Phased-array antenna amplitude-phase compensation method fusing genetic algorithm and rotation vector

Abstract

The invention specifically discloses a phased-array antenna amplitude-phase compensation method fusing a genetic algorithm and a rotation vector, and relates to the technical field of antenna calibration. The method comprises the following steps: S1, acquiring amplitude-phase response of each unit of the phased-array antenna in a cover-free state and a cover state to obtain a phase distortion matrix; s2, constructing an electromagnetic disturbance model, and establishing a function relationship between phase distortion and space coordinates; s3, constructing an array response model based on a rotation vector, and representing the excitation weight of each unit as a complex form; s4, performing global optimization by adopting a genetic algorithm; s5, decoding the optimal individual into an amplitude-phase excitation value of each channel, and outputting an optimal amplitude-phase parameter; and S6, re-testing the directional diagram, and evaluating the zero-depth recovery condition and the main lobe offset. The advantages of the genetic algorithm and the rotating vector method are fused, the beam pointing error is small, the convergence speed is high, multi-frequency-point, multi-polarization and full-beam-position automatic testing can be achieved, and remarkable strategic and economic benefits are achieved.

Description

Technical Field

[0001] This invention relates to the field of antenna calibration technology, and in particular to a phased array antenna amplitude and phase compensation method that integrates genetic algorithm and rotation vector. Background Technology

[0002] Phased array antennas, with their fast beam scanning, multi-beamforming, high gain, and strong anti-interference capabilities, have been widely used in various electronic information fields such as radar, communication, and navigation. To adapt to the aerodynamic performance, thermal protection, or structural installation requirements of different application scenarios, radomes have gradually evolved into irregularly shaped structures such as conical radomes and lifting body radomes, with some radomes integrating heat insulation components to ensure stable antenna operation in complex environments. However, the integrated design of the radome and the phased array antenna places higher demands on antenna amplitude and phase compensation technology. As a key aspect of ensuring the performance of phased array antennas, amplitude and phase compensation aims to offset phase distortion caused by various factors, ensuring that the pointing accuracy and null depth of the antenna pattern meet design standards.

[0003] Currently, the mainstream phase compensation methods for phased array antennas in the industry are represented by the rotating vector method and the gradient descent method. The traditional rotating vector method is based on the electromagnetic propagation characteristics of conventionally shaped radomes. Its core idea is to achieve amplitude and phase parameter calibration through vector rotation and scaling. However, this method has weak adaptability to phase distortion, especially in the case of irregularly shaped radomes. Due to the large incident angle and thick edge fins of the radome, strong nonlinear and large-amplitude phase distortion will be caused. The refraction and reflection effects of electromagnetic waves at the interfaces of multiple media such as air, radome, and heat shield are more complex. The traditional rotating vector method is difficult to accurately describe such complex distortion laws, resulting in insufficient compensation accuracy and inability to effectively correct beam pointing deviation and zero-depth descent problems in the radiation pattern.

[0004] Gradient descent, a commonly used optimization-based compensation algorithm, gradually approximates the optimal amplitude and phase parameters through iterative optimization. However, this algorithm suffers from significant convergence speed limitations. When facing complex phase distortion scenarios, it requires over 450 iterations to achieve a basic compensation effect and is prone to getting trapped in local optima, making it difficult to meet the rapid calibration requirements of engineering applications. Furthermore, the overall compensation effect of existing amplitude and phase compensation methods is limited. After calibration, the antenna null depth index typically only recovers to around 25dB, far below the design level of 40dB without a shield. The radiation pattern distortion problem remains unresolved, affecting the antenna's target detection accuracy and signal resolution.

