion of the corresponding nodes. The beamforming improved path is obtained by combining the paths pointed by the direction with the minimum index.

[0020] As a further solution of the present invention, the step of acquiring the robust beamforming control instruction is specifically as follows:

[0021] S311: Improve the path according to the beamforming, extract the discrete phase set in the path direction, convert it into a node phase sequence matrix, and determine whether the phase jump is within the limit range based on the direction interval, and establish a phase constraint equalization amount;

[0022] S312: Calling the phase constraint equalization amount, extracting the amplitude and phase errors of the node transmission waveform, normalizing the amplitude and phase errors respectively, identifying the error density fluctuation range, and obtaining an amplitude and phase offset distribution coefficient group;

[0023] S313: According to the amplitude-phase offset distribution coefficient group and in combination with the discrete phase constraint sequence, the amplitude-phase coupling value, phase adjustment amplitude and control update frequency of the node are extracted, the interference level is set as the denominator stability factor, the node gain adjustment interval is identified, the interference control intensity index is calculated, and the directional node configuration with the optimal value is used as the control basis to generate robust beamforming control instructions.

[0024] As a further solution of the present invention, the steps of acquiring the beamforming adjustment instruction set are specifically as follows:

[0025] S411: calling the robust beamforming control instruction, comparing the current beam direction output of the UAV cluster with the target beam direction, judging by the main lobe angle deviation, side lobe amplitude offset and pattern trend deviation, and obtaining the beam direction consistency deviation value;

[0026] S412: Adjust the current cluster beamforming control parameter group according to the beam direction consistency deviation value, analyze the difference in directional output amplitude and interference suppression capability under the differentiated parameter adjustment results, calculate the beam deviation adjustment metric value, select the control parameter combination, and obtain the optimal beam steering adjustment amount;

[0027] S413: Calling the optimal beam control adjustment amount, combining the current interference suppression vector feedback, reconstructing the matching relationship between the differential directional interference component amplitude and the cluster response, optimizing the beamforming adjustment sequence, and obtaining a beamforming adjustment instruction set.

[0028] As a further embodiment of the present invention, the method further comprises step S5:

[0029] S5: Based on the beamforming adjustment instruction set, collect consistency changes between the beamforming output and the desired beam direction in consecutive periods, analyze whether the change trend is in a convergence state, determine the stability period after the beamforming is executed, and output a beamforming maintenance status tag;

[0030] The beamforming maintenance status tag includes beam direction consistency, beamforming stability identifier, and interference suppression consistency.

[0031] As a further solution of the present invention, the step of acquiring the beamforming holding state tag is specifically as follows:

[0032] S511: Based on the beamforming adjustment instruction set, collect the consistency changes between the beamforming output output by the UAV cluster array antenna configuration model and the desired beam direction, extract the beam direction consistency and overlap length, analyze the variation amplitude of the overlap length within the period, and generate a periodic overlap amplitude sequence;

[0033] S512: Calling the periodic overlapping amplitude sequence, extracting the overlapping amplitude difference sequence within consecutive periods, marking the stable trend according to the polarity of the difference change, and combining it with the set beamforming stability threshold to determine whether the overlapping change converges to a single trend, thereby obtaining the beamforming fusion stability trend value;

[0034] S513: Based on the beamforming fusion stability trend value, identify the amplitude-phase error coupling relationship between the Cramer-Rao bound and the low-precision quantization waveform within the stability period, and combine the beam direction consistency length ratio to determine whether the UAV cluster is stably tracking the beamforming, and output the beamforming maintenance status label.

[0035] The UAV cluster intelligent beamforming system includes:

[0036] The array state recognition module calculates the corresponding distances and position vectors between nodes based on the coordinate data of the UAV cluster nodes. It extracts the time offset term by combining the time synchronization marker, compares the angle range between the interference signal incident and the target echo direction, analyzes the amplitude and phase discrete errors of the low-precision quantized waveform, and generates a node collaborative offset feature set.

[0037] The error path extraction module performs directional correlation analysis on the amplitude discreteness and phase offset items between nodes based on the node cooperative offset feature set, selects the path sequence with an angle with the interference direction, compares the phase mutation trends of adjacent nodes, extracts the interference suppression direction combination, and generates the interference avoidance path vector group;

[0038] The beam command generation module analyzes the node phase arrangement pattern in the direction combination based on the interference avoidance path vector group, compares the phase jump amplitudes of adjacent combinations, cross-validates the amplitude boundary nodes and the waveform main lobe angles, extracts the direction sequence with boundary continuity, and reconstructs the angle in combination with the target echo angle to generate an intelligent beam control command group;

[0039] The direction consistency comparison module, based on the intelligent beam control instruction group, performs frame-level difference analysis between the real-time direction of the beam output and the desired target direction, extracts direction segments with offset angles greater than a threshold, classifies direction change trends, analyzes the control instruction response rate and offset amplitude, and establishes a beam execution difference label set;

[0040] The beamforming state determination module is based on the beamforming execution difference label set, identifies the offset direction change curve in multiple cycles, extracts the difference range and the number of convergence cycles, performs cycle comparison, determines the start and end sections of the continuous stabilization interval, and outputs the beamforming holding state label.

[0041] Compared with the prior art, the advantages and positive effects of the present invention are:

[0042] In the present invention, by jointly modeling the node position information, antenna pointing deviation and synchronization error in the cluster, the multi-dimensional impact of quantization accuracy on beam performance can be effectively identified, and the subtle deviations of amplitude and phase can be further captured to achieve accurate separation of the features between the echo and interference waveforms, thereby improving the credibility of the basic shaping parameters. In the transmission waveform compensation stage, error identification and anti-interference direction screening are carried out in parallel, which promotes the simultaneous improvement of waveform accuracy recovery and interference avoidance. The establishment of the improved path introduces a directional optimization mechanism, which quickly corrects the directional offset caused by low-precision quantization error and enhances the robustness of the beam control process. By extracting the discrete phase constraint sequence and combining it with error distribution analysis, the efficiency of control vector solution can be improved in a discontinuous amplitude and phase control environment, and the fusion of the alternating direction penalty method improves the convergence speed and global consistency of the solution space. In the control instruction execution stage, an adaptive feedback adjustment closed loop is formed by comparing the consistency of the current output with the expected direction in real time, ensuring that the directional deviation can be quickly corrected in dynamic scenarios and enhancing the adaptability to complex interference signals. In stability judgment, through periodic analysis of convergence trends, a beam pointing tag with continuous maintenance capability is constructed to improve control accuracy and energy focusing efficiency in long-term operation. The entire process is integrated with multi-level error modeling and multi-stage feedback control to enhance the dynamic matching capability of beam consistency maintenance and anti-interference stability, showing higher execution reliability and directional control accuracy in scenarios with strong node instability and volatile synchronization errors. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is a schematic diagram of the main steps of the present invention;

