X-band radar antenna intelligent horizontal correction method and system

Abstract

The invention discloses an intelligent horizontal correction method and system for an X-band radar antenna, and belongs to the technical field of data analysis. The invention provides a radar antenna array calibration scheme based on a multi-dimensional state attribute vector set and a machine learning algorithm, through deep analysis of sensing measurement state data, X-waveband performance characteristics of the antenna array can be accurately identified, and specific correction decision guidance is provided for each performance characteristic. Therefore, intelligent and automatic calibration of the radar antenna array can be realized, the calibration accuracy and efficiency are remarkably improved, and a more reliable guarantee is provided for wide application of a radar system.

Description

Technical Field

[0001] The present application belongs to the field of data analysis technology, and specifically relates to an X-band radar antenna intelligent horizontal correction method and system. Background Art

[0002] In the field of radar technology, accurate and efficient calibration of antenna arrays is the key to ensuring stable system performance. However, traditional calibration methods often rely on manual intervention and regular maintenance, which is not only time-consuming and labor-intensive, but may also introduce errors due to human factors. Especially in complex electromagnetic environments and changing climatic conditions, the performance of radar antennas may be affected by a variety of factors and gradually decline, resulting in reduced monitoring accuracy and reliability. Therefore, there is an urgent need for a technical solution that can automatically and intelligently calibrate radar antenna arrays to improve the efficiency and accuracy of calibration and reduce operation and maintenance costs. Summary of the invention

[0003] The present application provides an X-band radar antenna intelligent horizontal correction method and system, which can solve or partially solve the technical problems involved in the above-mentioned background technology.

[0004] The present application embodiment provides an X-band radar antenna intelligent horizontal correction method, which is applied to an AI correction system. The method includes:

[0005] Acquire sensor measurement state data of the target radar antenna array, and determine a multidimensional state attribute vector set of the sensor measurement state data;

[0006] Determining a target performance feature identification result based on the multidimensional state attribute vector set, wherein the target performance feature identification result includes a plurality of first antenna performance feature units, and a feature weight of each first antenna performance feature unit indicates a confidence level that the first antenna performance feature unit is an X-band performance feature;

[0007] Determining a target proofreading feature recognition result based on the multi-dimensional state attribute vector set, wherein the target proofreading feature recognition result includes a plurality of second antenna performance feature units, each second antenna performance feature unit corresponds to one of the first antenna performance feature units, and a feature weight of each second antenna performance feature unit indicates a correction decision feature matched by the first antenna performance feature unit corresponding to the second antenna performance feature unit;

[0008] Determine X-band performance characteristics and horizontal correction expected characteristics based on the target performance characteristic recognition result and the target calibration characteristic recognition result;

[0009] The distribution information and deviation information of the data to be adjusted are determined in the sensor measurement state data according to the X-band performance characteristics and the expected characteristics of the horizontal correction, and the target radar antenna array is calibrated based on the distribution information and the deviation information.

[0010] In one implementation, determining a target performance feature identification result based on the multidimensional state attribute vector set includes:

[0011] Performing a feature embedding operation on the multidimensional state attribute vector set to determine a multidimensional state attribute vector set that has completed feature embedding, wherein the multidimensional state attribute vector set that has completed feature embedding includes a plurality of third antenna performance feature units;

[0012] For each third antenna performance characteristic unit in the multidimensional state attribute vector set that has completed feature embedding, determine the target pointing angle characteristic variable of the multidimensional state attribute vector set that has completed feature embedding in the pointing dimension, determine the target phase angle characteristic variable of the multidimensional state attribute vector set that has completed feature embedding in the phase dimension, and determine the characteristic weight of the first antenna performance characteristic unit in the target performance characteristic identification result based on the target pointing angle characteristic variable and the target phase angle characteristic variable.

[0013] In one implementation, determining the target pointing angle feature variable of the multidimensional state attribute vector set in the pointing dimension for completing feature embedding includes:

[0014] Traversing the feature weights of all third antenna performance feature units between the third antenna performance feature unit and the first pointing feature tag of the multidimensional state attribute vector set based on the pointing dimension, and determining the maximum value among the traversed feature weights as the first target pointing angle feature variable, and / or traversing the feature weights of all third antenna performance feature units between the third antenna performance feature unit and the second pointing feature tag of the multidimensional state attribute vector set based on the pointing dimension, and determining the maximum value among the traversed feature weights as the second target pointing angle feature variable, determining the target phase angle feature variable of the multidimensional state attribute vector set that completes feature embedding in the phase dimension includes:

[0015] Traversing the characteristic weights of all third antenna performance characteristic units between the third antenna performance characteristic unit and the first phase characteristic tag of the multidimensional state attribute vector set based on the phase dimension, and determining the maximum value among the traversed characteristic weights as the first target phase angle characteristic variable, and / or traversing the characteristic weights of all third antenna performance characteristic units between the third antenna performance characteristic unit and the second phase characteristic tag of the multidimensional state attribute vector set based on the phase dimension, and determining the maximum value among the traversed characteristic weights as the second target phase angle characteristic variable, determining the characteristic weight of the first antenna performance characteristic unit in the target performance characteristic identification result according to the target pointing angle characteristic variable and the target phase angle characteristic variable includes:

[0016] The first target pointing angle characteristic variable and / or the second target pointing angle characteristic variable and the first target phase angle characteristic variable and / or the second target phase angle characteristic variable are weighted to determine the characteristic weight of the first antenna performance characteristic unit corresponding to the third antenna performance characteristic unit in the target performance characteristic identification result.

[0017] In one implementation, the target performance feature identification result includes at least one performance attention indicator, and the feature weight of the first antenna performance feature unit in each performance attention indicator indicates the confidence that the first antenna performance feature unit is an X-band performance feature of a target of a corresponding performance category.

[0018] In one implementation, determining a target proofreading feature recognition result based on the multi-dimensional state attribute vector set includes:

[0019] A feature embedding operation is performed on the multidimensional state attribute vector set to obtain the target calibration feature identification result, wherein for the X-band performance characteristics of each performance type, the target calibration feature identification result includes at least one performance attention indicator, and the feature weight indication of the second antenna performance characteristic unit of each performance attention indicator corresponds to the first antenna performance characteristic unit of the second antenna performance characteristic unit as a performance correction reference under the set correction rule matched by the X-band performance characteristics of the performance type.

[0020] In one implementation, determining the X-band performance characteristic and the horizontal correction expected characteristic based on the target performance characteristic recognition result and the target calibration characteristic recognition result includes:

[0021] Determine, based on the feature weight of each first antenna performance feature unit in the target performance feature identification result, a discrimination score that the first antenna performance feature unit belongs to the X-band performance feature;

[0022] The first antenna performance feature unit whose discrimination score is greater than the set discrimination score is determined as the X-band performance feature; and based on the distribution characteristics of the X-band performance feature in the target performance feature identification result, the horizontal correction expected feature corresponding to the X-band performance feature is determined in the target proofreading feature identification result.

[0023] In one implementation, determining the distribution information and deviation information of the data to be adjusted in the sensor measurement state data according to the X-band performance characteristics and the expected characteristics of the horizontal correction includes:

[0024] Determine a characteristic association description between the multidimensional state attribute vector set and the sensor measurement state data;

[0025] Determining distribution information and initial noise information of the data to be adjusted in the sensor measurement state data according to the X-band performance characteristics, the expected characteristics of the horizontal correction, and the characteristic association description;

[0026] Determining a noise disturbance for the initial noise information based on the multidimensional state attribute vector set;

[0027] The deviation information of the data to be adjusted is determined in the sensor measurement state data according to the noise disturbance and the initial noise information.

[0028] In one implementation, the X-band performance characteristics and the horizontal correction desired characteristics include X-band performance characteristics of the data to be adjusted for at least one time series tag and horizontal correction desired characteristics of the data to be adjusted for the performance category.

[0029] In one implementation, the sensor measurement status data is radar antenna array status data collected by a built-in measurement device. Before determining the multidimensional state attribute vector set of the sensor measurement status data, the method also includes: adjusting the original characteristic value of each antenna performance characteristic unit in the sensor measurement status data.

[0030] In one implementation, determining the multidimensional state attribute vector set of the sensor measurement state data includes:

[0031] Processing the sensor measurement state data through at least one feature embedding branch to obtain sensor measurement state data with feature embedding completed;

[0032] The sensor measurement state data that has completed feature embedding is subjected to at least one feature pooling and at least one feature derivation branch to determine a multidimensional state attribute vector set of the sensor measurement state data, wherein the feature granularity of the multidimensional state attribute vector set is smaller than that of the sensor measurement state data.