[0005] Meanwhile, existing compensation methods lack adaptability and engineering practicality. On the one hand, there is a lack of dedicated calibration algorithms for irregularly shaped radomes, making it impossible to match the changes in electromagnetic propagation characteristics caused by the unique structure of irregularly shaped radomes. On the other hand, traditional compensation methods have low integration with testing procedures, making it difficult to adapt to the massive testing requirements brought about by the characteristics of phased array antennas such as large beam scanning range, wide bandwidth, dual polarization, and multiple wave positions. This results in cumbersome and inefficient testing and compensation procedures, failing to meet the engineering application requirements in equipment development, finalization, and mass production, becoming a key technical bottleneck restricting the full performance of phased array antennas. Summary of the Invention

[0006] The purpose of this invention is to propose a phased array antenna amplitude and phase compensation method that integrates genetic algorithm and rotation vector to solve the problems of phased array antenna pattern performance degradation caused by irregular radomes, poor adaptability of existing compensation algorithms, slow convergence speed and low testing efficiency. This method achieves high-precision, fast amplitude and phase compensation that is adaptable to complex irregular radome scenarios and meets the needs of engineering applications.

[0007] To achieve the above objectives, this invention proposes a phase compensation method for phased array antennas that integrates genetic algorithms and rotation vectors. The specific steps are as follows: Step S1: Collect the amplitude and phase responses of each element of the phased array antenna in both uncovered and covered states to obtain the phase distortion matrix; Step S2: Based on the collected amplitude and phase data, construct an electromagnetic disturbance model and establish a functional relationship between phase distortion and spatial coordinates; Step S3: Construct an array response model based on rotation vectors, representing the excitation of each antenna element as a vector on the complex plane; Step S4: Construct a genetic algorithm for global optimization. The genetic algorithm takes minimizing beam pointing error as its core objective, combines pattern fidelity and sidelobe level to design a fitness function, and uses the rotation vector method to calculate the trigonometric function terms in the fitness function. Step S5: Decode the optimal individual into the amplitude and phase excitation values ​​of each channel, and output the optimal amplitude and phase parameters; Step S6: Retest the radiation pattern, evaluate the zero-depth recovery and main lobe offset, verify the compensation effect, and optimize through a closed-loop feedback mechanism.

[0008] Preferably, in step S1, the phase distortion matrix formula is as follows: ; in, The phase distortion matrix, For the first i Line number j Phase distortion value of column cell, i , j It is an integer.

[0009] Preferably, in step S2, the functional relationship between phase distortion and spatial coordinates is as follows: ; in, , , , , These are the fitting coefficients. For unit coordinates, It is a wavenumber vector.

[0010] Preferably, in step S3, the rotating vector array response model includes the combined effects of radome distortion, element mutual coupling, and temperature drift. This is achieved through fitting calibration data. k Each antenna element is in the direction The rotational vector response is: ; in, For the first k Each antenna element is in the direction Rotational vector response under the following conditions For the first The incentive magnitude of each unit, For the first The original phase of each unit, The phase distortion is caused by the radome. The phase difference introduced by the mutual coupling effect between units. The phase difference is caused by temperature drift. For the first The radiation pattern function of each antenna element; The total array response is: ; in, For the total array response, This represents the total number of elements in the array antenna. For the first The position vector of each unit in space. This is the phase term caused by the path difference.

[0011] Preferably, in the array response model of the rotation vector, the first... i The compensated excitation of each unit is represented as follows: ; in, For the first i The excitation value after compensation for each unit For the unit excitation amplitude, For the original phase, The compensation phase to be optimized; Compensation phase to be optimized satisfy: ; in, This is the maximum amplitude limit for the hardware phase shifter.

[0012] Preferably, in step S4, the genetic algorithm includes encoding, population initialization, fitness function definition, genetic operations, and elite retention strategy; The coding uses real numbers, with each individual representing a group. Parameter combinations; The initial population size is set to 50-100, and the initial incentive weight parameters are randomly generated.

[0013] The termination condition for a genetic algorithm is: the fitness improvement is less than 1e for 10 consecutive generations. -4 Or the number of iterations reaches 200 generations.

[0014] Preferably, the fitness function is calculated using the following formula: ; in, For fitness value, This represents the actual beam pointing angle corresponding to the current individual in the population. For the desired beam pointing angle, The preset beam pointing error tolerance has a range of values. , For sidelobe level, These are the weighting coefficients.

[0015] Preferably, the genetic operations include selection, crossover, and mutation; wherein the selection operation adopts tournament selection, the crossover operation adopts simulated binary crossover, and the mutation operation adopts polynomial mutation.

