Phased array antenna calibration method and calibration system

Abstract

The present application relates to the technical fields of antenna calibration, and specifically includes a phased array antenna calibration method and a calibration system, which comprises: establishing a three-dimensional near-field calibration reference, and approximating an initial error matrix; determining a beam pointing deviation through grouping excitation, and distributing a spatial angle error component; generating a dynamic calibration weight matrix; deploying a distributed calibration node, and performing mutual coupling effect bias measurement and calibration decision-making. The technical problem of static compensation coefficient being difficult to adapt to dynamic working conditions, being unable to effectively separate thermal drift, mechanical deformation and electromagnetic interference, and affecting the stability of the communication link is solved. Through the establishment of a three-dimensional near-field calibration reference, the initial error matrix is formulated in combination with phase difference and amplitude error, the grouping excitation and angle error are cooperatively optimized, the least square method and the deep separable convolutional neural network are combined to effectively separate the environmental disturbance and dynamically compensate, the distributed calibration node is deployed to improve the mutual coupling calibration efficiency between array elements, and the technical effect of ensuring the stability of the communication link is further ensured.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of antenna calibration, in particular to a phased array antenna calibration method and a calibration system. BACKGROUND

[0002] With the extension of the communication frequency band to millimeter wave and terahertz, the number of phased array antenna elements has increased dramatically, and the influence of the mutual coupling effect between array elements and environmental disturbance has been significantly enhanced due to the expansion of the array scale. The conventional static calibration method cannot effectively separate the coupling effects of thermal drift, mechanical deformation and electromagnetic interference, resulting in invalidation of the calibration results with the change of working conditions, which seriously affects the quality of the communication link. In addition, the signal interference between adjacent subarrays has not been effectively suppressed, which further exacerbates the aliasing and distortion of the signal in the multi-base station coordination and complex electromagnetic environment, greatly reduces the beamforming effect, and causes the bit error rate of the communication system to rise sharply, even causing the communication link to be interrupted, which greatly limits the application performance of the phased array antenna in complex scenarios.

[0003] In summary, the prior art has the technical problem that the static compensation coefficient is difficult to adapt to dynamic working conditions, and cannot effectively separate thermal drift, mechanical deformation and electromagnetic interference, affecting the stability of the communication link. SUMMARY

[0004] The present application provides a phased array antenna calibration method and a calibration system to solve the technical problem that the static compensation coefficient in the prior art is difficult to adapt to dynamic working conditions, and cannot effectively separate thermal drift, mechanical deformation and electromagnetic interference, affecting the stability of the communication link.

[0005] In view of the above problems, the technical solution of the present application is as follows:

[0006] In a first aspect, the present application provides a phased array antenna calibration method, wherein the method comprises: establishing a three-dimensional near-field calibration reference based on phased array antenna configuration information, receiving the phase difference and amplitude error of the antenna element, and formulating an initial error matrix; grouping and exciting the phased array elements, determining the beam pointing deviation of each group of phased array elements through the far-field direction vector, and configuring the spatial angle error component; based on the initial error matrix and the spatial angle error component, using the least squares method to determine the compensation coefficient, extracting the environmental disturbance compensation factor through the deep separable convolutional neural network, separating the coupling effects of thermal drift, mechanical deformation and electromagnetic interference, and generating a dynamic calibration weight matrix; according to the dynamic calibration weight matrix, combining the beam pointing accuracy requirement to deploy distributed calibration nodes, and iteratively optimizing the calibration parameters in the multi-base station coordination scenario, while performing mutual coupling effect bias measurement and calibration decision.

[0007] Preferably, a first adaptive calibration index is configured according to the amplitude mutual coupling effect between the elements of the phased array antenna; a second adaptive calibration index is configured according to the phase mutual coupling effect between the elements of the phased array antenna; a test period is determined based on the first adaptive calibration index and the second adaptive calibration index; during the test period, the phase difference of each phased antenna element is determined according to the received signal amplitude of the individual signal transmitted by each phased antenna element, and the mutual coupling effect bias measurement and calibration decision are made: the coupling coefficient between adjacent elements is calculated by port S parameter measurement, the bias of each phased antenna element is determined, and adaptive calibration is performed by fuzzy decision.

[0008] Preferably, the phased antenna array is divided into N independent sub-arrays, each independent sub-array containing P phased antenna elements, and the isolation degree of the independent sub-arrays meets the isolation degree limitation condition; according to the time sequence control logic, the N independent sub-arrays are polled and excited by the high-speed radio frequency switch.

[0009] Preferably, the time interval between the excitation of adjacent sub-arrays corresponding to the N independent sub-arrays meets the time interval limitation condition; at the same time, a preset initial phase weight is applied to each independent sub-array according to the pointing deviation of the independent sub-array beam pointing direction from the target direction.

[0010] Preferably, a first time factor is determined according to the operating frequency of the phased array antenna; a second time factor is determined by using the free space propagation model, which is the time required for the energy of the signal of the first adjacent sub-array to attenuate to a level that meets the set interference threshold value after being transmitted to the second adjacent sub-array; a third time factor corresponding to the switching time of the high-speed radio frequency switch and a fourth time factor corresponding to the signal processing delay are determined, and the time interval limitation condition is set in combination with the first time factor and the second time factor.

[0011] Preferably, the pointing deviation of the independent sub-array beam pointing direction from the target direction is decomposed into an azimuth angle error component and a pitch angle error component; the spatial angle error component is configured by the azimuth angle error component and the pitch angle error component.

[0012] Preferably, according to the number and arrangement of the phased array antenna elements, a spherical scanning frame with a radius of U times the center wavelength of the operating frequency band is built in a microwave darkroom; based on the spherical scanning frame, a vector network analyzer and a double-ridge horn probe are configured to collect full-space field strength data through a preset scanning path; taking the center of the array as the coordinate origin, the physical position parameters of each phased antenna element are associated with the full-space field strength data to obtain a calibration reference database.

[0013] Preferably, the target function is configured as the mean square error of the beam pattern and the target beam pattern, and the constraint condition of the target function is the compensation coefficient adjustment range; based on the target function, a super-determined equation group is determined by using the weighted least square method, and the weight coefficients are dynamically allocated according to the far field attenuation characteristics of the distances of the phased antenna array elements to the array center.

[0014] Preferably, the mapping relationship between the environmental parameters and the amplitude and phase errors in the historical calibration data is taken as a training sample; a separation thermal drift branch, a mechanical deformation branch and an electromagnetic interference branch are set in the output layer of the deep separable convolutional neural network to decouple the coupling effects and draft the dynamic calibration weight matrix.

[0015] In the second aspect of the present application, a phased array antenna calibration system is provided, wherein the system comprises: an initial error matrix drafting module: based on phased array antenna configuration information, a three-dimensional near-field calibration reference is established, the phase difference and amplitude error of the antenna element are received, and an initial error matrix is drafted; a grouping excitation module: the phased antenna array elements are grouped and excited, the beam pointing deviation of each group of phased antenna array elements is determined through the far-field direction vector, and the spatial angle error component is configured; a coupling analysis module: based on the initial error matrix and the spatial angle error component, the compensation coefficient is determined by using the least square method, the environmental disturbance compensation factor is extracted by using the deep separable convolutional neural network, the coupling effects of thermal drift, mechanical deformation and electromagnetic interference are separated, and the dynamic calibration weight matrix is generated; a calibration decision module: according to the dynamic calibration weight matrix, the distributed calibration nodes are deployed in combination with the beam pointing accuracy requirement, the calibration parameters are iteratively optimized in the multi-base station cooperative scene, and at the same time, the mutual coupling effect bias measurement and calibration decision are performed.