[0033] In one implementation, the multidimensional state attribute vector set, the target performance feature recognition result, and the target calibration feature recognition result are generated by at least one long short-term memory model, and the at least one long short-term memory model is debugged by the following steps:

[0034] Determine a debugging sample set, wherein the debugging sample set includes at least one sensor measurement state data sample, and the at least one sensor measurement state data sample carries prior distribution information and prior deviation information of prior data to be adjusted;

[0035] For each sensor measurement status data sample:

[0036] Determine a multi-dimensional state attribute vector set sample of the sensor measurement state data sample;

[0037] Determine a target performance feature recognition result sample based on the multi-dimensional state attribute vector set sample;

[0038] Determine a target proofreading feature recognition result sample based on the multi-dimensional state attribute vector set sample;

[0039] Determine an X-band performance characteristic sample and a horizontal correction expected characteristic sample based on the target performance characteristic recognition result sample and the target calibration characteristic recognition result sample;

[0040] Determining a distribution information sample and a deviation information sample of a data sample to be adjusted in the sensor measurement state data sample according to the X-band performance characteristic sample and the horizontal correction expected characteristic sample;

[0041] The model weights of the at least one long short-term memory model are optimized so that the difference between the distribution information sample and the deviation information sample of the data sample to be adjusted in the sensor measurement state data sample and the prior distribution information and the prior deviation information of the prior data to be adjusted converges.

[0042] In one implementation, the difference includes at least one of the following comparison errors: a distribution comparison error between the distribution information sample and the prior distribution information; and a deviation comparison error between the deviation information sample and the prior deviation information.

[0043] An embodiment of the present application provides an AI correction system, comprising at least one processor and a memory; the memory stores computer-executable instructions; the at least one processor executes the computer-executable instructions stored in the memory, so that the at least one processor performs the above method.

[0044] An embodiment of the present application provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the above method are implemented.

[0045] The embodiment of the present application proposes a radar antenna array calibration scheme based on a multi-dimensional state attribute vector set and a machine learning algorithm. By deeply analyzing the sensor measurement state data, the X-band performance characteristics of the antenna array can be accurately identified, and specific correction decision guidance can be provided for each performance characteristic. In this way, intelligent and automated calibration of the radar antenna array can be achieved, the accuracy and efficiency of calibration can be significantly improved, and a more reliable guarantee can be provided for the widespread application of radar systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 A flowchart of an X-band radar antenna intelligent horizontal correction method provided in an embodiment of the present application.

[0047] Figure 2 A schematic diagram of the structure of an AI correction system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0049] The terms "first", "second", etc. in this application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by "first", "second", etc. are usually a class, and the number of objects is not limited. For example, the first object can be one or more. In addition, in this application, "and / or" represents at least one of the connected objects, and the character " / " generally indicates that the objects associated with each other are in an "or" relationship.

[0050] Figure 1 An X-band radar antenna intelligent horizontal correction method is shown, which is applied to an AI correction system. The method includes the following steps 110 to 150.

[0051] In modern radar systems, X-band radars are favored for their powerful penetration and high resolution. However, to ensure the continued efficient operation of such radar systems, it is crucial to accurately horizontally calibrate their antenna arrays. Traditional correction methods often rely on manual intervention and regular maintenance, which is not only time-consuming and labor-intensive, but may also introduce errors due to human factors. Based on this, the AI ​​correction system in the embodiment of the present application can automatically analyze the electronic level data, identify and predict the horizontal deviation of the antenna, and thus guide the system to make precise adjustments. The following describes in detail the implementation process of the embodiment of the present application through steps 110-150 and related application scenario examples.

[0052] Step 110: Acquire sensor measurement state data of the target radar antenna array, and determine a multi-dimensional state attribute vector set of the sensor measurement state data.

[0053] For example, the AI ​​correction system first obtains the sensor measurement status data of the target radar antenna array in real time through the sensors and measuring devices built into the radar system. The sensor measurement status data covers multiple key parameters of the antenna array, including but not limited to pointing angle, signal strength, phase difference, ambient temperature, humidity, and possible mechanical vibration. These sensor measurement status data are organized into a multidimensional state attribute vector set, which provides a basis for subsequent analysis and processing.

[0054] Step 120: determine a target performance characteristic identification result based on the multidimensional state attribute vector set, wherein the target performance characteristic identification result includes a plurality of first antenna performance characteristic units, and a characteristic weight of each first antenna performance characteristic unit indicates a confidence level that the first antenna performance characteristic unit is an X-band performance characteristic.

[0055] Once sufficient sensor measurement state data has been collected, the AI ​​correction system begins to analyze these sensor measurement state data to determine the target performance feature identification results. This process involves in-depth mining of the multi-dimensional state attribute vector set to identify key features that reflect the performance of the antenna array. These features are divided into several first antenna performance feature units, each of which represents a specific X-band performance feature, such as signal transmission efficiency, phase consistency, or pointing accuracy. Each first antenna performance feature unit is assigned a feature weight that indicates the confidence of the feature unit as an X-band performance feature, that is, its importance in the overall performance evaluation.

[0056] Step 130: determine a target proofreading feature recognition result based on the multi-dimensional state attribute vector set, wherein the target proofreading feature recognition result includes a plurality of second antenna performance feature units, each second antenna performance feature unit corresponds to one of the first antenna performance feature units, and a feature weight of each second antenna performance feature unit indicates a correction decision feature matched by the first antenna performance feature unit corresponding to the second antenna performance feature unit.

[0057] After identifying the antenna performance characteristics, the AI ​​correction system further analyzes this data to determine the target calibration feature identification results. This process aims to find the correction decision feature corresponding to each first antenna performance feature unit. Therefore, each first antenna performance feature unit corresponds to a second antenna performance feature unit, which contains the correction strategy information required to achieve performance optimization. Similarly, each second antenna performance feature unit is also assigned a feature weight that reflects the effectiveness and applicability of the corresponding correction decision feature.

[0058] Step 140: Determine X-band performance characteristics and horizontal correction expected characteristics based on the target performance characteristic recognition result and the target calibration characteristic recognition result.

[0059] Based on the target performance feature identification results and the target calibration feature identification results, the AI ​​correction system can comprehensively evaluate the current performance status of the antenna array and determine the key X-band performance characteristics and the required horizontal correction expected characteristics. This step is a bridge between performance analysis and actual correction actions. It ensures that the correction strategy can solve the specific problems of the antenna array in a targeted manner, thereby improving the accuracy and efficiency of the correction.

[0060] Step 150: Determine distribution information and deviation information of the data to be adjusted in the sensor measurement state data according to the X-band performance characteristics and the expected characteristics of the horizontal correction, and calibrate the target radar antenna array based on the distribution information and the deviation information.

[0061] Finally, the AI ​​correction system further analyzes the distribution information and deviation information of the data to be adjusted in the sensor measurement status data based on the determined X-band performance characteristics and the expected characteristics of the horizontal correction. This information provides the system with clear guidance on how to adjust the antenna array to achieve optimal performance. Based on these analyses, the AI ​​correction system can automatically perform calibration procedures and adjust the control parameters of the antenna array, such as phase control codes, phase shifter settings, etc., to achieve accurate horizontal correction. After the calibration is completed, the AI ​​correction system will also verify the correction results to ensure that the performance of the antenna array has been effectively improved.

[0062] Taking airport surface surveillance radar as an example, this type of airport surface surveillance radar usually needs to operate continuously in complex electromagnetic environments and changeable climate conditions to ensure real-time monitoring of key areas such as airport runways and aprons. However, due to environmental factors (such as temperature fluctuations, wind shear) and mechanical wear, the horizontal pointing accuracy of the radar antenna may gradually decrease, thus affecting the accuracy and reliability of monitoring.

[0063] In this case, the AI ​​correction system is integrated into the radar control software to achieve intelligent horizontal correction of the antenna array. The AI ​​correction system first collects the sensing measurement status data of the antenna array in real time through built-in sensors and measuring equipment, including pointing angle, signal strength, phase difference, and ambient temperature and humidity. Then, the AI ​​correction system uses machine learning algorithms to analyze this data and identify key features that reflect antenna performance, such as signal transmission efficiency and phase consistency.

[0064] Next, the AI ​​correction system determines the corresponding correction decision characteristics based on these performance characteristics, such as adjusting the phase control code to improve phase consistency or adjusting the antenna pointing angle to optimize signal transmission efficiency. After determining the key X-band performance characteristics and the desired characteristics of the required horizontal correction, the AI ​​correction system begins the calibration procedure. It automatically adjusts the control parameters of the antenna array to achieve precise horizontal correction, and monitors the performance changes during the calibration process in real time to ensure the accuracy and stability of the calibration results.

[0065] By introducing the AI ​​correction system, the horizontal correction process of the airport surface surveillance radar antenna has achieved significant intelligence and automation. The AI ​​correction system can automatically identify and predict the horizontal deviation of the antenna, thereby guiding precise adjustment and significantly improving the efficiency and accuracy of calibration. At the same time, due to the reduction of manual intervention, the operation and maintenance costs of the AI ​​correction system have also been effectively reduced, providing a more reliable guarantee for the safe operation of the airport.

[0066] In summary, the application of AI correction system in intelligent horizontal correction of X-band radar antennas demonstrates its great potential and value. By integrating advanced AI algorithms and hardware acceleration technology, the AI ​​correction system can achieve efficient and accurate correction of antenna arrays, providing strong support for the widespread application of airport surface surveillance radars.

[0067] It is worth mentioning that steps 120 to 130 are intended to accurately identify the performance characteristics of the target radar antenna array and further determine the corresponding correction strategy by deeply analyzing and processing the multi-dimensional state attribute vector set.