[0016] The preferred elite retention strategy is to retain the top 10% of the best individuals in each generation to accelerate the convergence speed and prevent premature convergence.

[0017] Preferably, in step S6, the closed-loop feedback mechanism is as follows: if the error exceeds the limit after retesting, the local fine-tuning genetic algorithm is started to reduce the search space to accelerate convergence and form a closed-loop adaptive compensation.

[0018] Therefore, this invention proposes a phase compensation method for phased array antennas that integrates genetic algorithms and rotation vectors, with the following beneficial effects: (1) This invention integrates genetic algorithm and rotating vector method, combined with electromagnetic disturbance model to accurately describe phase distortion, and achieves zero-depth index recovery from 15dB to 40dB, consistent with the level before the cover, and beam pointing error from 8° Reduced to 0.03 ° The following method effectively solves the problem of radiation pattern performance degradation caused by irregularly shaped shields.

[0019] (2) By introducing an elite retention strategy, the convergence speed of the genetic algorithm is increased by 3 times compared with the traditional gradient descent method. The number of iterations is only 200 generations, which is far lower than the 450 generations or more of the existing technology, thus meeting the needs of rapid calibration in engineering.

[0020] (3) This invention is applicable to complex radome scenarios such as large incident angle, thick edge strip wing, and pointed cone shape, filling the gap in China where there is no phased array pointing accuracy calibration with radome.

[0021] (4) This invention can be integrated into the rapid calibration and testing equipment for irregularly shaped domes of missile-borne phased array radars, realizing automated testing of multiple frequency points, multiple polarizations, and all wave positions. Combined with the fully closed-loop hardware trigger control data acquisition method, the testing efficiency is improved by tens of times compared with the traditional solution. It can be widely used in new generation missiles, aircraft and radar equipment, and has significant strategic value and economic benefits.

[0022] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0023] Figure 1 This is a flowchart of a phased array antenna amplitude and phase compensation method that integrates genetic algorithm and rotation vector according to the present invention. Detailed Implementation

[0024] To make the technical solutions, advantages, and objectives of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below. The described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the described embodiments of the present invention without creative effort are within the protection scope of the present invention.

[0025] Unless otherwise defined, the technical or scientific terms used in this invention shall have the ordinary meaning as understood by one of ordinary skill in the art to which this invention pertains.

[0026] Example like Figure 1 As shown, this invention provides a phase compensation method for phased array antennas that integrates genetic algorithms and rotation vectors. The specific steps are as follows: Step S1: Using a quad-arm helical antenna guided by a six-axis robotic arm, the amplitude and phase responses of the 256-channel phased array antenna are acquired in both uncovered and covered states. The operating bandwidth is 4GHz, and the mode is dual-polarized. The phase distortion matrix is ​​obtained. The formula for the phase distortion matrix is ​​as follows: ; in, The phase distortion matrix, For the first i Line number j Phase distortion value of column cell, i , j It is an integer.

[0027] Step S2: Based on the collected amplitude and phase data, construct an electromagnetic disturbance model and establish a functional relationship between phase distortion and spatial coordinates to predict the distortion trend of unmeasured points; the functional relationship between phase distortion and spatial coordinates is as follows: ; in, , , , , These are the fitting coefficients. For unit coordinates, It is a wavenumber vector.

[0028] Step S3: Construct an array response model based on rotation vectors, representing the excitation of each antenna element as a vector on the complex plane; The rotating vector array response model incorporates the combined effects of radome distortion, element coupling, and temperature drift. Through fitting with calibration data, the first... k Each antenna element is in the direction The rotational vector response is: ; in, For the first k Each antenna element is in the direction Rotational vector response under the following conditions For the first The incentive magnitude of each unit, For the first The original phase of each unit, The phase distortion is caused by the radome. The phase difference introduced by the mutual coupling effect between units. The phase difference is caused by temperature drift. For the first The radiation pattern function of each antenna element; The total array response is: ; in, For the total array response, This represents the total number of elements in the array antenna. For the first The position vector of each unit in space. This is the phase term caused by the path difference.