[0044] Figure 2 This is a flow chart for obtaining basic beamforming parameters in the present invention;

[0045] Figure 3 This is a flow chart for obtaining the beamforming improvement path in the present invention;

[0046] Figure 4 This is a flowchart for obtaining robust beamforming control instructions in the present invention;

[0047] Figure 5 This is a flow chart for obtaining a beamforming adjustment instruction set in the present invention;

[0048] Figure 6 This is a flow chart of obtaining the beamforming holding state tag in the present invention. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0050] In the description of the present invention, it should be understood that the terms "length," "width," "up," "down," "front," "back," "left," "right," "vertical," "horizontal," "top," "bottom," "inside," "outside," and the like, indicating positions or relationships, are based on the positions or relationships shown in the accompanying drawings and are intended only to facilitate the description of the present invention and simplify the description. They do not indicate or imply that the devices or elements referred to must have a specific orientation, be constructed, or operate in a specific orientation. Therefore, they should not be construed as limiting the present invention. Furthermore, in the description of the present invention, "plurality" means two or more, unless otherwise expressly and specifically defined.

[0051] Example 1

[0052] See also Figure 1 The present invention provides a technical solution: a UAV cluster intelligent beamforming method, comprising the following steps:

[0053] S1: Obtain the corresponding position information, antenna pointing angle deviation, and time synchronization error of the nodes in the UAV cluster array. Analyze the impact of quantization accuracy on beamforming through joint modeling, extract the amplitude and phase characteristics of the low-precision quantization waveform, and combine the interference signal model and target echo model to generate the basic beamforming parameters.

[0054] S2: Based on the basic beamforming parameters, the amplitude and phase error compensation of the node's transmitted waveform is identified. By screening the anti-interference performance optimization direction of the low-precision quantized waveform, the beamforming improvement path is obtained.

[0055] S3: Based on the beamforming improvement path, the discrete phase constraint sequence parallel optimization theory is extracted to analyze the amplitude and phase error distribution characteristics in waveform design. Combined with the alternating direction penalty optimization theory, robust beamforming control instructions are generated.

[0056] S4: Call the robust beamforming control command to compare the consistency of the real-time beamforming output with the desired beam direction. If there is a deviation, adjust the beamforming parameters and the interference suppression feedback to obtain the beamforming adjustment instruction set;

[0057] S5: Based on the beamforming adjustment instruction set, collect the consistency changes between the beamforming output and the desired beam direction in consecutive cycles, analyze whether the change trend is in a convergence state, determine the stability period after the beamforming is executed, and output the beamforming maintenance status label.

[0058] The basic beamforming parameters include position error range, time synchronization error range, antenna pointing deviation range, interference signal strength, and target echo strength. The beamforming improvement path includes waveform amplitude and phase error compensation, anti-interference performance optimization direction, and beam pointing consistency. The robust beamforming control instructions include amplitude and phase error distribution characteristics, discrete phase constraints, and alternating direction penalty method optimization results. The beamforming adjustment instruction set includes beamforming parameter adjustment amount, interference suppression feedback correction amount, and beam direction consistency correction amount. The beamforming maintenance status label includes beam direction consistency, beamforming stability identifier, and interference suppression consistency.

[0059] See also Figure 2 The steps for obtaining the basic beamforming parameters are as follows:

[0060] S111: Obtain the corresponding position information, antenna pointing angle deviation, and time synchronization error of the nodes in the UAV cluster array. Analyze the impact of quantization accuracy on beamforming through joint modeling, extract the amplitude and phase characteristics of the low-precision quantization waveform, and combine the interference signal model and target echo model to generate the basic beamforming parameters.

[0061] In a drone swarm array, each node's location information, antenna pointing angle deviation, and time synchronization error significantly impact beamforming accuracy. To acquire location information, each node's three-dimensional coordinates can be obtained using GPS sensors or visual positioning. For example, node 1 has coordinates (x1, y1, z1), and node 2 has coordinates (x2, y2, z2). Antenna pointing angle deviation is measured using pointing sensors mounted on the antennas. This allows the antenna pointing angle deviation of each node to be determined. For example, the antenna angle deviation of node 1 is Δθ1, and that of node 2 is Δθ2. Time synchronization error can be detected using time synchronization equipment or protocols. For example, GPS clock synchronization can be used to correct the clock deviation of each node to ensure that the time error is within an acceptable range. Joint modeling methods, such as Kalman filtering and particle filtering, can be used to integrate the acquired location information, angle deviation, and time error to form a comprehensive error model. This model helps quantify the impact of beamforming accuracy. For example, it can analyze the beam offset between node 1 and node 2 under different time synchronization errors Δt1 and Δt2. The error model is combined with the amplitude and phase characteristics of the low-precision quantized waveform. Low-precision quantization algorithms (e.g., 8-bit quantization) are used to process the waveform and extract the amplitude and phase characteristics. By combining the interference signal model with the target echo model, the signal propagation and echo reflection from the target object can be simulated. Ultimately, basic beamforming parameters such as the center frequency and gain of the beam are generated. These basic parameters provide a basis for subsequent beamforming strategies.

[0062] S112: Based on the basic beamforming parameters, a UAV cluster array antenna configuration model is constructed. The amplitude and phase error distribution characteristics of the differentiated quantization precision waveform under non-ideal conditions are analyzed. The variation patterns of key beamforming parameters are identified, and an initial beamforming parameter set is generated.

[0063] A UAV swarm array antenna configuration model is constructed, which includes information such as the antenna layout, pointing angle, and antenna gain of each node. This construction process incorporates an array geometry model, taking into account the relative position and orientation of the antennas. The beam patterns of each node are combined to form the cluster array's composite beam, specifically including the radiation pattern and gain of each antenna. By simulating waveforms with different quantization accuracies, such as 16-bit, 8-bit, or 4-bit quantization, their performance under non-ideal conditions can be compared, and the amplitude and phase error characteristics of the different quantization accuracies can be analyzed. For example, 8-bit quantization results in reduced amplitude accuracy, causing beam direction errors, while 16-bit quantization more accurately maintains the beam direction. The analysis involves calculating the amplitude and phase errors of each node at different quantization accuracies and, using an error propagation model, assessing how these errors affect the final beam shape. This analysis identifies the variation patterns of key beamforming parameters, such as gain, beamwidth, and pointing angle, allowing the generation of an initial set of beamforming parameters, such as the initial determination of the center frequency, gain, and beamwidth.