[0068] First, in step 120, the AI ​​correction system will use advanced machine learning algorithms to perform in-depth mining and analysis based on the collected multi-dimensional state attribute vector set. In this process, the AI ​​correction system will identify several key first antenna performance feature units, which can fully reflect the performance of the antenna array in the X-band. In order to quantify the importance of each feature unit in the overall performance evaluation, the AI ​​correction system will assign a feature weight to each first antenna performance feature unit, which actually indicates the confidence of the feature unit as an X-band performance feature, that is, its accuracy and reliability in describing antenna performance.

[0069] Secondly, in step 130, the AI ​​correction system will further utilize the multi-dimensional state attribute vector set to determine the target calibration feature recognition result. The core of this step is to find a corresponding second antenna performance feature unit for each first antenna performance feature unit that has been identified. This corresponding second antenna performance feature unit actually contains the correction strategy information required to achieve performance optimization, which guides the AI ​​correction system on how to adjust the antenna array to improve its performance. In order to evaluate the effectiveness of each correction strategy, the AI ​​correction system assigns a feature weight to each second antenna performance feature unit. This weight reflects the effectiveness and applicability of the corresponding correction decision feature, that is, the possibility of successfully improving antenna performance in practical applications.

[0070] Through these two steps, the AI ​​correction system can not only accurately identify the key performance characteristics of the antenna array in the X-band, but also provide specific and quantitative correction strategy guidance for each performance characteristic. This in-depth analysis and processing based on a multi-dimensional state attribute vector set ensures the pertinence and effectiveness of the correction strategy, thereby significantly improving the performance correction efficiency and accuracy of the antenna array.

[0071] On this basis, step 150 focuses on how to utilize the identified X-band performance characteristics and the desired characteristics of horizontal correction to accurately calibrate the target radar antenna array.

[0072] First, the AI ​​correction system will deeply mine the sensor measurement status data based on the identified X-band performance characteristics and horizontal correction expectations. In this process, the AI ​​correction system will pay special attention to those data points that are closely related to the performance characteristics and correction expectations to reveal the distribution information and deviation information of the data to be adjusted. The distribution information describes the distribution state of the data in multidimensional space, while the deviation information reflects the difference between the current antenna array performance and the expected performance.

[0073] Secondly, the AI ​​correction system will develop targeted calibration strategies based on the mined distribution information and deviation information. This strategy aims to gradually reduce the deviation by adjusting the control parameters of the antenna array, such as phase control codes, phase shifter settings, etc., so that the performance of the antenna array gradually approaches the expected level. When formulating the calibration strategy, the AI ​​correction system will comprehensively consider various factors, such as the feasibility of the adjustment, the impact on other parts of the AI ​​correction system, and the timeliness of the calibration to ensure the smooth progress of the calibration process.

[0074] Finally, the AI ​​correction system will execute the calibration strategy and perform actual calibration operations on the target radar antenna array. During the calibration process, the AI ​​correction system will monitor the performance changes of the antenna array in real time to ensure the accuracy and stability of the calibration results. Once the calibration is completed, the system will also verify the calibration results and evaluate the effectiveness of the calibration by comparing them with the expected performance. If the calibration results meet expectations, the AI ​​correction system will end the calibration process; if they do not meet expectations, the AI ​​correction system will adjust the calibration strategy according to the actual situation and continue to perform calibration operations until a satisfactory calibration effect is achieved.

[0075] It can be seen that the technical solution recorded in step 150 achieves accurate calibration of the target radar antenna array by deeply mining the sensor measurement state data, formulating and executing targeted calibration strategies. In this way, not only the efficiency and accuracy of the calibration are improved, but also a strong guarantee is provided for the stable operation of the radar system.

[0076] The embodiment of the present application achieves a comprehensive and in-depth analysis of the antenna array performance by acquiring the sensor measurement state data of the target radar antenna array and constructing a multi-dimensional state attribute vector set. By further determining the target performance feature identification results and the target calibration feature identification results, the X-band performance characteristics can be accurately identified, and specific correction decision guidance can be provided for each performance feature. Based on these identification results, the distribution information and deviation information of the data to be adjusted can be accurately determined, thereby achieving accurate calibration of the target radar antenna array. In summary, the embodiment of the present application significantly improves the efficiency and accuracy of calibration, reduces the need for manual intervention, and provides a strong guarantee for the stable operation of the radar system.

[0077] In some optional embodiments, determining a target performance characteristic identification result based on the multidimensional state attribute vector set includes: performing a feature embedding operation on the multidimensional state attribute vector set to determine a multidimensional state attribute vector set that has completed feature embedding, wherein the multidimensional state attribute vector set that has completed feature embedding includes a plurality of third antenna performance characteristic units; for each third antenna performance characteristic unit in the multidimensional state attribute vector set that has completed feature embedding, determining a target pointing angle characteristic variable of the multidimensional state attribute vector set that has completed feature embedding in the pointing dimension, determining a target phase angle characteristic variable of the multidimensional state attribute vector set that has completed feature embedding in the phase dimension, and determining a characteristic weight of the first antenna performance characteristic unit in the target performance characteristic identification result based on the target pointing angle characteristic variable and the target phase angle characteristic variable.

[0078] Applying this embodiment, the process of determining the target performance feature identification result based on the multidimensional state attribute vector set first involves a feature embedding operation on the multidimensional state attribute vector set, aiming to convert the original state attribute vector that may contain redundancy and noise into a more expressive and discriminative feature vector. Through this feature embedding, the AI ​​correction system can extract key features that are closely related to antenna performance from high-dimensional data, laying a solid foundation for subsequent performance feature identification.

[0079] Exemplarily, the feature embedding operation is implemented by applying a series of mathematical transformations and mapping functions to the multidimensional state attribute vector set. These transformations and functions may be linear or nonlinear, depending on the characteristics of the data and the type of features required. During the feature embedding process, the AI ​​correction system attempts to retain the information that is most critical to antenna performance identification, while removing or reducing the interference of irrelevant or redundant information. After the feature embedding is completed, the multidimensional state attribute vector set will be converted into a new feature vector set containing several third antenna performance feature units.

[0080] Next, for each third antenna performance characteristic unit in the multidimensional state attribute vector set that completes feature embedding, the AI ​​correction system will further determine its target pointing angle characteristic variable in the pointing dimension. The pointing dimension is an important aspect of radar antenna performance, which reflects the pointing capability of the antenna beam in space. By determining the target pointing angle characteristic variable, the AI ​​correction system is able to evaluate the radiation intensity and pointing accuracy of the antenna in a specific direction. The implementation of this step may rely on advanced signal processing algorithms and pattern recognition techniques to ensure accurate estimation of the pointing angle.

[0081] At the same time, the AI ​​correction system also determines the target phase angle characteristic variables in the phase dimension of the multidimensional state attribute vector set that completes the feature embedding. The phase dimension is another key performance indicator that reflects the phase relationship between the elements in the antenna array. The correct phase relationship is the basis for achieving antenna beamforming and pointing control. Therefore, by determining the target phase angle characteristic variables, the AI ​​correction system can evaluate the phase consistency and phase control capabilities of the antenna array. The implementation of this step also requires the use of precise mathematical models and algorithms to ensure accurate measurement and calculation of the phase angle.

[0082] Finally, the AI ​​correction system will determine the feature weight of the first antenna performance feature unit in the target performance feature identification result based on the target pointing angle feature variable and the target phase angle feature variable. The feature weight is a quantitative indicator that reflects the confidence and importance of each first antenna performance feature unit in describing the antenna X-band performance. By comprehensively considering the feature variables in the pointing dimension and the phase dimension, the AI ​​correction system can more comprehensively evaluate the contribution of each performance feature unit to the overall performance and assign corresponding feature weights accordingly. This process may require the application of machine learning algorithms or statistical models to achieve accurate calculation and allocation of feature weights.

[0083] It can be seen that the process of determining the target performance feature identification results based on a multi-dimensional state attribute vector set is a comprehensive data analysis and processing process. Through feature embedding operations, determination of feature variables in the pointing dimension and phase dimension, and calculation and allocation of feature weights, the AI ​​correction system can achieve accurate identification and quantitative evaluation of the X-band performance of the radar antenna. This technical solution provides an important basis and support for subsequent radar antenna calibration and optimization. It not only improves the accuracy and efficiency of calibration, but also lays a solid foundation for the stable operation and performance improvement of the radar system.