[0029] In the array response model of the rotation vector, the first i The compensated excitation of each unit is represented as follows: ; in, For the first i The excitation value after compensation for each unit For the unit excitation amplitude, For the original phase, The compensation phase to be optimized; Compensation phase to be optimized satisfy: ; in, This is the maximum amplitude limit for the hardware phase shifter.

[0030] Step S4: Construct a genetic algorithm for global optimization. The genetic algorithm takes minimizing beam pointing error as its core objective. It designs the fitness function by combining pattern fidelity and sidelobe level, and uses the rotation vector method to calculate the trigonometric function terms in the fitness function, replacing the traditional lookup table method or Taylor expansion, so as to reduce the computational complexity in the iteration process of the genetic algorithm. The genetic algorithm includes encoding, population initialization, fitness function definition, genetic operations, and elite retention strategy.

[0031] The encoding uses real numbers, with each individual representing a group. Parameter combinations.

[0032] The initial population size is set to 50-100, and the initial incentive weight parameters are randomly generated; in this embodiment, the initial population size is set to 80.

[0033] The fitness function is calculated as follows: ; in, For fitness value, This represents the actual beam pointing angle corresponding to the current individual in the population. For the desired beam pointing angle, The preset beam pointing error tolerance has a range of values. , For sidelobe level, These are the weighting coefficients.

[0034] Genetic operations include selection, crossover, and mutation; among them, selection uses tournament selection, crossover uses simulated binary crossover, and mutation uses polynomial mutation. The elite retention strategy retains the top 10% of the best individuals in each generation to accelerate convergence and prevent premature convergence.

[0035] The termination condition for a genetic algorithm is: the fitness improvement is less than 1e for 10 consecutive generations. -4 Or the number of iterations reaches 200 generations.

[0036] Step S5: Output the optimal amplitude and phase parameters, decode the optimal individual into the amplitude and phase excitation values ​​of each channel, and write them into the control register of the digital beamformer or phase shifter.

[0037] Step S6: Retest the radiation pattern, evaluate the zero-depth recovery and main lobe offset, verify the compensation effect, and optimize through a closed-loop feedback mechanism.

[0038] The closed-loop feedback mechanism is as follows: if the error exceeds the limit after retesting, the local fine-tuning genetic algorithm is activated to reduce the search space to accelerate convergence and form a closed-loop adaptive compensation.

[0039] The present invention also provides a phased array antenna compensation system, comprising: a control center (including a genetic algorithm compensation engine), an FPGA beamformer, a T / R component array, a data acquisition module, a feedback verification module, a vector network analyzer or a receiving probe array, and a six-axis robotic arm and a microwave probe; The control center is used to run a compensation algorithm that combines genetic algorithms and rotation vectors, and is responsible for parameter calculation and issuing control commands. FPGA beamformers are used to receive optimal amplitude and phase parameters to achieve amplitude and phase adjustment of antenna elements; The T / R module array includes adjustable attenuators and phase shifters, which adjust the excitation amplitude and phase of each unit according to the control signal; The data acquisition module, in conjunction with a vector network analyzer or a receiving probe array, a six-axis robotic arm, and a microwave probe, acquires amplitude and phase response data under both shielded and shielded conditions. The feedback verification module is used to collect test data and feed it back to the control center to realize the verification of compensation effect and closed-loop optimization.

[0040] It is worth noting that all contents not described in detail in this invention are existing technologies and are well known to those skilled in the art.

[0041] Therefore, this invention provides a phased array antenna amplitude and phase compensation method that integrates genetic algorithm and rotation vector method. By integrating genetic algorithm and rotation vector method, combined with electromagnetic disturbance modeling, multi-dimensional space optimization and elite retention strategy, it achieves high-precision, fast amplitude and phase compensation that is adaptable to complex irregular radome scenarios, meets the needs of engineering applications and has significant strategic and economic benefits.