[0016] To sum up, one or more technical solutions provided in the present application establish a three-dimensional near-field calibration reference, draft an initial error matrix in combination with the phase difference and amplitude error, cooperatively optimize the grouping excitation and angle error, effectively separate the environmental disturbance and dynamically compensate in combination with the least square method and the deep separable convolutional neural network, deploy the distributed calibration nodes, improve the mutual coupling calibration efficiency between the array elements, and further guarantee the technical effect of affecting the stability of the communication link. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 A flowchart of the phased array antenna calibration method is provided for the present application.

[0018] Figure 2 A structural schematic diagram of the phased array antenna calibration system is provided for the present application.

[0019] Explanation of reference signs: initial error matrix drafting module M100, grouping excitation module M200, coupling analysis module M300, and calibration decision module M400. DETAILED DESCRIPTION

[0020] Embodiment one, the application is described in detail below, as shown in the accompanying drawings, the application provides a phased array antenna calibration method, wherein the method comprises: Figure 1

[0021] S1: based on the phased array antenna configuration information, establish a three-dimensional near-field calibration reference, receive the phase difference, amplitude error of the antenna unit, and draft an initial error matrix; S2: group excitation is performed on the phased array antenna elements, and the beam pointing deviation of each group of phased array antenna elements is determined through the far-field direction vector, and the spatial angle error component is configured.

[0022] Specifically, the three-dimensional near-field calibration reference refers to a three-dimensional space calibration reference system constructed in the near-field region of the phased array antenna array surface, which is used to accurately measure and characterize the radiation characteristics of the antenna elements; the initial error matrix is drafted based on the collected phase difference and amplitude error of the antenna elements, and the error information is integrated into a matrix form through spherical wave expansion, so as to facilitate subsequent processing; group excitation refers to dividing the phased array antenna elements into multiple groups according to certain rules, and sequentially applying excitation signals to each group to make it produce a radiation field; the far-field direction vector is a unit vector describing the direction of antenna radiation, which is used to determine the beam pointing in the far-field measurement; the beam pointing deviation refers to the deviation angle between the actual beam pointing and the ideal beam pointing; the spatial angle error component is the error component in the direction of azimuth angle and elevation angle, etc. decomposed from the beam pointing deviation according to the spatial coordinate system, so as to more carefully analyze and calibrate.

[0023] Execution steps: in the process of establishing a three-dimensional near-field calibration reference and drafting an initial error matrix, according to the phased array antenna configuration information such as the number of elements, arrangement mode, etc., a spherical scanning frame is built in a microwave darkroom, a vector network analyzer and a double-ridge horn probe are configured, full-space field strength data is collected, the center of the array surface is taken as the coordinate origin, the physical position parameters of each phased array antenna element are associated with the collected field strength data, a calibration reference database is constructed, and reliable basis is provided for subsequent error analysis; through the spherical wave expansion method, the phase difference and amplitude error of the antenna elements are integrated into the initial error matrix, the element value represents the amplitude and phase error of the corresponding element, and the initial error matrix can comprehensively reflect the initial error state of the antenna elements, providing basic data for subsequent calibration compensation.

[0024] ​The phased antenna array elements are grouped and excited, and further, the phased antenna array surface is divided into N independent sub-arrays, each sub-array containing P phased antenna array elements, and the isolation degree between independent sub-arrays needs to meet the isolation degree limit condition (such as isolation degree ≥ 30 dB) to reduce the mutual interference between sub-arrays; according to the time sequence control logic, the N independent sub-arrays are polled and excited in turn by using a high-speed radio frequency switch, and the excitation time of each sub-array is T seconds, for example, if the phased array antenna has 100 elements and is divided into 4 sub-arrays, each sub-array contains 25 elements, and at a working frequency of 10 GHz, each sub-array is excited for 1 second through the switching of the high-speed radio frequency switch, and the excitation of all sub-arrays is completed in turn.

[0025] During the excitation process, the far-field radiation pattern of each sub-array is obtained by a far-field measurement device to determine the beam pointing deviation of each group of phased antenna array elements, and the deviation is decomposed into azimuth angle error components and elevation angle error components through coordinate transformation, and then the spatial angle error components are configured to accurately position the beam pointing of each group of elements, provide key spatial error information for subsequent dynamic calibration weight matrix generation, so that the calibration process can compensate for specific spatial angle errors, effectively improving the calibration accuracy and antenna performance.

[0026] S3: Based on the initial error matrix and the spatial angle error components, a compensation coefficient is determined using the least squares method, an environmental disturbance compensation factor is extracted by a deep separable convolutional neural network to separate the coupling effects of thermal drift, mechanical deformation and electromagnetic interference, and a dynamic calibration weight matrix is generated; S4: According to the dynamic calibration weight matrix, combine the beam pointing accuracy requirement to deploy distributed calibration nodes, and in the multi-base station cooperative scene, iteratively optimize the calibration parameters, and at the same time, perform mutual coupling effect bias measurement and calibration decision.

[0027] Specifically, the initial error matrix is a matrix integrated by mathematical methods such as spherical wave expansion based on the collected phase difference and amplitude error of the antenna elements, used to represent the initial error state of the antenna elements; the spatial angle error component is the error component in the direction of azimuth angle and elevation angle decomposed from the beam pointing deviation according to the spatial coordinate system, so as to analyze and calibrate more carefully; the least squares method is used to find the best fitting parameters by minimizing the sum of squares of errors; the deep separable convolutional neural network refers to the combination of deep convolutional layers and point convolutional layers, which can efficiently extract features from multi-dimensional data.

[0028] The dynamic calibration weight matrix is a matrix calibrated according to an environmental disturbance compensation factor and an initial error, and is used to adjust the amplitude and phase states of the antenna elements in real time to adapt to the dynamically changing environment; the distributed calibration node refers to a plurality of calibration device nodes deployed in the phased array antenna system, which can collect the working state parameters of the elements in the region in real time; the multi-base station cooperation refers to the information interaction and cooperative work among a plurality of base stations to achieve more extensive calibration and optimization; the mutual coupling effect refers to the mutual influence between the antenna elements due to electromagnetic coupling, which can change the amplitude and phase characteristics of the elements and affect the overall performance of the antenna; the bias measurement is a process of quantitatively measuring the amplitude and phase bias of the elements caused by the mutual coupling effect; the calibration decision is a judgment process of deciding whether calibration is needed according to the bias measurement results and calibration requirements.