[0084] In the next step, determining the target pointing angle characteristic variable of the multidimensional state attribute vector set that has completed feature embedding in the pointing dimension includes: traversing the characteristic weights of all third antenna performance characteristic units between the third antenna performance characteristic unit and the first pointing characteristic tag of the multidimensional state attribute vector set based on the pointing dimension, and determining the maximum value among the traversed characteristic weights as the first target pointing angle characteristic variable, and / or traversing the characteristic weights of all third antenna performance characteristic units between the third antenna performance characteristic unit and the second pointing characteristic tag of the multidimensional state attribute vector set based on the pointing dimension, and determining the maximum value among the traversed characteristic weights as the second target pointing angle characteristic variable, determining the target phase angle characteristic variable of the multidimensional state attribute vector set that has completed feature embedding in the phase dimension includes: traversing the characteristic weights of all third antenna performance characteristic units between the third antenna performance characteristic unit and the first phase characteristic tag of the multidimensional state attribute vector set based on the phase dimension. The characteristic weights of all third antenna performance characteristic units between the third antenna performance characteristic unit and the second phase characteristic tag of the multidimensional state attribute vector set are traversed based on the phase dimension, and the maximum value among the traversed characteristic weights is determined as the first target phase angle characteristic variable, and / or the characteristic weights of all third antenna performance characteristic units between the third antenna performance characteristic unit and the second phase characteristic tag of the multidimensional state attribute vector set are traversed based on the phase dimension, and the maximum value among the traversed characteristic weights is determined as the second target phase angle characteristic variable, and the characteristic weight of the first antenna performance characteristic unit in the target performance characteristic identification result is determined according to the target pointing angle characteristic variable and the target phase angle characteristic variable, including: weighting the first target pointing angle characteristic variable and / or the second target pointing angle characteristic variable with the first target phase angle characteristic variable and / or the second target phase angle characteristic variable, and determining the characteristic weight of the first antenna performance characteristic unit corresponding to the third antenna performance characteristic unit in the target performance characteristic identification result.

[0085] Based on this embodiment, determining the target pointing angle characteristic variable of the multidimensional state attribute vector set in the pointing dimension for completing feature embedding is a key process, which involves a comprehensive examination of the feature weights between the third antenna performance characteristic unit and the pointing feature label in the multidimensional state attribute vector set. Exemplarily, the AI ​​correction system will first traverse the feature weights of all third antenna performance characteristic units between the third antenna performance characteristic unit and the first pointing feature label of the multidimensional state attribute vector set based on the pointing dimension. The traversal process aims to find the third antenna performance characteristic unit that best matches the first pointing feature label, that is, the unit whose feature weight reaches the maximum value. This maximum value is determined as the first target pointing angle characteristic variable, which represents the strongest association between the third antenna performance characteristic unit and the first pointing feature label in the pointing dimension.

[0086] At the same time, and / or to represent another possible situation, the AI ​​correction system will also traverse the feature weights of all third antenna performance feature units between the third antenna performance feature unit and the second pointing feature tag of the multidimensional state attribute vector set based on the pointing dimension. Similarly, the purpose of the traversal process is to find the third antenna performance feature unit that best matches the second pointing feature tag, and the maximum value of its feature weight is determined as the second target pointing angle feature variable. This variable reflects the strongest association between the third antenna performance feature unit and the second pointing feature tag in the pointing dimension.

[0087] After determining the target pointing angle characteristic variable, the AI ​​correction system will then determine the target phase angle characteristic variable of the multidimensional state attribute vector set that completes the feature embedding in the phase dimension. This process is similar to the processing of the pointing dimension, but the focus is on the phase feature label. Exemplarily, the AI ​​correction system will traverse the feature weights of all third antenna performance characteristic units between the third antenna performance characteristic unit and the first phase feature label of the multidimensional state attribute vector set based on the phase dimension, and determine the maximum value of the traversed feature weights as the first target phase angle characteristic variable. This variable represents the strongest association between the third antenna performance characteristic unit and the first phase feature label in the phase dimension.

[0088] Similarly, and / or indicates that the AI ​​correction system will also consider another situation, that is, traversing the characteristic weights of all third antenna performance characteristic units between the third antenna performance characteristic unit and the second phase characteristic label of the multidimensional state attribute vector set based on the phase dimension. The purpose of the traversal process is to find the third antenna performance characteristic unit that best matches the second phase characteristic label, and the maximum value of its characteristic weight is determined as the second target phase angle characteristic variable. This variable reflects the strongest association between the third antenna performance characteristic unit and the second phase characteristic label in the phase dimension.

[0089] Finally, based on the determined target pointing angle characteristic variable and target phase angle characteristic variable, the AI ​​correction system will further determine the characteristic weight of the first antenna performance characteristic unit in the target performance characteristic identification result. This step is achieved by weighting the first target pointing angle characteristic variable and / or the second target pointing angle characteristic variable with the first target phase angle characteristic variable and / or the second target phase angle characteristic variable. The specific method of weighting may depend on the actual application scenario and specific needs, but the core purpose is to comprehensively consider the characteristic variables in the pointing dimension and the phase dimension to determine the confidence and importance of the first antenna performance characteristic unit in the overall performance evaluation. Through this weighting process, the AI ​​correction system can derive the characteristic weight of the first antenna performance characteristic unit corresponding to the third antenna performance characteristic unit, thereby providing important basis and support for subsequent radar antenna calibration and optimization.

[0090] Such a design determines the target characteristic variables in the pointing dimension and phase dimension of the multidimensional state attribute vector set that completes feature embedding, and determines the feature weights in the target performance feature identification results based on these variables. This is a comprehensive data analysis and processing process. This process not only involves an in-depth traversal and investigation of the multidimensional state attribute vector set, but also requires a comprehensive consideration of the characteristic variables in different dimensions and their interrelationships. Through this technical solution, the AI ​​correction system can achieve accurate identification and quantitative evaluation of radar antenna performance, providing strong support for subsequent antenna calibration and optimization. This not only improves the accuracy and efficiency of calibration, but also lays a solid foundation for the stable operation and performance improvement of the radar system.

[0091] In some examples, the target performance feature identification result includes at least one performance attention indicator, and the feature weight of the first antenna performance feature unit in each performance attention indicator indicates the confidence that the first antenna performance feature unit is an X-band performance feature of the target of the corresponding performance category.

[0092] It can be understood that in actual application, the target performance feature identification result is a comprehensive and in-depth evaluation product, which not only includes multiple dimensions of radar antenna performance, but also quantifies the characteristics and importance of these performances through specific performance indicators. Among them, a key component is the performance attention indicator, which reveals different aspects of antenna performance characteristics in a unique way.

[0093] The performance attention index in the embodiment of the present application is a quantitative measure used to express the confidence of the X-band performance characteristics exhibited by the first antenna performance characteristic unit under a specific performance category. "Confidence" in the embodiment of the present application is a common term in statistics and machine learning, which reflects the degree of trust the AI ​​correction system has in a certain result or prediction. In the performance evaluation of radar antennas, the higher the confidence, the more confident the AI ​​correction system is in the performance of the antenna in a specific performance.

[0094] Based on this, the performance attention index is constructed based on the feature embedding and feature weight calculation of the multi-dimensional state attribute vector set mentioned above. Specifically, when the AI ​​correction system performs feature embedding operations on the multi-dimensional state attribute vector set, the AI ​​correction system is actually converting these high-dimensional raw data that may contain redundancy and noise into more expressive and discriminative feature vectors. Each element in these feature vectors can be regarded as a quantitative representation of a specific aspect of antenna performance.

[0095] Next, after determining the multidimensional state attribute vector set that completes feature embedding, the AI ​​correction system will further calculate the target characteristic variables of each third antenna performance characteristic unit in the pointing dimension and phase dimension, such as the target pointing angle characteristic variable and the target phase angle characteristic variable. These characteristic variables actually describe the radiation intensity and pointing accuracy of the antenna in a specific direction, as well as the phase relationship between the elements in the antenna array. By calculating these characteristic variables, the AI ​​correction system can have a more comprehensive understanding of the performance characteristics of the antenna.

[0096] However, it is not enough to simply identify the performance characteristics of the antenna, the AI ​​correction system also needs to identify the importance of these characteristics in the overall performance evaluation. This is where the feature weight calculation comes in. By calculating the feature weight of each third antenna performance feature unit, the AI ​​correction system is able to quantify the confidence and importance of that unit in describing the antenna's X-band performance. These feature weights actually tell the AI ​​correction system which performance characteristics are most critical to the antenna performance evaluation and which ones can be ignored or considered secondary.

[0097] Finally, when the AI ​​correction system has these feature weights, the AI ​​correction system can construct a performance attention index. Specifically, the AI ​​correction system maps the feature weight of each third antenna performance feature unit to the corresponding first antenna performance feature unit to obtain the feature weight of the first antenna performance feature unit. This feature weight is actually part of the performance attention index, which represents the confidence of the first antenna performance feature unit in the target X-band performance feature of the corresponding performance category.

[0098] Through this process, the AI ​​correction system can obtain a target performance feature recognition result that includes multiple performance attention indicators. This result not only tells the AI ​​correction system which performance the antenna performs well in, but also tells the AI ​​correction system the importance of these performances in the overall evaluation. This is very valuable for subsequent radar antenna calibration and optimization, because it can help the AI ​​correction system locate the problem more accurately and optimize and improve it more targetedly.

[0099] It can be seen that the performance attention index is a very important concept, which provides the AI ​​correction system with a new way to quantify and evaluate the performance of radar antennas. By calculating the feature weights of each first antenna performance feature unit, the AI ​​correction system can have a more comprehensive understanding of the performance characteristics of the antenna and more accurately evaluate the importance of these characteristics in the overall performance. This provides strong support for subsequent radar antenna calibration and optimization.