[0064] S113: Based on the initial beamforming parameter set, combined with error analysis and the coupling relationship between multi-dimensional amplitude and phase errors of the low-precision quantized waveform, the beamforming error compensation requirement is evaluated and basic beamforming parameters are generated.

[0065] Combining error analysis methods, the multidimensional amplitude and phase errors of low-precision quantized waveforms are analyzed. This process first requires evaluating the error propagation at each node and describing the coupling relationship between amplitude error and phase error through a mathematical model. For example, amplitude error causes beam gain distortion, while phase error causes beam direction deviation. By establishing an error compensation model, the errors can be quantified and compensated. Methods such as adaptive filtering and gradient descent are used to optimize beamforming accuracy and ensure collaborative operation between nodes. During the evaluation, it is necessary to analyze the contribution of each error and its impact on beamforming. For example, Monte Carlo simulation is used to simulate the error and determine the changes in beam accuracy under different quantization accuracies. If the error exceeds the preset compensation range, for example, the beam direction error is greater than 5 degrees, compensation is considered necessary. The impact of the error can be reduced by adjusting parameters such as the antenna direction or gain. Through compensation measures, more accurate basic beamforming parameters such as adjusted antenna gain, precise beam direction, and beam width are generated. These parameters will be used for subsequent beamforming optimization.

[0066] Table 1: Example data of node position, antenna angle deviation and time synchronization error

[0067]

[0068] As shown in Table 1, the table lists the location information, antenna angle deviation, time synchronization error, and quantized accuracy of different nodes. It also lists the beam gain and beam direction error of each node. This data can be used for further error analysis and beamforming optimization in practical applications.

[0069] See also Figure 3 ,The specific steps for obtaining the beamforming improved path are:

[0070] S211: Based on the basic beamforming parameters, the transmission waveforms of the UAV nodes in the cluster are extracted, the amplitude and phase errors are calculated, and the deviation range is identified in combination with the quantitative benchmark to obtain the node amplitude and phase error mapping value group;

[0071] First, the transmission frequency band, main control direction angle and initial amplitude and phase configuration data of the nodes in the drone cluster are clarified. In practical applications, for example, a cluster of 10 drones, each node is assigned an independent initial transmission frequency of 5.8GHz carrier signal, using a fixed direction antenna, and the initial phase is set between 0° and 330°, with each 30° as a group, and the initial amplitude is based on the unit of 1.0. The amplitude error δA and phase error δφ of each node during actual transmission need to be collected and converted into a numerical sequence. For example, the amplitude of the third node should be 1.0 under ideal conditions, but the actual detection is 0.94, and the phase deviation is 7°. The amplitude and phase error is obtained as (-0.06, +7°). Repeat this operation to complete the error acquisition of 10 nodes. Then, according to the set quantization level standard, the amplitude is divided into 0.05 intervals, a total of 20 levels, and the phase is divided into 5° The interval has 72 levels in total. The quantization error offset is calculated based on the distance between it and the adjacent quantization point. For example, the amplitude of -0.06 corresponds to a quantization offset of one level, and the phase of +7° corresponds to an offset of two levels. In this way, the error of each node is uniformly coded and processed to determine whether it is within the allowable offset range. For example, the maximum tolerance of the amplitude is set to ±0.10, and the maximum tolerance of the phase is set to ±10°. If the node offset exceeds the limit, it is marked as an abnormal node. In this processing process, for the "judgment" operation, a numerical comparison is required based on the upper and lower limits of the maximum tolerance interval. For the "calculation" operation, the difference between the actual amplitude and phase values ​​and the theoretical values ​​is used for conversion and classified into the standard interval. The quantization error is obtained based on the ratio of the relative error to the level. Finally, all node mapping values ​​are output in the form of (node ​​number, amplitude error level, phase error level) to obtain the node amplitude and phase error mapping value group.

[0072] S212: Calling the node amplitude and phase error mapping value group, monitoring the interference response of the node with the largest deviation value, calculating the directional gain change and sidelobe energy distribution of the waveform under the interference environment, screening the stable waveform direction nodes by the change amplitude, and obtaining the interference steady-state direction waveform set;

[0073] The two nodes with the largest and smallest error margins are selected for comparison. In a practical project, for example, the error of node 2 is (-0.02, -3°) and the error of node 7 is (-0.12, +13°). Node 7 clearly has the largest deviation, necessitating interference response analysis. Specifically, a directional gain sweep is performed on each node in an experimental environment. The mainlobe gain values ​​in different directions are measured in 15° steps from 0° to 360°. The power distribution of the corresponding sidelobes is monitored to obtain the interference energy concentration area and record the frequency band occupancy ratio. In this process, the "calculation" operation requires the explicit use of the gain difference ΔG = Gactual – Gtheoretical to obtain the impact of directional changes. The "screening" operation is based on the interference concentration interval standard. For example, a node with a sidelobe energy ratio exceeding 30% is considered unstable, while a node with a stable direction is considered stable. Furthermore, the node's directional projection maintenance range angle is counted to determine whether it exceeds 180°. If so, the node is classified as stable. This series of processing ultimately selects waveform node combinations with stable directional responses under interference conditions, generating a set of steady-state interference directional waveforms.

[0074] S213: Based on the interference steady-state directional waveform set, extract waveform samples that have consistent directional angle retention and frequency domain coupling performance. Combined with the gain distribution difference and frequency domain energy spread of the corresponding node, the formula is used:

[0075]

[0076] Calculate the directional response interference suppression index, combine the paths in the direction pointed by the minimum index, and obtain the beamforming improved path;

[0077] Where Ψ represents the directional response interference suppression index, G i represents the directional gain in the direction of the i-th group of waveforms, is the waveform directional gain mean, Δφ i is the amplitude and phase error value of the i-th group of waveforms, σ i is the frequency domain energy expansion of the i-th group of waveforms, Q i is the quantization error amplitude of the i-th group of waveforms, and n is the number of sample waveform groups;

[0078] The directional response interference suppression index is an indicator used to evaluate the comprehensive anti-interference capability of each transmission waveform in a UAV cluster under a specific direction combination;