[0029] Error analysis is performed according to the initial error matrix and the spatial angle error component. The error analysis involves information such as phase difference, amplitude error and beam pointing deviation of the antenna elements. The phased array antenna has N elements, and the spatial angle error component of the initial error matrix is an N x 2 matrix (corresponding to azimuth and elevation error). Based on the error data, the least squares method is used to determine the compensation coefficient. The objective function is defined as the mean square error of the actual beam pattern and the ideal pattern, and the adjustment range of the compensation coefficient is constrained (amplitude 0.5-2 times, phase -180-180°). The weighted least squares method is used to solve the overdetermined equation set. The weight coefficient is dynamically allocated according to the far and near field attenuation characteristics of the element distance from the center, and then the Levenberg-Marquardt iterative optimization compensation coefficient is used. The calculation is stopped when the iteration residual is ≤1e-6, and the initial compensation table is generated. The compensation coefficient can be accurately determined to provide key parameters for subsequent calibration, effectively improve the calibration accuracy, reduce the beam pointing error after compensation to an acceptable range, and significantly improve the radiation performance of the antenna.

[0030] At the same time, the environmental disturbance compensation factor is extracted by a deep separable convolutional neural network. Further, a neural network containing 3 deep convolutional layers and 2 point convolutional layers is constructed, and multi-dimensional time series data of environmental parameters (such as temperature range -40℃-85℃, vibration frequency 0Hz-50Hz, electromagnetic interference intensity -100dBm-0dBm) are input. The mapping relationship between the environmental parameters and the amplitude and phase errors in the historical calibration data is used as a training sample, and the Adam optimizer is used for model training. The loss function is the root mean square error.

[0031] Three branches are arranged at the network output layer to output compensation factors corresponding to thermal drift, mechanical deformation and electromagnetic interference respectively, to realize decoupling of coupling effects, to collect environmental data in real time and input the model, to update compensation factors synchronously, to improve compensation accuracy, to effectively improve the stability and reliability of the antenna system in a dynamic environment, and to make the calibration result better adapt to changes in the actual working environment.

[0032] According to the dynamic calibration weight matrix, distributed calibration nodes are deployed and multi-base station collaborative optimization is performed. Further, according to the beam pointing accuracy index, multiple calibration nodes are configured, the nodes are synchronized through optical fiber (time delay ≤10 ns) to ensure the efficiency and real-time performance of data transmission, each node is integrated with an amplitude-phase detector and a processor to collect working state parameters of the array elements in the region, such as phase and amplitude information, a master-slave calibration architecture is set, the master node collects data of the slave nodes and generates global calibration instructions, the slave nodes execute local parameter adjustment, and fine calibration of the antenna array elements is realized.

[0033] In the multi-base station collaborative scenario, the synchronization update of the cross-base station calibration parameters is realized through timestamp alignment to ensure the consistency and collaboration of the antenna calibration between multiple base stations, the mutual coupling effect bias measurement and calibration decision are simultaneously performed, the array element mutual coupling model is established, the coupling coefficients between adjacent array elements are measured through port S parameters to generate a mutual coupling matrix, the bias of each array element is calculated according to the mutual coupling matrix, and the targeted calibration is triggered when the bias exceeds the limit; the calibration priority is generated by comprehensively considering the beam pointing error, the mutual coupling bias and the environmental disturbance level through fuzzy decision, and the calibration with the highest priority is executed to ensure the availability of the antenna system. In the above steps, the calibration efficiency and accuracy are improved through reasonable deployment of calibration nodes and distributed collaborative optimization, the mutual coupling effect between array elements and the multi-base station collaboration problem are effectively solved, and the performance and stability of the phased array antenna system in a complex scenario are further improved, so that it can better meet the requirements of modern communication systems for high precision and high reliability.

[0034] Further, the mutual coupling effect bias measurement and calibration decision are performed, and the method of the present application includes the following steps:

[0035] According to the amplitude mutual coupling effect between the phased array antenna elements, a first adaptive calibration index is configured; according to the phase mutual coupling effect between the phased array antenna elements, a second adaptive calibration index is configured; based on the first adaptive calibration index and the second adaptive calibration index, a test period is determined; in the test period, the phase difference of each phased antenna element is determined according to the received signal amplitude of the individual signal transmission of each phased antenna element, and the mutual coupling effect bias measurement and calibration decision are performed: the coupling coefficients between adjacent array elements are calculated through port S parameter measurement, the bias of each phased antenna element is determined, and adaptive calibration is performed through fuzzy decision.

[0036] Specifically, the adaptive calibration index is based on the mutual coupling effect characteristics between the elements of the phased array antenna, dynamically measures the deviation degree of the antenna phase amplitude state and the ideal state, specifically, the first adaptive calibration index reflects the influence of amplitude mutual coupling effect on the amplitude of each element, and the second adaptive calibration index reflects the influence of phase mutual coupling effect on the phase of each element; by real-time monitoring and analyzing the mutual coupling effect between the elements, the adaptive calibration index can dynamically reflect the error state of the antenna system, and provide quantitative basis for subsequent calibration decision.

[0037] The port S parameter, i.e. the scattering parameter, is a parameter system for describing the characteristics of radio frequency and microwave networks. In the phased array antenna, the port S parameter is used to quantify the signal transmission and reflection characteristics between the antenna ports. By measuring the S parameter, the coupling coefficient between adjacent elements can be calculated to reflect the degree of electromagnetic coupling between the elements. The coupling coefficient is a key indicator for measuring the strength of the mutual coupling effect between the elements. The larger the value, the stronger the mutual coupling between the elements. For example, when the coupling coefficient between two elements is large, the signal of one element will significantly affect the amplitude and phase characteristics of the other element, resulting in an increase in the mutual coupling effect bias.

[0038] Fuzzy decision-making is a decision-making method based on fuzzy logic, which is used to handle uncertain information. In the calibration of the phased array antenna, the fuzzy decision-making algorithm considers multiple factors such as beam pointing error, mutual coupling bias, environmental disturbance level, etc. to generate a calibration priority. These factors are often difficult to describe with precise mathematical models. Fuzzy decision-making converts these factors into fuzzy sets through fuzzy rules and membership functions, thereby achieving comprehensive evaluation of the calibration item priority and effectively handling multi-factor and uncertain decision-making problems, ensuring reasonable calibration decisions in complex environments.

[0039] Execution steps: when configuring the adaptive calibration index, two independent index systems are constructed based on the amplitude mutual coupling effect and the phase mutual coupling effect of the elements of the phased array antenna. Further, for the amplitude mutual coupling effect, the amplitude bias of each element is determined by measuring the coupling coefficient between the elements, and then the first adaptive calibration index is constructed. For the phase mutual coupling effect, the second adaptive calibration index is constructed based on the coupling coefficient and phase difference measurement. For example, in a phased array antenna with N elements, the coupling coefficients between each element and other elements form an N x N mutual coupling matrix. By analyzing the eigenvalues and eigenvectors of this matrix, the strength and influence range of the mutual coupling effect can be quantified. In a typical phased array antenna, if the coupling coefficient between two adjacent elements is greater than 0.1, it is considered that there is significant mutual coupling effect between the two adjacent elements and calibration is needed. Based on these two calibration indexes, combined with historical calibration data and the working state of the antenna, the optimal test period is predicted using machine learning algorithms. The optimal test period of the phased array antenna is once every hour, ensuring timely capture of changes in the mutual coupling effect and thereby achieving adaptive calibration.