[0100] In some alternative embodiments, determining a target proofreading feature identification result based on the multidimensional state attribute vector set includes: performing a feature embedding operation on the multidimensional state attribute vector set to obtain the target proofreading feature identification result, wherein for the X-band performance characteristics of each performance type, the target proofreading feature identification result includes at least one performance attention indicator, and the feature weight indication of the second antenna performance characteristic unit of each performance attention indicator corresponds to the first antenna performance characteristic unit of the second antenna performance characteristic unit as a performance correction reference under the set correction rule matched by the X-band performance characteristics of the performance type.

[0101] Based on this embodiment, an innovative technical solution based on a multi-dimensional state attribute vector set is proposed for the performance calibration and feature recognition of radar antennas. The core of this solution is that by performing feature embedding operations on the multi-dimensional state attribute vector set, the AI ​​correction system can obtain an information-rich and compact target calibration feature recognition result. This result not only fully reflects the performance characteristics of the radar antenna, but also provides the AI ​​correction system with an in-depth understanding and guidance of antenna performance calibration through an innovative way: performance attention indicators.

[0102] First, the AI ​​correction system needs to perform feature embedding operations on the multi-dimensional state attribute vector set. This step is the basis of the entire technical solution. Its purpose is to convert the original high-dimensional state attribute vector that may contain redundancy and noise into a more expressive and discriminative feature vector. Each element in these feature vectors can be regarded as a quantitative representation of a specific aspect of antenna performance. Through feature embedding, the AI ​​correction system can better capture and describe the performance characteristics of the antenna, providing strong support for subsequent performance proofreading and feature recognition.

[0103] After completing the feature embedding operation, the AI ​​correction system obtains a feature vector set rich in antenna performance information. However, this feature vector set is still relatively abstract and difficult to interpret directly. In order to more intuitively understand and evaluate the performance of the antenna, the AI ​​correction system needs to further build performance attention indicators.

[0104] The performance attention index is a key innovation in the technical solution of the AI ​​correction system. It is actually a quantitative measurement used to express the confidence of the performance correction reference under the set correction rule that the first antenna performance characteristic unit corresponding to the second antenna performance characteristic unit matches the X-band performance characteristics of the performance type under a specific performance category. The "confidence" in the embodiment of the present application reflects the degree of trust the AI ​​correction system has in the performance correction reference, while the "correction rules" are a series of correction standards and strategies formulated by the AI ​​correction system based on the performance characteristics and actual needs of the antenna.

[0105] Furthermore, the AI ​​correction system will calculate the characteristic weight of each second antenna performance characteristic unit for each performance category of the X-band performance characteristics. This characteristic weight actually quantifies the importance of the second antenna performance characteristic unit in describing the antenna performance and its confidence as a performance correction reference.

[0106] When calculating feature weights, the AI ​​correction system will take into account multiple factors, including but not limited to the antenna's radiation intensity, pointing accuracy, phase relationship, etc. These factors are key factors that affect antenna performance, and are also the focus of the AI ​​correction system when performing performance correction. By calculating the feature weights of each second antenna performance feature unit, the AI ​​correction system can obtain a comprehensive and quantitative performance evaluation result. This result not only tells the AI ​​correction system which performance the antenna performs well or poorly, but also tells the AI ​​correction system the importance of these performances in the overall evaluation and their confidence as a reference for performance correction.

[0107] Finally, when the AI ​​correction system has these feature weights, the AI ​​correction system can construct a performance attention index. Specifically, the AI ​​correction system will map the feature weight of each second antenna performance feature unit to the corresponding first antenna performance feature unit, thereby obtaining the feature weight of the first antenna performance feature unit. This feature weight is actually part of the performance attention index, which represents the confidence of the performance correction reference under the set correction rule that matches the first antenna performance feature unit on the X-band performance characteristics of the corresponding performance type.

[0108] Through this process, the AI ​​correction system can obtain a target calibration feature recognition result that includes multiple performance attention indicators. This result not only fully reflects the performance characteristics of the antenna, but also provides the AI ​​correction system with an in-depth understanding and guidance of antenna performance correction through performance attention indicators. The AI ​​correction system can formulate targeted correction strategies and optimization plans based on these indicators, thereby more effectively improving the performance and stability of the antenna.

[0109] In this way, through the feature embedding operation and the construction of performance attention indicators, a new radar antenna performance calibration and feature recognition method is provided for the AI ​​correction system. It not only improves the accuracy and comprehensiveness of performance evaluation, but also provides new ideas and directions for better understanding and improving the performance of radar antennas.

[0110] In another alternative embodiment, determining the X-band performance characteristic and the expected characteristic for horizontal correction based on the target performance characteristic identification result and the target proofreading characteristic identification result includes: determining a discrimination score that the first antenna performance characteristic unit belongs to the X-band performance characteristic based on the characteristic weight of each first antenna performance characteristic unit in the target performance characteristic identification result; determining the first antenna performance characteristic unit having a discrimination score greater than a set discrimination score as the X-band performance characteristic; and determining the expected characteristic for horizontal correction corresponding to the X-band performance characteristic in the target proofreading characteristic identification result based on the distribution characteristics of the X-band performance characteristic in the target performance characteristic identification result.

[0111] It can be understood that the task faced by this embodiment is to determine the X-band performance characteristics and the corresponding horizontal correction desired characteristics based on the target performance characteristic recognition results and the target calibration characteristic recognition results. The core of this task is to accurately identify which ones belong to the X-band from a large number of performance characteristics, and further find out the horizontal correction desired characteristics corresponding to these X-band performance characteristics, so as to provide guidance for subsequent radar antenna performance optimization and correction.

[0112] First, the AI ​​correction system starts with the target performance feature identification result. This target performance feature identification result is a complex data set containing multiple first antenna performance feature units and their feature weights. Each first antenna performance feature unit represents the performance characteristics of the antenna in a certain aspect, and the feature weight quantifies the importance of this feature in the overall performance evaluation. In order to determine which first antenna performance feature units belong to the X-band performance characteristics, the AI ​​correction system needs to calculate the discrimination score of each unit.

[0113] The calculation of the discrimination score is a process that takes into account both feature weights and performance category information. Specifically, the AI ​​correction system calculates a score that reflects the likelihood that the unit belongs to the X-band performance feature based on the feature weights of each first antenna performance feature unit and the distribution of these weights in the overall data set. The higher the score, the more likely the unit is to be part of the X-band performance feature.

[0114] After completing the calculation of the discrimination score, the AI ​​correction system can determine which first antenna performance characteristic units truly belong to the X-band performance characteristics based on the set discrimination score threshold. Specifically, the AI ​​correction system will select the first antenna performance characteristic units with a discrimination score greater than the set threshold as representatives of the X-band performance characteristics. These selected units are the performance characteristics that the AI ​​correction system needs to focus on in subsequent analysis.

[0115] However, it is not enough to simply determine the X-band performance characteristics. The AI ​​correction system also needs to find the expected horizontal correction characteristics corresponding to these characteristics. This requires the AI ​​correction system to turn to the target calibration feature recognition result. This target calibration feature recognition result is also a data set containing multiple performance feature units and their feature weights, but it focuses more on information on antenna performance correction.

[0116] In the target calibration feature recognition results, each performance feature unit has one or more corresponding correction rules. These correction rules are based on the actual performance characteristics and correction requirements of the antenna, and they describe how to correct the performance of the antenna to achieve the desired level. The task of the AI ​​correction system is to find the horizontal correction expected features corresponding to the X-band performance characteristics in these correction rules.

[0117] To accomplish this task, the AI ​​correction system first needs to analyze the distribution characteristics of the X-band performance characteristics in the target performance characteristics identification results. This includes understanding the proportion of these characteristics in the overall data set, the correlation between them, and their importance in antenna performance evaluation. Through these analyses, the AI ​​correction system can have a more comprehensive and in-depth understanding of the X-band performance characteristics.

[0118] The AI ​​correction system then matches this information with the correction rules in the target calibration feature recognition results. Specifically, the AI ​​correction system looks for correction rules that are closely related to the X-band performance characteristics and can effectively correct them. The performance characteristics described by these correction rules are the expected characteristics of horizontal correction that the AI ​​correction system is looking for. They represent the ideal state or expected level that the AI ​​correction system hopes to achieve when performing performance correction on the antenna.

[0119] Through this process, the AI ​​correction system can accurately determine the X-band performance characteristics and the corresponding horizontal correction expected characteristics based on the target performance feature identification results and the target calibration feature identification results. This result not only provides the AI ​​correction system with important information about antenna performance, but also provides strong guidance for subsequent radar antenna performance optimization and correction work. The AI ​​correction system can formulate targeted correction strategies and optimization plans based on this information, thereby more effectively improving the performance and stability of the antenna. At the same time, the results can also serve as an important reference in the antenna design and manufacturing process, helping to better understand and improve the performance characteristics of radar antennas.

[0120] Under some preferred technical ideas, determining the distribution information and deviation information of the data to be adjusted in the sensor measurement state data based on the X-band performance characteristics and the expected characteristics of the horizontal correction includes: determining the feature association description between the multidimensional state attribute vector set and the sensor measurement state data; determining the distribution information and initial noise information of the data to be adjusted in the sensor measurement state data based on the X-band performance characteristics, the expected characteristics of the horizontal correction and the feature association description; determining the noise disturbance for the initial noise information based on the multidimensional state attribute vector set; and determining the deviation information of the data to be adjusted in the sensor measurement state data based on the noise disturbance and the initial noise information.