[0079] It is necessary to further extract waveform samples with consistent direction retention and similar frequency domain coupling, and jointly evaluate their directional gain difference, spectrum spread, quantization deviation level value and amplitude and phase error. In this process, in order to ensure the consistency of the operation participants, it is necessary to first perform a dimension unification operation and normalize the participants of different units. The specific process is as follows: First, collect the measured directional gain G of each waveform sample at the main control direction angle θi , the unit is dBi. During the measurement process, each group of waveforms is aligned with the main control direction in turn, and its main lobe peak is recorded through high-frequency antenna testing. Then the average value of multiple samples is calculated as The normalization operation is to convert G i and The difference between Form the relative deviation factor, and then obtain the frequency domain expansion σ of each group of waveforms i This parameter represents the breadth of frequency energy distribution. Taking the main frequency 5.8GHz as the center, the power density of each frequency point within ±50MHz is measured, and the standard deviation of the frequency energy is calculated. After the unit is unified into MHz, it is normalized by the maximum acceptable expansion (set to 10MHz), that is, σ i ′=σ i / 10; Furthermore, for the amplitude-phase offset value Δφ i , calculated by amplitude deviation and phase deviation, whose units are unit amplitude (dimensionless) and degree respectively. Before merging, the phase error needs to be converted into amplitude equivalent deviation. Assuming that the phase unit error weight is set to 0.01 of the amplitude deviation, that is, 1° phase error is equivalent to 0.01 unit amplitude difference, so after conversion, it is unified into a dimensionless value, and the final quantization deviation level Q i Map the error interval to the corresponding level. For example, the amplitude interval is ±0.1, which is divided into 5 levels, and the phase interval is ±15°, which is divided into 5 levels. After merging, the error weights are weighted and merged to obtain the level code.

[0080] Substitute the actual parameters into the example, take the sample number n = 4 groups of waveforms, and set them respectively:

[0081] G1=10.2dBi, G2=9.5dBi, G3=11.1dBi, G4=10.0dBi, average value

[0082] Δφ1=0.06,Δφ2=0.04,Δφ3=0.08,Δφ4=0.05 (all unified to amplitude equivalent values);

[0083] σ1=3.2MHz,σ2=2.8MHz,σ3=4.0MHz,σ4=2.0MHz, which are 0.32, 0.28, 0.40, and 0.20 respectively after normalization;

[0084] Q1=2, Q2=1, Q3=3, Q4=1;

[0085] Substituting into the formula, the calculation steps are as follows:

[0086] Item 1:

[0087] Item 2:

[0088] Item 3:

[0089] Item 4:

[0090] Total: Ψ = 0 + 2.476 + 6.618 + 0.971 = 10.065

[0091] The results show that the directional response interference suppression index Ψ under the current sample combination is 10.065. By comparing the relative values ​​of Ψ under multiple waveform combinations, the minimum value path is taken as the directional combination path with minimum interference. The corresponding direction is the UAV cluster beamforming adjustment path set. Through the joint normalization processing of directional gain deviation, amplitude and phase error and spectrum expansion, the collaborative evaluation between physical quantities of different scales is achieved, thereby improving the accuracy and stability of interference path screening.

[0092] See also Figure 4 ,The specific steps for obtaining the robust beamforming control instructions are:

[0093] S311: Improve the path based on beamforming, extract the discrete phase set in the path direction, convert it into a node phase sequence matrix, and determine whether the phase jump is within the limit based on the direction interval to establish the phase constraint equalization;

[0094] The spatial direction configuration of all nodes in the path is extracted from the cluster control, and the main transmission angle of each node in the path action direction is recorded through the antenna direction control module, and the direction area is encoded based on the preset discrete angle step. For example, 0° to 360° is divided into 72 segments, and each segment is 5° apart. The action direction of each node in the current path can be mapped to the corresponding interval code, and the phase information corresponding to the actual transmission signal of each node is further obtained. The intermediate frequency acquisition device with a sampling frequency of 5MHz is used to capture the node sending signal in real time, and the signal main frequency phase value is extracted by fast Fourier transform. The phase value of each node is combined to form a phase sequence matrix, and then the matrix is ​​rearranged according to the direction interval code to ensure that the order is consistent with the spatial layout direction of the node. To maintain consistency, the phase jump values ​​between adjacent nodes are calculated row by row in the matrix. The jump amplitude is obtained using the form of Δφ = |φn-φn-1| and compared with the phase control standard deviation threshold. If the Δφ value is within the preset control range (for example, ≤20°), the phase structure meets the beam control stability requirements. Further, by statistically analyzing the difference between the mean phase jump and the mean phase of all nodes, an inter-node phase offset balance index is constructed to determine whether the offset difference falls within the control interval of ±10°. If so, it is determined to be a balanced structure; otherwise, it is marked as unbalanced. The overall phase sequence is then aggregated and the number of valid jumps and the jump stability ratio are counted. Finally, the control stability judgment criterion is constructed with the jump frequency and stability ratio as variables to obtain the phase constrained balance value.

[0095] S312: Call the phase constraint equalizer to extract the amplitude and phase errors of the node's transmitted waveform, normalize the amplitude and phase errors respectively, identify the error density fluctuation range, and obtain the amplitude and phase offset distribution coefficient group;

[0096] The amplitude and phase error data set is extracted from the node waveform used for beamforming. The amplitude error is calculated by subtracting the ideal amplitude from the actual amplitude. The phase error is obtained by demodulating the signal carrier phase and comparing it with the ideal value. Taking the ideal amplitude as 1 and the ideal phase as 0° as an example, if the actual amplitude of a node is 0.92 and the phase is 6°, the amplitude error is -0.08 and the phase error is 6°. The unit scale is unified by normalization, and the amplitude error is directly retained. The decimal value represents the deviation. The angle error is converted by 1° to 0.01 amplitude unit, that is, 6° is converted to 0.06. The combination of the two forms an amplitude and phase offset value of 0.14. Then the above operations are performed on all node waveforms in turn to form The amplitude and phase offset vector set is further statistically analyzed for the maximum, minimum, average and standard deviation of the amplitude and phase offset values, and the density distribution interval is divided. For example, the offset values ​​are divided into [0, 0.05), [0.05, 0.10), [0.10, 0.15), etc. The number of nodes in each interval is counted, and the slope and concentration of the node density distribution curve are calculated to establish the waveform error density distribution function. Then, by setting the offset concentration judgment threshold, for example, the number of nodes in the offset concentration exceeding 60% is considered as a concentrated state, and less than 40% is considered as a discrete state, the node offset structure distribution map is constructed, and finally a multidimensional error indicator set describing the waveform error diffusion characteristics is formed to obtain the amplitude and phase offset distribution coefficient group.