[0040] In the test cycle, by controlling each phased antenna array element to emit a signal with a known amplitude and phase, other array elements receive the signal and measure its amplitude and phase, according to the amplitude change of the received signal, using mutual coupling model and least square method, etc. algorithm, determine the phase difference of the array element, such as the array element emits a signal with amplitude V0 and phase θ0, the amplitude of the signal received by the adjacent array element is V1, and the phase is θ1, then the coupling coefficient and the phase difference between the array element and other array elements can be determined by the ratio of V1 to V0 and the difference of θ1 to θ0, combined with the mutual coupling model.

[0041] By port S parameter measurement, the coupling coefficient matrix between adjacent array elements is determined, and the bias of each phased antenna array element is determined. The element S ij in the coupling coefficient matrix represents the coupling effect of the i-th array element on the j-th array element. By analyzing the amplitude and phase of S ij , the bias model can be constructed to quantify the bias of each array element. Further, if the bias of a certain array element exceeds a predetermined threshold (such as 5%), a targeted calibration process is triggered, and a fuzzy decision algorithm is used to integrate beam pointing error, mutual coupling bias, environmental disturbance level and other factors to generate a calibration priority. In the above steps, by real-time monitoring and analyzing the mutual coupling effect, the calibration strategy is dynamically adjusted to ensure that the antenna system always maintains high precision and high stability in complex environments.

[0042] Further, the phased antenna array elements are grouped and excited, and the method of the present application comprises:

[0043] The phased antenna array surface is divided into N independent sub-arrays, each independent sub-array containing P phased antenna array elements, and the isolation degree of the independent sub-arrays meets the isolation degree limit condition; according to the time sequence control logic, the N independent sub-arrays are polled and excited by high-speed radio frequency switches.

[0044] Specifically, the independent sub-array refers to the phased antenna array surface divided according to certain rules, and the sub-arrays have high isolation between each other. Each independent sub-array contains a certain number of phased antenna elements. Such division helps to reduce the mutual interference between sub-arrays and facilitate separate excitation and calibration. The isolation degree limitation condition refers to the minimum isolation degree requirement between sub-arrays to ensure that each independent sub-array does not interfere with each other when working. It is usually expressed in decibels (dB) to ensure that the signal leakage between sub-arrays is within an acceptable range. The timing control logic refers to the rules for controlling each independent sub-array according to the pre-set time sequence and logical relationship, which is used to coordinate the action of high-speed radio frequency switches, so that each sub-array can be excited in turn to avoid signal conflict and mutual interference. The polling excitation refers to exciting each independent sub-array in turn according to certain timing control logic, so that each sub-array transmits signals within a specified time to measure and analyze its performance respectively.

[0045] The execution steps are as follows: the phased antenna array surface is divided into N independent sub-arrays and polling excitation is performed. Further, according to the array layout and performance requirements of the phased array antenna, the number N of sub-arrays and the number P of elements contained in each sub-array are determined. For example, the phased array antenna array surface has 100 elements, which is divided into 4 independent sub-arrays, and each sub-array contains 25 elements. In the division process, the arrangement and wiring of the elements are optimized to ensure that the isolation degree between independent sub-arrays meets the isolation degree limitation condition. For example, by increasing the spacing between sub-arrays, using electromagnetic shielding measures, etc., the isolation degree between sub-arrays is increased to more than 35 dB, effectively reducing the signal interference between sub-arrays.

[0046] According to the timing control logic, the N independent sub-arrays are polled and excited using high-speed radio frequency switches. If the timing control logic requires each sub-array to be excited for 1 ms and the excitation interval between adjacent sub-arrays is 0.5 ms, the switching time of the high-speed radio frequency switch needs to be less than 1 μs to meet the timing requirements. Further, by controlling the switching of the high-speed radio frequency switch, excitation signals are applied to each independent sub-array in turn to make it emit radio frequency signals. For example, after the first sub-array is excited for 1 ms, the high-speed radio frequency switch is quickly switched to the second sub-array, and after an interval of 0.5 ms, the second sub-array is excited. In this way, the polling excitation of all sub-arrays is completed.

[0047] Through this polling excitation method, the radiation characteristic data of each sub-array, including the directional diagram, phase difference, amplitude error, etc., can be obtained, and the performance and error of each sub-array can be analyzed. The above steps provide a grouping data basis for subsequent error analysis and calibration parameter optimization, so that the calibration process can be optimized for the specific situation of each sub-array, improving the calibration efficiency and accuracy. At the same time, it also provides a convenient condition for the deployment of distributed calibration nodes and the cooperative optimization of multiple base stations.

[0048] Further, the method of the present application further comprises:

[0049] The time interval between the excitations of the adjacent sub-arrays corresponding to the N independent sub-arrays satisfies a time interval limitation condition; and a preset initial phase weight is applied to each independent sub-array according to a pointing deviation of the beam pointing direction of the independent sub-array from the target direction.

[0050] Specifically, the time interval between the excitations of the adjacent sub-arrays is the time interval between two adjacent sub-arrays when the independent sub-arrays are polled and excited, which needs to satisfy the time interval limitation condition, i.e. the minimum time interval requirement determined according to the working frequency of the phased array antenna, the signal propagation characteristics, and the switching speed of the device, etc., to ensure that the signal emitted by the previous sub-array has been switched by the high-speed radio frequency switch and the next sub-array can work normally before the signal emitted by the previous sub-array interferes with the next sub-array; the preset initial phase weight is a phase compensation value set for each independent sub-array in advance according to the pointing deviation of the beam pointing direction of the independent sub-array from the target direction, which is used to adjust the initial phase of the beam when the sub-array is excited, so that the beam pointing direction is closer to the target direction, the beam pointing deviation is reduced, and the pattern performance of the antenna is improved.

[0051] Execution steps: To ensure that the time interval between the excitations of the adjacent sub-arrays satisfies the time interval limitation condition, further, a first time factor is determined according to the working frequency of the phased array antenna, such as 0.1 μs corresponding to the signal period when the working frequency is 10 GHz; the distance corresponding to the signal propagation from emission to energy attenuation to the signal collection of the adjacent sub-arrays meeting the set interference threshold value is calculated by using the free space propagation model, and then a second time factor is determined, such as requiring the interference signal to be attenuated to below -60 dBm, so that the signal propagation distance of 2 m at this working frequency satisfies the attenuation requirement, and the corresponding second time factor is the time required for the signal to propagate 2 m in the air; a third time factor corresponding to the switching time of the high-speed radio frequency switch, and a fourth time factor corresponding to the signal processing delay are determined; the time interval between the excitations of the adjacent sub-arrays is set by comprehensively considering the above time factors, to ensure that the next sub-array is excited after the high-speed radio frequency switch is switched and the signal interference is attenuated to an acceptable level, to avoid the mutual interference of the signals between the sub-arrays and to ensure the accuracy of the measurement data.

[0052] Meanwhile, according to the deviation degree of the pointing direction of each independent subarray from the target direction, a preset initial phase weight is applied to each independent subarray, and according to the relationship between the beam pointing direction and the phase weight, the preset initial phase weight that needs to be applied to the subarray is determined. Specifically, if it is found through measurement that the beam pointing direction of a certain independent subarray deviates from the target direction by 2° in the azimuth angle and 1.5° in the elevation angle, the corresponding phase adjustment amount is π / 90 radian (assuming that the wavelength is λ and the element spacing is λ / 2) for each 1° deviation in the azimuth angle, and the same applies to the elevation angle. Therefore, the independent subarray needs to adjust the phase by 2xπ / 90 radian in the azimuth angle and 1.5xπ / 90 radian in the elevation angle. By adding the corresponding phase compensation in the excitation signal, the beam pointing direction of the independent subarray is closer to the target direction. In the above steps, the preset initial phase weight reduces the beam pointing deviation, improves the directivity gain and anti-interference capability of the antenna, provides more accurate initial conditions for the subsequent dynamic calibration weight matrix generation and distributed calibration node optimization, and ensures the efficiency of the calibration process and the accuracy of the calibration result.