[0121] It can be understood that the task faced by this technical idea is to accurately determine the distribution information and deviation information of the data to be adjusted in the sensor measurement state data. Accurately determining the distribution information and deviation information of the data to be adjusted in the sensor measurement state data requires the guidance of the X-band performance characteristics and the expected characteristics of the horizontal correction, and also requires making full use of the inherent connection between the multidimensional state attribute vector set and the sensor measurement state data.

[0122] First, the AI ​​correction system needs to determine the characteristic association description between the multidimensional state attribute vector set and the sensor measurement state data. This is a key prerequisite step because it provides a basic framework for the subsequent analysis of the AI ​​correction system. The multidimensional state attribute vector set is a data set containing multiple state attributes, each of which describes the state or characteristics of the radar antenna or related system in a certain aspect. The sensor measurement state data is the data obtained by actual measurement, which reflects the actual performance of the antenna or system under specific conditions. By deeply analyzing these two sets of data, the AI ​​correction system can discover the inherent connection and law between them, which is the characteristic association description.

[0123] With the feature association description, the AI ​​correction system can further determine the distribution information and initial noise information of the data to be adjusted in the sensor measurement state data based on the X-band performance characteristics, the expected characteristics of horizontal correction, and the feature association description. The X-band performance characteristics are the key performance characteristics that the AI ​​correction system focuses on, and they represent the key performance of the antenna in the X-band. The expected horizontal correction characteristics are the target state that the AI ​​correction system expects to achieve when correcting the antenna performance. By combining these two characteristics and feature association descriptions, the AI ​​correction system can locate those data to be adjusted that are related to the X-band performance characteristics and the expected horizontal correction characteristics in the sensor measurement state data, and understand their distribution and initial noise level.

[0124] Next, the AI ​​correction system needs to determine the noise disturbance used for the initial noise information based on the multidimensional state attribute vector set. Noise disturbance is an important concept that represents the random fluctuations in the sensor measurement state data due to various factors (such as measurement errors, environmental interference, etc.). By utilizing the information in the multidimensional state attribute vector set, the AI ​​correction system can make more accurate estimates and modeling of noise disturbances. Specifically, the AI ​​correction system can analyze the relationship between the various attributes in the multidimensional state attribute vector set and the noise in the sensor measurement state data, so as to find out those attributes that have a significant impact on the noise, and use this information to build a model of noise disturbance.

[0125] With the model of noise disturbance, the AI ​​correction system can determine the deviation information of the data to be adjusted in the sensor measurement state data based on the noise disturbance and the initial noise information. The deviation information reflects the degree of difference between the data to be adjusted and its true value or expected value. By combining the noise disturbance and the initial noise information, the AI ​​correction system can more accurately evaluate the deviation of the data to be adjusted. Specifically, the AI ​​correction system can use the model of noise disturbance to simulate the random fluctuations in the sensor measurement state data, and compare this fluctuation with the initial noise information to obtain the deviation information of the data to be adjusted.

[0126] In this way, by deeply analyzing the intrinsic relationship between the multidimensional state attribute vector set and the sensor measurement state data, and making full use of the guidance of the X-band performance characteristics and the expected characteristics of the horizontal correction, it is possible to accurately determine the distribution information and deviation information of the data to be adjusted in the sensor measurement state data. This result not only provides the AI ​​correction system with important information about the antenna performance, but also provides strong guidance for the subsequent radar antenna performance optimization and correction work. The AI ​​correction system can formulate targeted correction strategies and optimization plans based on this information, thereby more effectively improving the performance and stability of the antenna.

[0127] In some other examples, the X-band performance characteristics and the horizontal correction desired characteristics include X-band performance characteristics of the data to be adjusted for at least one time series tag and horizontal correction desired characteristics of the data to be adjusted for the performance category.

[0128] In practical applications, it is crucial to process and analyze X-band performance characteristics and expected horizontal correction characteristics in a more specific and in-depth manner. These characteristics are not just abstract descriptions or summaries, but are specific to the data to be adjusted for each time series tag, providing more refined guidance for radar antenna performance optimization and correction.

[0129] First of all, the X-band performance characteristics refer to the performance of the radar antenna in the X-band. The operating frequency band of the radar antenna is usually divided into multiple bands, and the X-band is an important part of it. The X-band performance characteristics are used to describe and quantify the various performance indicators of the antenna in the X-band, such as gain, directivity, beam width, etc. These performance indicators are directly related to the detection capability, positioning accuracy and anti-interference ability of the radar system, so they are the focus of attention in the antenna design and optimization process.

[0130] The expected horizontal correction feature is the target state that the AI ​​correction system expects to achieve when correcting the performance of the radar antenna. In practical applications, due to various factors, the performance of the radar antenna often deviates from its ideal state. In order to improve the performance of the antenna, the AI ​​correction system needs to correct it so that it is as close as possible to or reaches the performance level expected by the AI ​​correction system. The expected horizontal correction feature is used to describe this expected state, which provides the AI ​​correction system with the direction and target of correction.

[0131] Accordingly, the X-band performance characteristics and expected characteristics of horizontal correction are further refined to the data to be adjusted for each timing tag. Timing tag is an important concept that represents the time sequence or time series of data. During the operation of the radar system, the performance data of the antenna is continuously generated. These data are recorded in chronological order to form time series data. Each timing tag corresponds to a specific set of performance data, which reflects the performance status of the antenna at a certain moment or over a period of time.

[0132] Therefore, when it comes to X-band performance characteristics and horizontal correction expected characteristics including data to be adjusted for at least one timing tag, the AI ​​correction system is actually saying that these characteristics are not just a general description of antenna performance, but are specific to each timing tag, providing corresponding characteristics and expectations for performance data at each moment or within each period of time.

[0133] This design first improves the AI ​​correction system's ability to understand and analyze antenna performance. By viewing the performance characteristics and expected characteristics on each timing label, the AI ​​correction system can more clearly understand the performance of the antenna at different times or under different conditions, and how the AI ​​correction system needs to correct and optimize it. Secondly, it provides a more refined correction and optimization strategy for the AI ​​correction system. Since the data on each timing label has its own specific characteristics and expectations, the AI ​​correction system can formulate more targeted correction strategies and optimization plans based on this data, thereby more effectively improving the performance of the antenna.

[0134] Specifically, when the AI ​​correction system has the X-band performance characteristics and the expected characteristics of horizontal correction for the data to be adjusted for at least one timing tag, the AI ​​correction system can perform the following operations: First, the AI ​​correction system can extract and analyze the performance data on each timing tag. This includes calculating various performance indicators, identifying performance patterns, detecting performance anomalies, etc. Through these operations, the AI ​​correction system can gain an in-depth understanding of the performance of the antenna at different times or under different conditions, and provide basic data for subsequent correction and optimization work. Secondly, the AI ​​correction system can formulate a correction strategy based on the expected characteristics of horizontal correction. For each data on a timing tag, the AI ​​correction system can determine the target and direction of correction based on its expected characteristics. For example, if the expected feature is to increase the gain, the AI ​​correction system can formulate a corresponding strategy to adjust the parameters or structure of the antenna to improve its gain performance in the X-band. Then, the AI ​​correction system can implement the correction strategy and verify the results. After implementing the correction strategy, the AI ​​correction system needs to measure and evaluate the performance of the antenna again to verify the effectiveness of the correction strategy. If the correction result is as expected, the AI ​​correction system can apply the correction strategy to the data on other timing tags; if the correction result is not as expected, the AI ​​correction system needs to adjust or reformulate the strategy. Finally, the AI ​​correction system can integrate and analyze the corrected data to evaluate the effectiveness of the entire correction process. By comparing the performance data before and after correction, the AI ​​correction system can understand the extent to which the correction strategy improves antenna performance and provide guidance for subsequent optimization work.

[0135] It can be seen that the X-band performance characteristics and expected characteristics of horizontal correction are further refined to the data to be adjusted for each time series label. This processing method improves the AI ​​correction system's ability to understand and analyze antenna performance, and provides the AI ​​correction system with a more refined correction and optimization strategy. Through a series of operations such as feature extraction, strategy formulation, implementation verification, and effect evaluation, the AI ​​correction system can more effectively improve the performance of radar antennas and provide strong guarantees for their performance in practical applications.

[0136] In some further optional examples, the sensor measurement status data is radar antenna array status data collected by a built-in measurement device. Before determining a multidimensional state attribute vector set of the sensor measurement status data, the method further includes: adjusting the original characteristic value of each antenna performance characteristic unit in the sensor measurement status data.

[0137] Based on this embodiment, especially when these data come from radar antenna array status data collected by built-in measurement equipment, data processing is not limited to simple collection and organization, but involves more complex and sophisticated feature adjustment and construction of a multi-dimensional state attribute vector set.