[0042] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A phase and amplitude compensation method for phased array antennas that integrates genetic algorithms and rotation vectors, characterized in that, The specific steps are as follows: Step S1: Collect the amplitude and phase responses of each element of the phased array antenna in both uncovered and covered states to obtain the phase distortion matrix; Step S2: Based on the collected amplitude and phase data, construct an electromagnetic disturbance model and establish a functional relationship between phase distortion and spatial coordinates; Step S3: Construct an array response model based on rotation vectors, representing the excitation of each antenna element as a vector on the complex plane; Step S4: Construct a genetic algorithm for global optimization. The genetic algorithm takes minimizing beam pointing error as its core objective, combines pattern fidelity and sidelobe level to design a fitness function, and uses the rotation vector method to calculate the trigonometric function terms in the fitness function. Step S5: Decode the optimal individual into the amplitude and phase excitation values ​​of each channel, and output the optimal amplitude and phase parameters; Step S6: Retest the radiation pattern, evaluate the zero-depth recovery and main lobe offset, verify the compensation effect, and optimize through a closed-loop feedback mechanism.

2. The phase compensation method for a phased array antenna that integrates genetic algorithm and rotation vector according to claim 1, characterized in that, In step S1, the formula for the phase distortion matrix is ​​as follows: ； in, The phase distortion matrix, For the first i Line 1 j Phase distortion value of column cell, i , j It is an integer.

3. The phase compensation method for a phased array antenna that integrates genetic algorithm and rotation vector according to claim 2, characterized in that, In step S2, the functional relationship between phase distortion and spatial coordinates is as follows: ； in, , , , , These are the fitting coefficients. For unit coordinates, It is a wavenumber vector.

4. The phase compensation method for a phased array antenna that integrates genetic algorithm and rotation vector according to claim 3, characterized in that, In step S3, the rotating vector array response model incorporates the combined effects of radome distortion, element mutual coupling, and temperature drift. Through fitting with calibration data, the... k Each antenna element is in the direction The rotational vector response is: ； in, For the first k Each antenna element is in the direction Rotational vector response under the following conditions For the first The incentive magnitude of each unit, For the first The original phase of each unit, The phase distortion is caused by the radome. The phase difference introduced by the mutual coupling effect between units. The phase difference is caused by temperature drift. For the first The radiation pattern function of each antenna element; The total array response is: ； in, For the total array response, This represents the total number of elements in the array antenna. For the first The position vector of each unit in space. This is the phase term caused by the path difference.

5. The phase compensation method for a phased array antenna that integrates genetic algorithm and rotation vector according to claim 4, characterized in that, In the array response model of the rotation vector, the first i The compensated excitation of each unit is represented as follows: ； in, For the first i The excitation value after compensation for each unit For the unit excitation amplitude, For the original phase, The compensation phase to be optimized; Compensation phase to be optimized satisfy: ； in, This is the maximum amplitude limit for the hardware phase shifter.

6. The phase compensation method for a phased array antenna that integrates genetic algorithm and rotation vector according to claim 5, characterized in that, In step S4, the genetic algorithm includes encoding, population initialization, fitness function definition, genetic operations, and elite retention strategy; The coding uses real numbers, with each individual representing a group. Parameter combinations; The initial population size is set to 50-100, and the initial incentive weight parameters are randomly generated.

7. The phase compensation method for a phased array antenna that integrates genetic algorithm and rotation vector according to claim 6, characterized in that, The fitness function is calculated as follows: ； in, For fitness value, This represents the actual beam pointing angle corresponding to the current individual in the population. For the desired beam pointing angle, The preset beam pointing error tolerance has a range of values. , For sidelobe level, These are the weighting coefficients.

8. The phase and amplitude compensation method for a phased array antenna that integrates genetic algorithm and rotation vector according to claim 7, characterized in that, Genetic operations include selection, crossover, and mutation; among them, selection uses tournament selection, crossover uses simulated binary crossover, and mutation uses polynomial mutation.

9. The phase compensation method for a phased array antenna that integrates genetic algorithm and rotation vector according to claim 8, characterized in that, The elite retention strategy retains the top 10% of the best individuals in each generation to accelerate convergence and prevent premature convergence.

10. The phase compensation method for a phased array antenna that integrates genetic algorithm and rotation vector according to claim 9, characterized in that, In step S6, the closed-loop feedback mechanism is as follows: if the error exceeds the limit after retesting, the local fine-tuning genetic algorithm is started to reduce the search space to accelerate convergence and form a closed-loop adaptive compensation.

Images