[0053] Further, the time interval between the excitations of the adjacent subarrays corresponding to the N independent subarrays satisfies a time interval limitation condition, and the method comprises the following steps:

[0054] A first time factor is determined according to the operating frequency of the phased array antenna, a second time factor is determined by using a free space propagation model to determine the time required for the energy of the signal of the first adjacent subarray to attenuate to a set interference threshold value from being emitted, a third time factor corresponding to the switching time of a high-speed radio frequency switch and a fourth time factor corresponding to the signal processing delay are determined, and the time interval limitation condition is set in combination with the first time factor and the second time factor.

[0055] Specifically, the first time factor refers to a signal period or a related time parameter calculated based on the operating frequency of the phased array antenna, which is used to evaluate the basic time characteristics of the signal; the second time factor refers to the time required for the energy of the signal of the adjacent subarray to attenuate to the set interference threshold value from being emitted, which is determined according to the free space propagation model and is used to evaluate the attenuation time in the signal propagation process; the third time factor refers to the time required for the high-speed radio frequency switch to complete switching, which is a key parameter of the device performance; the fourth time factor refers to the delay time generated in the signal processing process, including the total time of the signal detection, amplification, filtering and other processing links; and the time interval limitation condition refers to setting the minimum time interval that must be satisfied between the excitations of the adjacent subarrays to ensure the accuracy of the signal acquisition and avoid mutual interference between the subarrays.

[0056] Execution step: according to the formula f=1 / T, the signal period is determined, which determines the size of the first time factor, for example, at a working frequency of 10GHz, the signal period is 0.1μs; according to the free space propagation model, combined with the interference threshold value, the signal propagation distance is calculated, and then converted into time, for example, at a frequency of 10GHz, the electromagnetic wave propagation speed C is 3×10 8 m / s, the signal propagation distance of 2m corresponds to a time of about 2 / (3×10 8 m / s), that is, 6.67ns; if the switching time of the high-speed radio frequency switch is 0.1μs; the signal processing delay is 0.2μs; the first time factor (0.1μs), the second time factor (6.67ns), the third time factor (0.1μs) and the fourth time factor (0.2μs) are added, and the total time interval is 0.4667μs, which is used as the time interval limitation condition of adjacent subarray excitation.

[0057] In the actual application process, it also includes checking whether the calculated time interval can effectively avoid signal interference between subarrays, which can be verified by experiment or simulation; verifying whether the switching time of the high-speed radio frequency switch and the signal processing delay meet the actual device performance; ensuring the accuracy and reliability of signal acquisition under the time interval limitation condition. Preferably, through the above steps, the reasonable setting of the interval between adjacent subarray excitations is ensured, signal interference is avoided, and the accuracy of calibration is improved.

[0058] Further, the beam pointing deviation of each group of phased antenna elements is determined, and the spatial angle error component is configured. The method of the application comprises:

[0059] According to the pointing deviation value of the independent subarray beam pointing and the target direction, the azimuth angle error component and the elevation angle error component are decomposed; the spatial angle error component is configured through the azimuth angle error component and the elevation angle error component.

[0060] Specifically, the pointing deviation value refers to the deviation of the actual beam pointing from the target pointing, which is expressed in angle; the azimuth angle error component refers to the included angle between the horizontal plane beam pointing and the target direction; the elevation angle error component refers to the included angle between the vertical plane beam pointing and the target direction; the spatial angle error component refers to the integration of the azimuth angle and the elevation angle error, which is a comprehensive index reflecting the spatial pointing deviation.

[0061] Execution step: by measuring the beam pointing of the independent subarray, the deviation from the target direction is determined, such as the target direction is azimuth 0°, elevation 45°, if the actual beam pointing is azimuth 2°, elevation 46.5°, the azimuth error is 2°, and the elevation error is 1.5°; configure the spatial angle error component, further, according to the formula, the square of the azimuth error plus the square of the elevation error, and then the square root of the spatial angle error is obtained, the spatial pointing deviation is quantified, which provides accurate data support for subsequent calibration.

[0062] In the above steps, by decomposing and configuring the error component, the key spatial error parameters are provided for the generation of the dynamic calibration weight matrix, which can adjust the phase weight of the corresponding independent subarray to reduce the deviation, thereby improving the antenna pattern gain and anti-interference ability, which helps to compensate for the specific spatial angle error in the subsequent calibration process, effectively improves the calibration accuracy and antenna performance, and ensures the quality of the communication link.

[0063] Further, based on the phased array antenna configuration information, a three-dimensional near-field calibration reference is established, and the method of the application further comprises:

[0064] According to the number and arrangement of the phased array antenna elements, a spherical scanning frame with a radius of U times the center wavelength of the working frequency band is built in the microwave anechoic chamber; based on the spherical scanning frame, a vector network analyzer and a double-ridge horn probe are configured, and full-space field strength data is collected through a preset scanning path; taking the center of the array surface as the coordinate origin, the physical position parameters of each phased antenna element are associated with the full-space field strength data to obtain a calibration reference database.

[0065] Specifically, the number and arrangement of the phased array antenna elements refer to the fact that the phased array antenna is composed of multiple antenna units (elements), and the number and arrangement (such as rectangular lattice, triangular lattice, etc.) determine the overall performance and calibration complexity of the antenna; the spherical scanning frame refers to a spherical structure built in the microwave anechoic chamber, which is used to support and move the measurement equipment, and its radius is determined according to the center wavelength of the working frequency band to ensure the accuracy of the measurement and the requirement of covering the full space; the vector network analyzer is an instrument for measuring the amplitude and phase characteristics of radio frequency and microwave components, which can accurately measure the amplitude and phase information of the antenna elements to provide key data support for calibration.

[0066] The double-ridge horn probe is a broadband and low-loss antenna probe used for receiving or transmitting signals on a spherical scanning frame to collect full-space field strength data of the phased array antenna.

[0067] The execution steps are as follows: according to the number and arrangement of the elements of the phased array antenna, a spherical scanning frame with a radius of M times the center wavelength of the working frequency band is built in the microwave darkroom. For example, if the center wavelength of the working frequency band of the phased array antenna is λ and M is 2, the radius of the spherical scanning frame is set to 2λ. Preferably, the calibration is performed in a darkroom environment to avoid external electromagnetic interference and ensure the accuracy of the measurement data. The vector network analyzer and the double-ridge horn probe are installed on the spherical scanning frame, and full-space field strength data is collected according to the preset scanning path. The vector network analyzer can accurately measure the amplitude and phase of the signal, while the double-ridge horn probe is responsible for receiving or transmitting the signal. The two work together to ensure that the collected field strength data is comprehensive and accurate.