[0138] First, the sensor measurement status data is directly collected by the built-in measurement equipment, which reflects the various states of the radar antenna array during actual operation. These data are raw and unprocessed, so they may contain various noises, errors, and redundant information. In order to extract useful information from these raw data, the AI ​​correction system needs to perform a series of processing and analysis.

[0139] Before determining the multidimensional state attribute vector set of the sensor measurement state data, an important step is to adjust the original eigenvalue of each antenna performance characteristic unit in the sensor measurement state data. This step is crucial because it directly affects the accuracy and effectiveness of the subsequent multidimensional state attribute vector set.

[0140] The original eigenvalues ​​are the most basic elements in the sensor measurement state data. They represent the original measured values ​​or calculated values ​​of the antenna performance characteristic unit in a certain aspect. However, these original eigenvalues ​​are often not the state attributes that the AI ​​correction system ultimately needs, because they may be affected by various factors and deviate from the true value. Therefore, the AI ​​correction system needs to adjust these original eigenvalues ​​to eliminate or reduce the impact of these factors.

[0141] The process of adjusting the original eigenvalues ​​can be a complex and delicate operation. It may involve data filtering, denoising, correction, normalization and other aspects. Filtering is to eliminate random noise and interference in the data to make the data smoother and more stable; denoising is to remove outliers and erroneous values ​​in the data to avoid their influence on subsequent analysis; correction is to correct the deviation and distortion in the data to make it closer to the true value; normalization is to convert the data to a unified scale for subsequent processing and analysis.

[0142] In the process of adjusting the original eigenvalues, the AI ​​correction system also needs to make full use of prior knowledge and empirical data. Prior knowledge refers to the AI ​​correction system's understanding and knowledge of the performance characteristics of the radar antenna, including the changing patterns, influencing factors, and correlations of the performance characteristics. Empirical data refers to the data and experience accumulated by the AI ​​correction system when processing and analyzing similar antenna performance characteristics in the past. By combining prior knowledge and empirical data, the AI ​​correction system can more accurately adjust the original eigenvalues ​​to make them more consistent with the actual antenna performance characteristics.

[0143] After completing the adjustment of the original eigenvalues, the AI ​​correction system can further construct a multidimensional state attribute vector set of the sensor measurement state data. A multidimensional state attribute vector set is a data set containing multiple state attributes, each of which describes the state or characteristics of the radar antenna or related system in a certain aspect. By constructing a multidimensional state attribute vector set, the AI ​​correction system can have a more comprehensive and in-depth understanding of the state and performance characteristics of the radar antenna array.

[0144] When constructing a multidimensional state attribute vector set, the AI ​​correction system needs to select appropriate state attributes based on actual needs and application scenarios. These state attributes should be able to fully reflect the state and performance characteristics of the radar antenna array, while being measurable and analyzable. For example, the AI ​​correction system can select the antenna's gain, directivity, beam width, standing wave ratio, etc. as state attributes, because they are all important indicators for describing antenna performance.

[0145] At the same time, the AI ​​correction system also needs to quantify and standardize the selected state attributes. Quantification refers to converting the original value of the state attribute into a numerical form for subsequent calculation and analysis; standardization is to eliminate the dimension and scale differences between different state attributes to make them comparable and additive. Through quantification and standardization, the AI ​​correction system can convert the selected state attributes into elements in a multidimensional state attribute vector set and provide a basis for its subsequent analysis and application.

[0146] It can be seen that when the sensor measurement state data is the radar antenna array state data collected by the built-in measurement device, the AI ​​correction system needs to adjust the original eigenvalues ​​of each antenna performance characteristic unit in the sensor measurement state data before determining its multidimensional state attribute vector set. This step is crucial and directly affects the accuracy and effectiveness of the subsequent multidimensional state attribute vector set. By adjusting the original eigenvalues, constructing a multidimensional state attribute vector set, and selecting appropriate state attributes for quantification and standardization, the AI ​​correction system can have a more comprehensive and in-depth understanding of the state and performance characteristics of the radar antenna array, and provide a solid foundation for subsequent analysis and application.

[0147] In some preferred embodiments, determining the multidimensional state attribute vector set of the sensor measurement state data includes: processing the sensor measurement state data through at least one feature embedding branch to obtain sensor measurement state data that has completed feature embedding; performing at least one feature pooling and at least one feature derivation on the sensor measurement state data that has completed feature embedding through at least one feature pooling branch and at least one feature derivation branch to determine the multidimensional state attribute vector set of the sensor measurement state data, wherein the feature granularity of the multidimensional state attribute vector set is smaller than that of the sensor measurement state data.

[0148] This embodiment focuses on how to determine the multidimensional state attribute vector set of sensor measurement state data. This process not only involves data processing and analysis, but also covers multiple key links such as feature embedding, feature pooling and feature derivation.

[0149] First of all, to determine the multi-dimensional state attribute vector set of the sensor measurement state data, the AI ​​correction system needs to effectively process and analyze these raw data. Feature embedding is the first step in this process. The purpose of feature embedding is to convert the high-dimensional and complex information in the sensor measurement state data into a more compact, low-dimensional and easy-to-process feature representation. This step is usually implemented through a feature embedding branch, which can perform deep feature extraction and conversion on the original data to obtain sensor measurement state data with complete feature embedding. These data processed by feature embedding not only retain the key information in the original data, but also facilitate subsequent feature pooling and feature derivation operations.

[0150] Secondly, after completing feature embedding, the AI ​​correction system needs to further perform feature pooling and feature derivation processing on the obtained sensor measurement state data. Feature pooling is an effective means of dimensionality reduction. It can reduce the dimension of data by merging or aggregating similar features while retaining key information. This step is usually completed by the feature pooling branch, which can further compress and refine the data after feature embedding to obtain a more compact and informative feature representation. Feature derivation generates new features based on existing features through certain mathematical transformations or combinations. These new features can often reveal hidden information or patterns in the data. Feature derivation is usually implemented by the feature derivation branch, which can further mine and explore the pooled data to obtain richer and more diverse feature representations.

[0151] Through feature pooling and feature derivation, the AI ​​correction system can obtain a series of feature representations with different granularities and levels, which together constitute a multidimensional state attribute vector set of the sensor measurement state data. It is worth noting that the feature granularity of this multidimensional state attribute vector set is smaller than the original sensor measurement state data. This means that the AI ​​correction system successfully converts high-dimensional and complex sensor measurement state data into a low-dimensional, compact and informative multidimensional state attribute vector set through a series of operations such as feature embedding, feature pooling and feature derivation.

[0152] It can be understood that the advantage of this technical solution is that it can effectively process and analyze sensor measurement state data, extract key feature information from it, and construct a multidimensional state attribute vector set with rich representation capabilities. This not only improves the data processing efficiency and analysis accuracy, but also provides strong data support for subsequent machine learning, pattern recognition and other tasks. At the same time, through the organic combination of feature embedding, feature pooling and feature derivation, the AI ​​correction system can more flexibly adjust and optimize the construction process of the multidimensional state attribute vector set to adapt to different application scenarios and actual needs.

[0153] In the specific implementation process, the AI ​​correction system can select appropriate feature embedding branches, feature pooling branches, and feature derivation branches according to the characteristics and analysis requirements of the sensor measurement state data. For example, for sensor measurement state data containing time series data, the AI ​​correction system can select a feature embedding branch based on a recurrent neural network to extract dynamic features in the time series; for sensor measurement state data containing image data, the AI ​​correction system can select a feature embedding branch based on a convolutional neural network to extract spatial features in the image. Similarly, the AI ​​correction system can select appropriate feature pooling methods and feature derivation strategies according to actual needs to construct a multi-dimensional state attribute vector set that best suits the current application scenario.

[0154] Designed in this way, the AI ​​correction system determines the multidimensional state attribute vector set of the sensor measurement state data through a series of operations such as feature embedding, feature pooling, and feature derivation. This technical solution not only improves the efficiency and accuracy of data processing and analysis, but also provides strong data support for subsequent machine learning, pattern recognition and other tasks. By flexibly selecting and combining different feature embedding branches, feature pooling branches, and feature derivation branches, the AI ​​correction system can adapt to different application scenarios and actual needs, and construct a multidimensional state attribute vector set with rich representation capabilities.

[0155] In some further preferred embodiments, the multidimensional state attribute vector set, the target performance feature identification result, and the target proofreading feature identification result are generated by at least one long short-term memory model, and the at least one long short-term memory model is debugged by the following steps: determining a debugging sample set, wherein the debugging sample set includes at least one sensor measurement state data sample, and the at least one sensor measurement state data sample carries prior distribution information and prior deviation information of prior data to be adjusted; for each sensor measurement state data sample: determining a multidimensional state attribute vector set sample of the sensor measurement state data sample; determining a target performance feature identification result sample based on the multidimensional state attribute vector set sample; determining a target proofreading feature identification result sample based on the multidimensional state attribute vector set sample; determining a target proofreading feature identification result sample based on the target performance feature identification The method comprises the steps of: determining an X-band performance feature sample and a horizontal correction expected feature sample based on the identification result sample and the target calibration feature identification result sample; determining a distribution information sample and a deviation information sample of the data sample to be adjusted in the sensor measurement state data sample according to the X-band performance feature sample and the horizontal correction expected feature sample; optimizing the model weights of the at least one long short-term memory model so that the difference between the distribution information sample and the deviation information sample of the data sample to be adjusted in the sensor measurement state data sample and the prior distribution information and prior deviation information of the prior data to be adjusted converges; exemplarily, the difference includes at least one of the following comparison errors: a distribution comparison error between the distribution information sample and the prior distribution information; and a deviation comparison error between the deviation information sample and the prior deviation information.