[0097] S313: Based on the amplitude-phase offset distribution coefficient group and the discrete phase constraint sequence, the node amplitude-phase coupling value, phase adjustment amplitude, and control update frequency are extracted. The interference level is set as the denominator stability factor to identify the node gain adjustment interval using the formula:

[0098]

[0099] Calculate the interference control strength index, use the optimal directional node configuration as the control basis, and generate robust beamforming control instructions;

[0100] Among them, α represents the amplitude-phase error coupling value, ε is the amplitude offset value, δ is the phase adjustment amplitude, β is the update frequency, γ is the interference level, μ j is the gain adjustment interval of the jth node, m is the total number of nodes, and Ω is the interference control strength index;

[0101] The error data and directional control parameters of each node are further integrated, and the interference control intensity index Ω is constructed through the normalized value after unified dimension. First, the amplitude-phase error coupling value α is obtained for each node. The amplitude error ε is set to the ideal amplitude of 1 as the benchmark, and is obtained by the difference between the measured value of the transmitted signal and the ideal amplitude. If the measured amplitude of the node is 0.91, then ε = 1-0.91 = 0.09; then the phase error is converted, with 1° converted to 0.01 amplitude units as the benchmark. If the measured phase offset is 7°, then the phase offset δ = 0.07. After merging, α = ε + δ = 0.09 + 0.07 = 0.16 is obtained. The control update frequency β of the node is determined. The total number of control instructions issued to the node within a 1-second period in the statistical control signaling channel is used as the frequency benchmark. If the statistical value is 4 times, then β = 4;

[0102] The interference level γ needs to be converted using the signal-to-noise ratio. The signal-to-noise ratio is obtained by the carrier-to-noise ratio detection circuit in the signal receiving module. Assuming the average signal-to-noise ratio of the node's frequency band is 14 dB, the linear ratio is 25.12, and the corresponding interference level is γ = 1 / 25.12 ≈ 0.0398, which is used as the suppression coefficient in the calculation.

[0103] Then adjust the gain interval μ j Sampling is performed, and the average time difference between two consecutive beam control of each node is taken as the interval unit, in seconds. For example, nodes 1-4 are 0.02s, 0.03s, 0.01s, and 0.02s respectively. In order to avoid coupling deviation caused by inconsistent dimensions between parameters, a unified normalization operation is required to normalize ε, δ, β, γ, μ j All of them are converted into dimensionless values. The specific method is as follows:

[0104] ε itself is the normalized amplitude difference, which is satisfied;

[0105] δ is converted based on the maximum allowable phase adjustment range of 30°: δ' = δ / 30 = 0.07 / 30 ≈ 0.00233;

[0106] β is based on the maximum control frequency of 10 times / s: β ’ =4 / 10=0.4;

[0107] Since γ is in the linear form of 1 / SNR, it is directly normalized;

[0108] μ j Based on the maximum interval of 0.05s: μ1'=0.02 / 0.05=0.4, μ2'=0.03 / 0.05=0.6, μ3'=0.01 / 0.05=0.2, μ4'=0.02 / 0.05=0.4,

[0109] Substituting the above values ​​into the formula:

[0110] The specific calculations are as follows: α = 0.16, ε = 0.09, δ' ≈ 0.00233, β' = 0.4, γ = 0.0398,

[0111] Calculate the numerator:

[0112] Calculate the denominator: 0.0398 + 1.6 = 1.6398;

[0113] have to:

[0114] By normalizing multiple participating terms with different physical dimensions, such as amplitude error, phase error, frequency, and time interval, and integrating them into the same expression, and setting the denominator to the superposition of the suppression factor and the adjustment cost, the final value Ω can quantitatively reflect the coordination between controlling interference and resource overhead for each node combination. The results show that Ω = 0.0914 under the current combination. This value is used as the interference control strength index to screen the optimal beamforming control path, and is used as the main control variable in the subsequent generation of robust beamforming control instructions to determine the path combination.

[0115] See also Figure 5 , the steps for obtaining the beamforming adjustment instruction set are as follows:

[0116] S411: Call the robust beamforming control command to compare the current beam direction output of the UAV cluster with the target beam direction, and make judgments based on the main lobe angle deviation, side lobe amplitude offset, and pattern trend deviation to obtain the beam direction consistency deviation value;

[0117] Extract the current beam direction output data of each node in the UAV cluster. The direction output data can be represented by the actual beam pointing angle information measured by the phased array antenna. During the flight of the UAV cluster, each UAV is in a dynamic adjustment state, and its direction output has a certain deviation. During the execution process, first obtain the target expected beam direction data. This data is preset by the ground command station or the mission module in the mission planning. The setting angle is 35° as the forward direction, and then compared with the current beam main lobe direction angle fed back by the sensor. For example, if the current detection beam main lobe angle is 38°, the angle deviation is 3°. Secondly, detect the amplitude data of the side lobe. Suppose the left side lobe amplitude is -18dB and the target setting amplitude is -25dB, then the offset is 7dB. Finally, extract the morphological variation of the beam pattern. The displacement changes of the main lobe and side lobe in the directional pattern in the continuous frame images are monitored in a sequential frame manner, converted into angle rate change values, and the trend deviation is derived in combination with the time difference. For example, if the main lobe offset rate is 1.2° / s and the target stable trend is 0° / s, there is a trend deviation of 1.2° / s. The above angle deviation, side lobe offset, and trend deviation are accumulated or weighted superimposed to form a comprehensive deviation. In this example, the superposition value can be 11.2, indicating that there is an 11.2 unit offset in the current beam direction. Combined with the mission scenario, if the UAV cluster is in a path tracking mission across an obstacle area, the direction deviation will lead to a signal coverage blind spot. Therefore, it is necessary to complete the extraction and comprehensive processing of each directional deviation parameter in this step to obtain the beam direction consistency deviation value.

[0118] S412: Adjust the current cluster beamforming control parameter group according to the beam direction consistency deviation value, analyze the difference in directional output amplitude and interference suppression capability under the differentiated parameter adjustment results, and use the formula:

[0119]

[0120] Calculate the beam deviation adjustment metric value, select the control parameter combination, and obtain the optimal beam steering adjustment amount;

[0121] Where D represents the beam deviation adjustment value, P represents the current main lobe response amplitude, R represents the target direction response, M represents the main lobe width change, and Z k represents the directional difference residual term of the kth group, and x represents the number of directional sample groups;

[0122] Obtain the interference vector feedback parameters of the airspace where the current drone cluster is located. This parameter is output by the distributed direction perception module in the drone array based on the direction detection information. The collected information includes the azimuth angle of the interference source, the main frequency range, the instantaneous power intensity, and the dynamic change rate of the direction. The azimuth angle can be obtained by estimating the arrival angle of the received signal by the antenna array, the main frequency value is extracted by spectrum analysis, and the instantaneous power is calculated based on the amplitude after the envelope detection of the received signal and the time window integration. For example, in the current feedback information, the azimuth angle of the interference source is 60°, the main frequency value is 2.38GHz, the power intensity is 23dB, and the direction change rate is 1.1° / s. Then, the beam direction setting corresponding to the optimal beam control adjustment amount obtained in the second section is called. The main lobe center of the beam direction is set to 75°, and the deviation angle from the interference source is 15°. According to the set direction avoidance critical angle threshold of 30°, the deviation is judged to be lower than the critical value and marked as a potential interference impact area. In order to compare the interference degree and beam control adaptability, the index value D needs to be introduced to evaluate the quality of the direction control adjustment.