[0068] The physical position parameters of each element of the phased array antenna are recorded with the center of the array surface as the coordinate origin, including the coordinate position, direction, and other information of the element. The collected full-space field strength data is associated with these physical position parameters to form a calibration reference database. The calibration reference database can provide radiation characteristic data of the phased array antenna at different positions and directions, providing a basis for subsequent error analysis and calibration. For example, the calibration reference database can be used to accurately analyze the radiation field strength of each element in different directions, and then determine the phase difference and amplitude error data in the initial error matrix, providing key support for the subsequent calibration steps.

[0069] In the above steps, by accurately building the spherical scanning frame, configuring the measurement equipment, and collecting the full-space field strength data, the calibration reference database is constructed, providing reliable data support for subsequent error analysis, dynamic calibration weight matrix generation, and optimization of distributed calibration nodes, ensuring the scientificity of the calibration process and the accuracy of the calibration results.

[0070] Further, the least squares method is used to determine the compensation coefficient, and the environmental disturbance compensation factor is extracted through the deep separable convolutional neural network. The method of the application further includes:

[0071] The target function is configured as the mean square error of the beam pattern and the target beam pattern, and the constraint condition of the target function is the adjustment range of the compensation coefficient; based on the target function, a super-determined equation group is determined by using a weighted least square method, and the weight coefficient is dynamically allocated according to the far and near field attenuation characteristics of the distance of each phased antenna array element from the array center.

[0072] Specifically, the target function is an index for measuring the difference between the beam pattern and the target beam pattern, which is represented by the mean square error here, and the target beam pattern is an ideal pattern, while the beam pattern is an actually measured or simulated pattern. The purpose of the target function is to optimize the calibration effect by minimizing the difference between the two; the mean square error is a statistical quantity for measuring the difference between the predicted value and the true value, and the calculation method is to square the difference between each predicted value and the true value and then take the average, which is used to measure the difference between the actual beam pattern and the target beam pattern.

[0073] The adjustment range of the compensation coefficient is the range of the compensation amplitude and phase adjustment allowed in the calibration process. The compensation coefficient is used to adjust the amplitude and phase state of the antenna array element to reduce the difference between the beam pattern and the target pattern. The adjustment range is usually expressed in amplitude multiples and phase angles, for example, the amplitude is between 0.5 and 2 times, and the phase is between -180° and 180°; the weighted least square method is an optimization algorithm used to solve super-determined equation groups in the presence of measurement errors or data uncertainties, by assigning a weight to each equation, the data points with smaller errors have a greater impact on the result, thereby improving the accuracy of the solution.

[0074] The super-determined equation group refers to an equation group in which the number of equations is greater than the number of unknowns. In this case, since the number of measurement data points is usually greater than the number of compensation coefficients to be determined, the least square method or other methods need to be used to solve it; the far and near field attenuation characteristics describe the law of the radiation field strength of the antenna array element changing with distance. The field strength decays faster in the near field region, while the field strength decays slower in the far field region. The weight coefficient is dynamically allocated according to the far and near distance of the array element from the array center according to its field of view attenuation characteristics, to reflect the influence of different array elements on the overall beam pattern.

[0075] The execution steps are as follows: the gain value of the actual beam pattern in the i-th direction and the gain value of the target beam pattern in the i-th direction are configured to the target function as the mean square error of the beam pattern and the target beam pattern, and the constraint condition is the adjustment range of the compensation coefficient, i.e., the amplitude is between 0.5 and 2 times, and the phase is between -180° and 180°.

[0076] Based on the configured objective function, the over-determined equation set is solved by using the weighted least squares method, and the weight coefficients are dynamically allocated according to the far and near field attenuation characteristics of the array element distance from the array center. Specifically, the standard deviation of the field strength measurement error of the jth array element is related to the distance of the array element from the array center. The field strength attenuation of the near-field array element is faster, and the measurement error is larger, so the weight coefficient is smaller. The field strength attenuation of the far-field array element is slower, and the measurement error is smaller, so the weight coefficient is larger. For example, in a phased array antenna containing 100 array elements, the weight coefficient of the array element near the center is 0.1, and the weight coefficient of the array element at the edge may be 1.0. By solving the over-determined equation set by the weighted least squares method, the optimal compensation coefficient can be obtained to minimize the objective function, significantly improving the accuracy of the beam pattern.

[0077] In the above steps, the compensation coefficient is accurately determined by a mathematical optimization method to minimize the difference between the beam pattern and the target pattern; by reasonably configuring the objective function and using the weighted least squares method, the contributions and measurement errors of different array elements are effectively considered, the accuracy and reliability of the calibration are improved, accurate compensation coefficients are provided for subsequent dynamic calibration weight matrix generation and environmental disturbance compensation, and the phased array antenna can maintain high performance under various working conditions.

[0078] Further, the environmental disturbance compensation factor is extracted by the depth separable convolutional neural network to separate the coupling effects of thermal drift, mechanical deformation and electromagnetic interference, and the dynamic calibration weight matrix is generated. The method of the application comprises:

[0079] The mapping relationship between the environmental parameters and the amplitude and phase errors in the historical calibration data is taken as the training sample; the output layer of the depth separable convolutional neural network is set to have a thermal drift branch, a mechanical deformation branch and an electromagnetic interference branch to decouple the coupling effects and draft the dynamic calibration weight matrix.

[0080] Specifically, the historical calibration data refers to the data accumulated by the phased array antenna in the past calibration process, which records the amplitude and phase errors of the antenna under different environmental conditions and the corresponding calibration parameters. These data are crucial for training the machine learning model, as they provide the mapping relationship between the environmental parameters and the amplitude and phase errors; the environmental parameters refer to external conditions that affect the performance of the phased array antenna, such as temperature, humidity, air pressure, vibration and electromagnetic interference, etc. Changes in these parameters will cause the amplitude and phase characteristics of the antenna to change, thereby affecting the beam pointing accuracy and system stability.

[0081] Amplitude and phase errors refer to the deviations between the actual amplitude and phase of the antenna elements and the ideal values. Amplitude errors are usually expressed in decibels (dB). These errors affect the beamforming and pattern of the antenna and need to be reduced through calibration. Deeply separable convolutional neural networks are a lightweight convolutional neural network structure that decomposes standard convolution into deep convolution and point-wise convolution, reducing computational complexity and the number of model parameters while maintaining model performance. This network structure is suitable for resource-constrained environments such as embedded systems.

[0082] Separating the thermal drift branch, mechanical deformation branch, and electromagnetic interference branch refers to setting three independent branches in the output layer of the deeply separable convolutional neural network, corresponding to the thermal drift, mechanical deformation, and electromagnetic interference three coupling effects. Each branch outputs a corresponding compensation factor for decoupling the effects of these three effects. Decoupling of coupling effects refers to separating the three interwoven effects of thermal drift, mechanical deformation, and electromagnetic interference to determine their contributions to amplitude and phase errors. The decoupled compensation factors can be more accurately used for calibration, improving the effectiveness of calibration. The dynamic calibration weight matrix is a matrix generated dynamically according to real-time environmental conditions and calibration requirements, used to adjust the amplitude and phase of the antenna elements to compensate for the effects of environmental disturbances and mutual coupling. The dynamic calibration weight matrix plays a key role in the calibration process of phased array antennas, ensuring high-performance operation of the antenna in different environments.