[0123] The parameter acquisition process is as follows:

[0124] P is the response amplitude in the current main lobe direction. This value can be obtained from the measured directivity pattern. Under the current beam adjustment amount, the output power in the main lobe direction is 1.85, normalized to 0.925 (maximum value is 2.0);

[0125] R is the target beam direction response, which is set to 2.0 by the mission planning instruction and normalized to 1.0;

[0126] M is the change in main lobe width. The current width measured by 3dB bandwidth is 14°, the target is set to 10°, the difference is 4°, and after normalization based on the maximum width of 20°, it is 0.2;

[0127] Z k is the residual sequence, which represents the direction difference values ​​measured multiple times in the current direction. Take three sample groups with Z1=0.11, Z2=0.09, and Z3=0.12. The normalization benchmark is the maximum acceptable direction residual of 0.2, and the corresponding normalized results are 0.55, 0.45, and 0.6 respectively;

[0128] Substitute the normalized values ​​into the formula for calculation:

[0129] Step 1: |PR=|0.925-1.0|=0.075;

[0130] Step 2, M 2 =0.04;

[0131] Step 3, Z k The sum is:

[0132] The fourth step is to extract the square root of

[0133] Step 5: Total D = 0.075 + 1.2806 = 1.3556;

[0134] The results show that the interference avoidance matching index of the current direction adjustment is 1.3556. If the set interference adaptability control benchmark value is 1.5, the current adjustment is still within the acceptable range, which means that the direction combination still has interference suppression adaptability. Finally, based on this judgment, the activated direction combinations in the beamforming control sequence are rearranged, and the current combination is set to the medium priority segment. A new UAV cluster beamforming direction configuration is generated according to the updated instruction set. As the output UAV cluster beam adjustment instruction set, the adjustment evaluation value is constructed by jointly calculating the direction response difference, the main lobe morphology change, and multiple difference residuals. This expands the direction avoidance evaluation from a single parameter judgment to a multi-factor fusion evaluation, thereby enhancing the robustness of the direction adaptation judgment.

[0135] S413: Calling the optimal beam steering adjustment value, combining the current interference suppression vector feedback, reconstructing the matching relationship between the amplitude of the interference component in the differentiated direction and the cluster response, optimizing the beamforming adjustment sequence, and obtaining a beamforming adjustment instruction set;

[0136] Combined with the feedback directional interference component vector, the interference information is obtained through array signal detection and interference source azimuth analysis. For example, a strong interference signal is detected at 45° north-east of the route, and its main frequency band is located in an area close to the communication frequency band. The amplitude value, frequency jitter range and angle change trend of the directional interference component are extracted to be 20dB, ±3MHz and 0.8° / s respectively. The directional interference feature is then combined with the current direction adjustment parameter for analysis. By constructing a comparative relationship, it is determined whether the current direction parameter has the ability to avoid interference. For example, if the current beam direction is adjusted to point to the east, it deviates from the interference direction by 45°. , which belongs to the effective avoidance interval. On the contrary, if the deviation is less than 20°, it is marked as a high interference risk, and the interference amplitude is judged whether it exceeds the interference tolerance amplitude benchmark value set by the system. The benchmark value is set to 15dB. If the current interference is 20dB, it exceeds the limit, triggering the reconstruction of the direction control sequence, adjusting the direction activation order in the original beamforming strategy, and rearranging the directional priority to achieve the strongest suppression in the direction with the highest interference intensity. In the mission scenario, if the UAV cluster is cruising in the enemy communication interference area, this step processing helps to update the direction command through real-time interference feedback, and finally obtain the UAV cluster beam adjustment instruction set.

[0137] See also Figure 6 ,The specific steps for obtaining the beamforming hold state tag are:

[0138] S511: Based on the beamforming adjustment instruction set, collect the consistency changes between the beamforming output output by the UAV cluster array antenna configuration model and the desired beam direction, extract the beam direction consistency and overlap length, analyze the variation amplitude of the overlap length within the period, and generate a periodic overlap amplitude sequence;

[0139] By monitoring the beamforming output of the antenna configuration model of the drone swarm array, the consistency of the beam output direction with the desired beam direction within each cycle can be determined. The beam output direction can be compared with the desired beam direction by comparing the actual radiation pattern of the antenna array. Methods such as angular error can be used to quantify directional consistency. For example, if the beam output direction of one node deviates from the desired direction by 0.5° within a cycle, while the deviation of another node is 0.8°, the beam directional consistency and overlap length are calculated. The overlap length is the length of the portion of the beam radiation range that overlaps with the desired beam and can be calculated using spatial coordinates and angles. For example, if the actual beam radiation range within a cycle is 20° and the desired beam is 18°, the overlap is 18°, i.e., the overlap length is 18°. Analyzing the variation in the overlap length over multiple cycles requires recording and comparing the overlap length for each cycle and statistically analyzing the variation between cycles. If the overlap length fluctuates greatly within a cycle, it may indicate poor stability, while small fluctuations indicate relative stability. Through the data, a cycle overlap amplitude sequence is generated. For example, the overlap amplitude of cycle 1 can be obtained as 0.9 degrees, and that of cycle 2 as 0.7 degrees, and so on. Finally, a sequence is constructed to reflect the changing trend of the overlap length.

[0140] S512: Call the periodic overlapping amplitude sequence, extract the overlapping amplitude difference sequence within consecutive periods, mark the stable trend according to the polarity of the difference change, and combine it with the set beamforming stability threshold to determine whether the overlapping change converges to a single trend, and obtain the beamforming fusion stability trend value;

[0141] The overlap amplitude difference sequence of consecutive cycles in this sequence is further extracted. By calculating the overlap amplitude difference between each pair of adjacent cycles, a difference sequence is obtained. For example, if the overlap amplitude of cycle 1 is 0.9 degrees and that of cycle 2 is 0.7 degrees, the difference is 0.2 degrees; the difference between cycles 2 and 3 is 0.1 degrees. Based on the polarity of the difference change, the stability trend is marked. For example, when the change in the difference sequence tends to zero (i.e., the difference is close to zero or the amplitude of change is small), it can be considered stable. Conversely, if the difference fluctuates greatly, it is marked as an unstable trend. Based on a set beamforming stability threshold (for example, set to 0.1 degrees), the overlap amplitude difference is determined to determine whether it has converged to a single trend. If the difference change is less than the threshold, the beamforming within that cycle is considered stable. If it is greater than the threshold, further optimization parameters need to be adjusted. Through this process, the beamforming fusion stability trend value is ultimately obtained. This value reflects the stability of the beamforming of the entire drone cluster. The closer the value is to 1, the more stable the beamforming is.

[0142] S513: Based on the beamforming fusion stability trend value, the coupling relationship between the Cramer-Rao bound and the amplitude and phase error of the low-precision quantization waveform within the stability period is identified. Combined with the beam direction consistency length ratio, it is determined whether the UAV cluster is stably tracking the beamforming and outputs the beamforming hold status label.

[0143] Based on the beamforming fusion stability trend value, it can be determined whether it has entered a stable state. This is analyzed by combining the Cramer-Rao bound with the amplitude and phase error coupling relationship of the low-precision quantized waveform. The Cramer-Rao bound is a theoretical limit that describes signal accuracy. Low-precision quantized waveforms will cause amplitude and phase errors, so it is necessary to analyze how the errors affect each other. For example, when the amplitude error is greater than the set threshold (such as 0.5dB) and the phase error exceeds 0.2°, the error in the beam direction will increase, thereby affecting tracking accuracy. The beam direction consistency length ratio is combined for judgment. The beam direction consistency length ratio refers to the ratio of the actual beam direction consistency to the expected beam direction consistency. If this ratio is low (for example, below 0.7), it indicates poor stability. Through comprehensive analysis, it is determined whether the drone cluster is stably tracking the beamforming. For example, if the fused stability trend value of the beamforming within the stability period is 1 and the beam direction consistency length ratio is greater than 0.8, it can be considered stable and can continuously track the beamforming. The beamforming maintenance status label is output. If the stability is good, "stable tracking" is output; if the beamforming fluctuates significantly, "unstable tracking" is output.

[0144] An unmanned aerial vehicle (UAV) swarm intelligent beamforming system, the system comprising:

[0145] The array state recognition module calculates the corresponding distances and position vectors between nodes based on the coordinate data of the UAV cluster nodes. It extracts the time offset term by combining the time synchronization marker, compares the angle range between the interference signal incident and the target echo direction, analyzes the amplitude and phase discrete errors of the low-precision quantized waveform, and generates a node collaborative offset feature set.

[0146] The error path extraction module performs directional correlation analysis on the amplitude discreteness and phase offset items between nodes based on the node cooperative offset feature set, selects the path sequence with an angle with the interference direction, compares the phase mutation trends of adjacent nodes, extracts the interference suppression direction combination, and generates the interference avoidance path vector group;

[0147] The beam command generation module analyzes the node phase arrangement pattern in the direction combination based on the interference avoidance path vector group, compares the phase jump amplitudes of adjacent combinations, cross-validates the amplitude boundary nodes and the waveform main lobe angles, extracts the direction sequence with boundary continuity, and reconstructs the angle in combination with the target echo angle to generate an intelligent beam control command group;

[0148] The direction consistency comparison module, based on the intelligent beam control instruction group, performs frame-level difference analysis between the real-time direction of the beam output and the desired target direction, extracts direction segments with offset angles greater than a threshold, classifies direction change trends, analyzes the control instruction response rate and offset amplitude, and establishes a beam execution difference label set;

[0149] The beamforming state determination module identifies the offset direction change curve within multiple cycles based on the beam execution difference label set, extracts the difference range and the number of convergence cycles, performs cycle comparison, determines the start and end sections of the continuous stabilization interval, and outputs the beamforming holding state label.

[0150] The above are merely preferred embodiments of the present invention and do not limit the present invention in any other form. Any technician familiar with the profession may use the technical content disclosed above to change or modify it into an equivalent embodiment with equivalent changes and apply it to other fields. However, any simple modification, equivalent change and modification made to the above embodiment based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. A UAV swarm intelligent beamforming method, characterized in that: The following steps are involved: S1: Obtain the corresponding position information, antenna pointing angle deviation, and time synchronization error of the nodes in the UAV cluster array. Analyze the impact of quantization accuracy on beamforming through joint modeling, extract the amplitude and phase characteristics of the low-precision quantization waveform, and combine the interference signal model and target echo model to generate the basic beamforming parameters. S2: Based on the basic beamforming parameters, the amplitude and phase error compensation amount of the node transmission waveform is identified, and the anti-interference performance optimization direction of the low-precision quantized waveform is screened to obtain the beamforming improvement path; S3: Based on the beamforming improvement path, the discrete phase constraint sequence parallel optimization theory is extracted, the amplitude and phase error distribution characteristics in the waveform design are analyzed, and the robust beamforming control instructions are generated by combining the alternating direction penalty optimization theory. S4: calling the robust beamforming control instruction, comparing the consistency between the real-time beamforming output and the desired beam direction, and if there is a deviation, adjusting the beamforming parameters and the interference suppression feedback to obtain a beamforming adjustment instruction set.

2. The UAV swarm intelligent beamforming method according to claim 1, characterized in that: The basic beamforming parameters include position error range, time synchronization error range, antenna pointing deviation range, interference signal strength, and target echo strength. The beamforming improvement path includes waveform amplitude and phase error compensation amount, anti-interference performance optimization direction, and beam pointing consistency. The robust beamforming control instructions include amplitude and phase error distribution characteristics, discrete phase constraints, and optimization results of the alternating direction penalty method. The beamforming adjustment instruction set includes beamforming parameter adjustment amount, interference suppression feedback correction amount, and beam direction consistency correction amount.

3. The UAV swarm intelligent beamforming method according to claim 1, characterized in that: The steps for obtaining the basic beamforming parameters are as follows: S111: Obtain the corresponding position information, antenna pointing angle deviation, and time synchronization error of the nodes in the UAV cluster array. Analyze the impact of quantization accuracy on beamforming through joint modeling, extract the amplitude and phase characteristics of the low-precision quantization waveform, and combine the interference signal model and target echo model to generate the basic beamforming parameters. S112: Based on the basic beamforming parameters, a UAV cluster array antenna configuration model is constructed, the amplitude and phase error distribution characteristics of the differentiated quantization precision waveform under non-ideal conditions are analyzed, the variation pattern of the key beamforming parameters is identified, and an initial beamforming parameter set is generated; S113: Based on the initial beamforming parameter group, combined with error analysis and the coupling relationship between multi-dimensional amplitude and phase errors of the low-precision quantized waveform, the beamforming error compensation requirement is evaluated to generate basic beamforming parameters.