[0083] Execution steps: Collect environmental parameters and corresponding amplitude and phase errors from historical calibration data, record amplitude and phase error data at different temperatures, vibration frequencies, and electromagnetic interference intensities, and establish a training sample set. Construct a deeply separable convolutional neural network, with the input layer receiving an environmental parameter vector, the middle layers extracting features through deep convolution and point-wise convolution, and the output layer setting three branches corresponding to the compensation factors for thermal drift, mechanical deformation, and electromagnetic interference. For example, the network structure can be: input layer (3 neurons) → deep convolution layer (32 filters, filter size 1x3) → point-wise convolution layer (64 filters) → deep convolution layer (64 filters, filter size 1x3) → point-wise convolution layer (128 filters) → output layer (3 branches, each branch corresponding to a compensation factor). During training, use the root mean square error (RMSE) as the loss function and use the Adam optimizer for optimization. Training data can be divided into 80% for training, 10% for validation, and 10% for testing to ensure the generalization ability of the model.

[0084] By inputting real-time environmental parameters into the trained deep separable convolutional neural network, the network outputs compensation factors of thermal drift, mechanical deformation and electromagnetic interference in three branches of the output layer respectively; according to the decoupled compensation factors, combined with the initial error matrix and the spatial angle error component, a dynamic calibration weight matrix is generated. The elements of the dynamic calibration weight matrix represent the amplitude and phase adjustment amount of each array element, which is used to calibrate the amplitude and phase state of the antenna. In the above steps, the machine learning model is trained using historical data to realize real-time compensation for environmental disturbances. Through the branch structure of the deep separable convolutional neural network, the effects of thermal drift, mechanical deformation and electromagnetic interference can be effectively decoupled to generate an accurate dynamic calibration weight matrix, which provides key support for the high-performance operation of phased array antennas in complex dynamic environments, and significantly improves the adaptability and reliability of the antenna system.

[0085] In summary, the beneficial effects of the embodiments of the present application are:

[0086] Based on the phased array antenna configuration information, a three-dimensional near-field calibration reference is established, the phase difference and amplitude error of the receiving antenna element are received, and an initial error matrix is drafted. The phased array elements are grouped and excited, the beam pointing deviation of each group of phased array elements is determined through the far-field direction vector, and the spatial angle error component is configured. Based on the initial error matrix and the spatial angle error component, the least squares method is used to determine the compensation coefficient, the environmental disturbance compensation factor is extracted through the deep separable convolutional neural network, the coupling effects of thermal drift, mechanical deformation and electromagnetic interference are separated, and a dynamic calibration weight matrix is generated. According to the dynamic calibration weight matrix, combined with the beam pointing accuracy requirement, a distributed calibration node is deployed, the calibration parameters are iteratively optimized in the multi-base station cooperative scenario, and at the same time, the mutual coupling effect bias measurement and calibration decision are performed. The present application provides a phased array antenna calibration method and a calibration system. By establishing a three-dimensional near-field calibration reference, combining the phase difference and amplitude error to draft an initial error matrix, and performing group excitation and angle error cooperative optimization, the environmental disturbance is effectively separated and dynamically compensated by combining the least squares method and the deep separable convolutional neural network, a distributed calibration node is deployed, the mutual coupling calibration efficiency between array elements is improved, and the technical effect of ensuring the stability of the communication link is further ensured.

[0087] Embodiment two, based on the same inventive concept as the phased array antenna calibration method in the foregoing embodiments, as Figure 2 shown, the embodiments of the present application provide a phased array antenna calibration system, wherein the system comprises:

[0088] The initial error matrix drafting module M100: based on the phased array antenna configuration information, a three-dimensional near-field calibration reference is established, the phase difference and amplitude error of the receiving antenna element are received, and an initial error matrix is drafted.

[0089] The grouping excitation module M200: group excitation of phased antenna array elements, determine the beam pointing deviation of each group of phased antenna array elements through the far-field direction vector, and configure the spatial angle error component.

[0090] The coupling analysis module M300: based on the initial error matrix and the spatial angle error component, determine the compensation coefficient using the least square method, extract the environmental disturbance compensation factor through the depth separable convolutional neural network, separate the coupling effects of thermal drift, mechanical deformation and electromagnetic interference, and generate a dynamic calibration weight matrix.

[0091] The calibration decision module M400: according to the dynamic calibration weight matrix, combined with the beam pointing accuracy requirement, deploy distributed calibration nodes, iteratively optimize calibration parameters in a multi-base station cooperative scenario, and at the same time, perform mutual coupling effect bias measurement and calibration decision.

[0092] Further, the calibration decision module M400 is also used to execute the following method:

[0093] According to the amplitude mutual coupling effect between the phased array antenna elements, a first adaptive calibration index is configured; according to the phase mutual coupling effect between the phased array antenna elements, a second adaptive calibration index is configured; based on the first adaptive calibration index and the second adaptive calibration index, a test period is determined; in the test period, according to the received signal amplitude of each phased antenna element separately transmitting a signal, the element phase difference is determined, the mutual coupling effect bias measurement and calibration decision are performed: the coupling coefficient between adjacent elements is calculated by port S parameter measurement, the bias of each phased antenna element is determined, and adaptive calibration is performed by fuzzy decision.

[0094] Further, the grouping excitation module M200 is used to execute the following method:

[0095] The phased antenna array surface is divided into N independent sub-arrays, each independent sub-array contains P phased antenna array elements, and the isolation degree of the independent sub-arrays meets the isolation degree limitation condition; according to the time sequence control logic, the N independent sub-arrays are polled and excited by high-speed radio frequency switches.

[0096] Further, the grouping excitation module M200 is also used to execute the following method:

[0097] The adjacent sub-array excitation interval corresponding to the N independent sub-arrays meets the time interval limitation condition; at the same time, according to the pointing deviation degree of the independent sub-array beam pointing and the target direction, a preset initial phase weight is applied to each independent sub-array.

[0098] Further, the grouping excitation module M200 is also used to execute the following method:

[0099] Determine a first time factor according to the operating frequency of the phased array antenna; determine a second time factor according to the time required for the energy of the first adjacent subarray signal to attenuate to the set interference threshold value from the emission to the collection of the second adjacent subarray signal by using the free space propagation model; determine a third time factor corresponding to the switching time of the high-speed radio frequency switch and a fourth time factor corresponding to the signal processing delay, and set the time interval limit condition in combination with the first time factor and the second time factor.

[0100] Further, the grouping excitation module M200 is also used to execute the following method:

[0101] According to the pointing deviation value of the independent subarray beam pointing and the target direction, decompose into an azimuth angle error component and a pitch angle error component; configure the spatial angle error component through the azimuth angle error component and the pitch angle error component.

[0102] Further, the initial error matrix formulation module M100 is also used to execute the following method:

[0103] According to the number and arrangement mode of the phased array antenna elements, build a spherical scanning frame with a radius of U times the center wavelength of the operating frequency band in the microwave darkroom; based on the spherical scanning frame, configure a vector network analyzer and a double-ridge horn probe to collect full-space field strength data through a preset scanning path; take the array center as the coordinate origin, associate the physical position parameters of each phased antenna element with the full-space field strength data to obtain a calibration reference database.