4. The UAV swarm intelligent beamforming method according to claim 3, characterized in that: The steps for obtaining the beamforming improved path are specifically as follows: S211: Based on the basic beamforming parameters, extract the transmission waveforms of the UAV nodes in the cluster, calculate the amplitude and phase errors, and combine the quantitative benchmark identification deviation range to obtain the node amplitude and phase error mapping value group; S212: Calling the node amplitude and phase error mapping value group, monitoring the interference response of the node with the largest deviation value, calculating the directional gain change and sidelobe energy distribution of the waveform under the interference environment, screening the stable waveform direction nodes by the change amplitude, and obtaining the interference steady-state direction waveform set; S213: Based on the interference steady-state directional waveform set, waveform samples with consistent directional angle retention and frequency domain coupling performance are extracted, and the directional response interference suppression index is calculated by combining the gain distribution difference and frequency domain energy expansion of the corresponding nodes. The beamforming improved path is obtained by combining the paths pointed by the direction with the minimum index.

5. The UAV swarm intelligent beamforming method according to claim 4, characterized in that: The steps for obtaining the robust beamforming control instruction are specifically as follows: S311: Improve the path according to the beamforming, extract the discrete phase set in the path direction, convert it into a node phase sequence matrix, and determine whether the phase jump is within the limit range based on the direction interval, and establish a phase constraint equalization amount; S312: Calling the phase constraint equalization value, extracting the amplitude and phase errors of the node transmission waveform, normalizing the amplitude and phase errors respectively, identifying the error density fluctuation range, and obtaining an amplitude and phase offset distribution coefficient group; S313: According to the amplitude-phase offset distribution coefficient group and in combination with the discrete phase constraint sequence, the amplitude-phase coupling value, phase adjustment amplitude and control update frequency of the node are extracted, the interference level is set as the denominator stability factor, the node gain adjustment interval is identified, the interference control intensity index is calculated, and the directional node configuration with the optimal value is used as the control basis to generate robust beamforming control instructions.

6. The UAV swarm intelligent beamforming method according to claim 5, characterized in that: The steps for obtaining the beamforming adjustment instruction set are specifically as follows: S411: calling the robust beamforming control instruction, comparing the current beam direction output of the UAV cluster with the target beam direction, judging by the main lobe angle deviation, side lobe amplitude offset and pattern trend deviation, and obtaining the beam direction consistency deviation value; S412: Adjust the current cluster beamforming control parameter group according to the beam direction consistency deviation value, analyze the difference in directional output amplitude and interference suppression capability under the differentiated parameter adjustment results, calculate the beam deviation adjustment metric value, select the control parameter combination, and obtain the optimal beam steering adjustment amount; S413: Calling the optimal beam control adjustment amount, combining the current interference suppression vector feedback, reconstructing the matching relationship between the differential directional interference component amplitude and the cluster response, optimizing the beamforming adjustment sequence, and obtaining a beamforming adjustment instruction set.

7. The UAV swarm intelligent beamforming method according to claim 1, characterized in that: The method further comprises step S5: S5: Based on the beamforming adjustment instruction set, collect consistency changes between the beamforming output and the desired beam direction in consecutive periods, analyze whether the change trend is in a convergence state, determine the stability period after the beamforming is executed, and output a beamforming maintenance status tag; The beamforming maintenance status tag includes beam direction consistency, beamforming stability identifier, and interference suppression consistency.

8. The UAV swarm intelligent beamforming method according to claim 7, characterized in that: The steps for obtaining the beamforming holding status tag are specifically as follows: S511: Based on the beamforming adjustment instruction set, collect the consistency changes between the beamforming output output by the UAV cluster array antenna configuration model and the desired beam direction, extract the beam direction consistency and overlap length, analyze the variation amplitude of the overlap length within the period, and generate a periodic overlap amplitude sequence; S512: Calling the periodic overlapping amplitude sequence, extracting the overlapping amplitude difference sequence within consecutive periods, marking the stable trend according to the polarity of the difference change, and combining it with the set beamforming stability threshold to determine whether the overlapping change converges to a single trend, thereby obtaining the beamforming fusion stability trend value; S513: Based on the beamforming fusion stability trend value, identify the amplitude-phase error coupling relationship between the Cramer-Rao bound and the low-precision quantization waveform within the stability period, and combine the beam direction consistency length ratio to determine whether the UAV cluster is stably tracking the beamforming, and output the beamforming maintenance status label.

9. A UAV swarm intelligent beamforming system, characterized by: The system is used to implement the UAV cluster intelligent beamforming method according to any one of claims 1 to 8, and the system includes: The array state recognition module calculates the corresponding distances and position vectors between nodes based on the coordinate data of the UAV cluster nodes. It extracts the time offset term by combining the time synchronization marker, compares the angle range between the interference signal incident and the target echo direction, analyzes the amplitude and phase discrete errors of the low-precision quantized waveform, and generates a node collaborative offset feature set. The error path extraction module performs directional correlation analysis on the amplitude discreteness and phase offset items between nodes based on the node cooperative offset feature set, selects the path sequence with an angle with the interference direction, compares the phase mutation trends of adjacent nodes, extracts the interference suppression direction combination, and generates the interference avoidance path vector group; The beam command generation module analyzes the node phase arrangement pattern in the direction combination based on the interference avoidance path vector group, compares the phase jump amplitudes of adjacent combinations, cross-validates the amplitude boundary nodes and the waveform main lobe angles, extracts the direction sequence with boundary continuity, and reconstructs the angle in combination with the target echo angle to generate an intelligent beam control command group; The direction consistency comparison module, based on the intelligent beam control instruction group, performs frame-level difference analysis between the real-time direction of the beam output and the desired target direction, extracts direction segments with offset angles greater than a threshold, classifies direction change trends, analyzes the control instruction response rate and offset amplitude, and establishes a beam execution difference label set; The beamforming state determination module identifies the offset direction change curve within multiple cycles based on the beam execution difference label set, extracts the difference range and the number of convergence cycles, performs cycle comparison, determines the start and end sections of the continuous stabilization interval, and outputs the beamforming holding state label.

Images