[0104] Further, the coupling analysis module M300 is also used to execute the following method:

[0105] Configure the target function as the mean square error of the beam pattern and the target beam pattern, and the constraint condition of the target function is the compensation coefficient adjustment range; based on the target function, determine an overdetermined equation set by using the weighted least squares method, and the weight coefficients are dynamically allocated according to the far field attenuation characteristics of each phased antenna element from the array center.

[0106] Further, the coupling analysis module M300 is also used to execute the following method:

[0107] Take the mapping relationship between the environmental parameters and the amplitude and phase errors in the historical calibration data as training samples; set a separate thermal drift branch, a mechanical deformation branch and an electromagnetic interference branch in the output layer of the deep separable convolutional neural network to decouple the coupling effects and formulate the dynamic calibration weight matrix.

[0108] In summary, any step can be stored in a computer memory without limitation as computer instructions or programs, and can be called and recognized by a computer processor without limitation, and no redundant limitation is made here.

[0109] Further, the above technical solutions only embody the preferred technical solutions of the technical solutions of the embodiments of the present application, and some variations made by the person skilled in the art to some parts thereof all embody the principles of the novel embodiments of the present application. Obviously, the person skilled in the art can make various modifications and variations to the present application without departing from the scope of the present application.

Claims

1. A method of calibrating a phased array antenna, characterized by, The method comprises: Based on the phased array antenna configuration information, a three-dimensional near-field calibration reference is established, the phase difference and amplitude error of the antenna unit are received, and an initial error matrix is formulated; Grouping excitation is performed on the phased antenna array elements, the beam pointing deviation of each group of phased antenna array elements is determined through the far-field direction vector, and the spatial angle error component is configured; Based on the initial error matrix and the spatial angle error component, the compensation coefficient is determined using the least squares method, the environmental disturbance compensation factor is extracted through the depth separable convolutional neural network, the coupling effects of thermal drift, mechanical deformation and electromagnetic interference are separated, and a dynamic calibration weight matrix is generated; According to the dynamic calibration weight matrix, combined with the beam pointing accuracy requirement, a distributed calibration node is deployed, the calibration parameters are iteratively optimized in the multi-base station cooperative scene, and at the same time, the mutual coupling effect bias measurement and calibration decision are performed.

2. The phased array antenna calibration method of claim 1, wherein, Before the mutual coupling effect bias measurement and calibration decision, the method comprises: According to the amplitude mutual coupling effect between the phased array antenna elements, a first adaptive calibration index is configured; according to the phase mutual coupling effect between the phased array antenna elements, a second adaptive calibration index is configured; Based on the first adaptive calibration index and the second adaptive calibration index, a test period is determined; In the test period, the element phase difference is determined according to the received signal amplitude of each phased antenna element separately transmitting a signal, and the mutual coupling effect bias measurement and calibration decision are performed: the coupling coefficient between adjacent elements is calculated through port S parameter measurement, the bias of each phased antenna element is determined, and adaptive calibration is performed by fuzzy decision.

3. The phased array antenna calibration method of claim 2, wherein, Grouping excitation is performed on the phased antenna array elements, the method comprises: The phased antenna array surface is divided into N independent sub-arrays, each independent sub-array contains P phased antenna array elements, and the isolation degree of the independent sub-arrays meets the isolation degree limitation condition; According to the timing control logic, the N independent sub-arrays are polled and excited through a high-speed radio frequency switch.

4. The phased array antenna calibration method of claim 3, wherein, The excitation interval of adjacent sub-arrays corresponding to the N independent sub-arrays meets the time interval limitation condition; At the same time, according to the pointing deviation of the independent sub-array beam pointing and the target direction, a preset initial phase weight is applied to each independent sub-array.

5. The phased array antenna calibration method of claim 4, wherein, The excitation interval of adjacent sub-arrays corresponding to the N independent sub-arrays meets the time interval limitation condition, and the method comprises: According to the operating frequency of the phased array antenna, a first time factor is determined; Using the free space propagation model, a second time factor is obtained, which is the time required for the energy attenuation of the signal from the first adjacent sub-array to the second adjacent sub-array to meet the set interference threshold value; The switching time of the high-speed radio frequency switch corresponds to a third time factor, the signal processing delay corresponds to a fourth time factor, and the first time factor and the second time factor are combined to set the time interval limitation condition.

6. The phased array antenna calibration method of claim 4, wherein, According to the pointing deviation of the independent sub-array beam pointing and the target direction, the pointing deviation is decomposed into an azimuth angle error component and an elevation angle error component; The spatial angle error component is configured through the azimuth angle error component and the elevation angle error component. ​ 7. The phased array antenna calibration method of claim 2, wherein, Based on the phased array antenna configuration information, a three-dimensional near-field calibration reference is established, and the method further comprises: According to the number and arrangement of the phased array antenna elements, a spherical scanning frame with a radius of U times the center wavelength of the working frequency band is built in a microwave darkroom; Based on the spherical scanning frame, a vector network analyzer and a double-ridge horn probe are configured to collect full-space field strength data through a preset scanning path; Taking the array center as the coordinate origin, the physical position parameters of each phased antenna element are associated with the full-space field strength data to obtain a calibration reference database.

8. The phased array antenna calibration method of claim 7, wherein, The least squares method is used to determine the compensation coefficient, and the environmental disturbance compensation factor is extracted through a deep separable convolutional neural network, and the method further comprises: The objective function is configured as the mean square error of the beam pattern and the target beam pattern, and the constraint condition of the objective function is the compensation coefficient adjustment range; Based on the objective function, a weighted least squares method is used to determine an overdetermined equation set, and the weight coefficients are dynamically allocated according to the far field attenuation characteristics of each phased array antenna element from the array center.

9. The phased array antenna calibration method of claim 8, wherein, The environmental disturbance compensation factor is extracted through a deep separable convolutional neural network, the coupling effects of thermal drift, mechanical deformation and electromagnetic interference are separated, and a dynamic calibration weight matrix is generated, the method comprising: The mapping relationship between the environmental parameters and the amplitude and phase errors in the historical calibration data is taken as a training sample; The output layer of the deep separable convolutional neural network is set to have a thermal drift branch, a mechanical deformation branch and an electromagnetic interference branch to decouple the coupling effects and formulate the dynamic calibration weight matrix.

10. A phased array antenna calibration system, characterized by, The system for implementing the steps of the phased array antenna calibration method of any one of claims 1-9, comprising: An initial error matrix formulation module: based on the phased array antenna configuration information, a three-dimensional near-field calibration reference is established, the phase difference and amplitude error of the antenna element are received, and an initial error matrix is formulated; A grouping excitation module: the phased antenna elements are grouped and excited, the beam pointing deviation of each group of phased antenna elements is determined through the far-field direction vector, and the spatial angle error component is configured; A coupling analysis module: based on the initial error matrix and the spatial angle error component, the least squares method is used to determine the compensation coefficient, the environmental disturbance compensation factor is extracted through a deep separable convolutional neural network, the coupling effects of thermal drift, mechanical deformation and electromagnetic interference are separated, and a dynamic calibration weight matrix is generated; A calibration decision module: according to the dynamic calibration weight matrix, a distributed calibration node is deployed according to the beam pointing accuracy requirement, the calibration parameters are iteratively optimized in a multi-base station cooperative scenario, and at the same time, the mutual coupling effect bias is measured and the calibration decision is made.

Images