[0156] In actual application, the generation of multidimensional state attribute vector sets, target performance feature recognition results, and target calibration feature recognition results are all achieved through at least one long short-term memory model. In order to ensure the effectiveness and accuracy of this model, a series of debugging was carried out.

[0157] First, you need to build a debugging sample set, which is the cornerstone of model debugging. This debugging sample set is not a random combination, but carefully selected, and contains at least one sensor measurement state data sample. Each sensor measurement state data sample is like a treasure, containing rich prior distribution information and prior deviation information of the prior data to be adjusted. This information is crucial for the subsequent debugging and optimization of the model, and it can guide the AI ​​correction system to find the right direction for model debugging.

[0158] Next, the AI ​​correction system conducts in-depth analysis and processing for each sensor measurement state data sample. The AI ​​correction system must first determine the multidimensional state attribute vector set sample of the sensor measurement state data sample. This step is to extract the key features in the data and lay the foundation for subsequent performance feature identification and proofreading feature identification. With the multidimensional state attribute vector set samples, the AI ​​correction system can further determine the target performance feature identification result samples and the target proofreading feature identification result samples. These two result samples are important references for the AI ​​correction system debugging model, and they will help the AI ​​correction system evaluate the recognition and proofreading capabilities of the model.

[0159] However, the debugging process of the AI ​​correction system does not stop there. The AI ​​correction system also needs to further determine the X-band performance characteristic samples and the horizontal correction expected characteristic samples based on the target performance characteristic recognition result samples and the target calibration characteristic recognition result samples. These two samples are the ultimate goals of the AI ​​correction system debugging model. They will guide the AI ​​correction system to optimize the model weights so that the model can more accurately identify and calibrate the key features in the sensor measurement state data.

[0160] Next, to optimize the model weights, the AI ​​correction system needs to determine the distribution information samples and deviation information samples of the data samples to be adjusted based on the performance of the X-band performance characteristic samples and the horizontal correction expected characteristic samples in the sensor measurement state data samples. These two information samples are important indicators for the AI ​​correction system to evaluate the performance of the model, and they will help the AI ​​correction system discover the deficiencies of the model in the recognition and proofreading process.

[0161] Finally, the AI ​​correction system enters the stage of optimizing the model weights. The goal of the AI ​​correction system is to converge the differences between the distribution information samples and deviation information samples of the data samples to be adjusted in the sensor measurement state data samples and the prior distribution information and prior deviation information of the prior data to be adjusted. In other words, the AI ​​correction system should make the performance of the model in the recognition and proofreading process closer to the expectations of the AI ​​correction system, making the output of the model more accurate and reliable. To achieve this goal, the AI ​​correction system will continuously optimize the weights of the model until the requirements of the AI ​​correction system are met.

[0162] During the optimization process, the AI ​​correction system will pay special attention to the distribution comparison error between the distribution information sample and the prior distribution information, as well as the deviation comparison error between the deviation information sample and the prior deviation information. These two comparison errors are key indicators for the AI ​​correction system to evaluate model performance. They will help the AI ​​correction system discover problems in the model's recognition and proofreading process, and guide the AI ​​correction system to perform targeted optimization and improvement.

[0163] It can be seen that the AI ​​correction system generates a multidimensional state attribute vector set, target performance feature recognition results, and target proofreading feature recognition results through at least one long short-term memory model. In order to ensure the effectiveness and accuracy of the model, the AI ​​correction system has carried out a series of debugging. The AI ​​correction system constructs a debugging sample set, extracts samples of multidimensional state attribute vector sets, determines samples of target performance feature recognition results and target proofreading feature recognition results, and further determines X-band performance feature samples and horizontal correction expected feature samples. Then, the AI ​​correction system optimizes the weights of the model based on these samples, so that the performance of the model in the recognition and proofreading process is closer to the expectations of the AI ​​correction system. This series of debugging and optimization steps ensures that the model of the AI ​​correction system can perform at its best in practical applications.

[0164] In addition, in combination with the foregoing, for some independently implementable technical solutions, the calibrating the target radar antenna array based on the distribution information and the deviation information includes: obtaining a global impact information set of the target radar antenna array based on the distribution information and the deviation information, wherein the global impact information set includes Q uninterrupted global impact information, and Q is an integer greater than or equal to 1; obtaining a mechanical electronic impact information set according to the global impact information set, wherein the mechanical electronic impact information set includes Q uninterrupted mechanical electronic impact information; based on the global impact information set, obtaining a global impact factor relationship vector network through a first relationship vector identification sub-algorithm included in a calibration control algorithm A set of global influencing factor relationship vector nets, wherein the global influencing factor relationship vector nets include Q global influencing factor relationship vector nets; based on the mechanical and electronic influence information set, the mechanical and electronic influence factor relationship vector net set is obtained through the second relationship vector identification sub-algorithm included in the calibration control algorithm, wherein the mechanical and electronic influence factor relationship vector net set includes Q mechanical and electronic influence factor relationship vector nets; based on the global influencing factor relationship vector net set and the mechanical and electronic influence factor relationship vector net set, the radar antenna array calibration label corresponding to the global influence information set is obtained through the calibration execution matching sub-algorithm included in the calibration control algorithm; and the target radar antenna array is dynamically calibrated according to the radar antenna array calibration label.

[0165] It is understandable that the process of calibrating the target radar antenna array must not only consider the physical characteristics of the radar antenna array itself, but also fully consider the various influencing factors in the actual operating environment.

[0166] First, it is necessary to obtain the global impact information set of the target radar antenna array based on the distribution information and the deviation information. This step is the basis of the calibration process, because the global impact information set contains the various external and internal factors that may affect the radar antenna array during operation, and these factors together constitute the global operating environment of the radar antenna array. The global impact information set includes Q continuous global impact information, where Q is an integer greater than or equal to 1, which means that the influencing factors considered by the AI ​​correction system are continuous and comprehensive, without missing any factors that may affect the performance of the radar antenna array.

[0167] Next, the AI ​​correction system obtains the mechanical and electronic impact information set based on the global impact information set. The mechanical and electronic impact information set is an important subset of the global impact information set, which focuses on those influencing factors directly related to the mechanical and electronic components of the radar antenna array. These influencing factors include but are not limited to wear, aging, electromagnetic interference, etc. of components, which have a direct and significant impact on the performance of the radar antenna array. The mechanical and electronic impact information set also includes Q uninterrupted mechanical and electronic impact information to ensure that the AI ​​correction system has a comprehensive understanding and grasp of every mechanical and electronic factor that may affect the performance of the radar antenna array.

[0168] With the global impact information set and the mechanical and electronic impact information set, the AI ​​correction system can further obtain the global impact factor relationship vector network set and the mechanical and electronic impact factor relationship vector network set through the calibration control algorithm. These two networks are the key to the calibration process. They describe and represent the interactions and relationships between various impact factors through complex relationship vector networks. The global impact factor relationship vector network set includes Q global impact factor relationship vector networks, each of which corresponds to a global impact information, which is used to reveal the intrinsic connection between the impact information and other impact information. Similarly, the mechanical and electronic impact factor relationship vector network set also includes Q mechanical and electronic impact factor relationship vector networks, which are used to reveal the interactions and relationships between mechanical and electronic impact information.

[0169] After obtaining the global influencing factor relationship vector network set and the mechanical and electronic influencing factor relationship vector network set, the AI ​​correction system can obtain the radar antenna array calibration label corresponding to the global influencing information set through the calibration execution matching sub-algorithm included in the calibration control algorithm. This calibration label is the direct output of the calibration process. It contains all the necessary calibration information and instructions to guide subsequent calibration operations. The acquisition of the calibration label is based on the in-depth analysis and comprehensive judgment of the global influencing factor relationship vector network set and the mechanical and electronic influencing factor relationship vector network set, ensuring the accuracy and effectiveness of the calibration label.

[0170] Finally, the AI ​​correction system can dynamically calibrate the target radar antenna array according to the radar antenna array calibration tag. Dynamic calibration is the final link in the calibration process and one of the most important links. It requires the AI ​​correction system to perform real-time and dynamic calibration operations on the radar antenna array in the actual operating environment to ensure that the radar antenna array always remains in the best working state. The dynamic calibration process is complex and delicate. It requires the AI ​​correction system to make full use of the rich information and data such as the global impact information set, mechanical and electronic impact information set, global impact factor relationship vector network set, and mechanical and electronic impact factor relationship vector network set obtained previously. Eliminate the adverse effects of various influencing factors on the performance of the radar antenna array through precise calibration operations and adjustments, thereby ensuring the accuracy and stability of the radar antenna array.

[0171] It can be seen that the process of calibrating the target radar antenna array is a comprehensive and complex process. It requires the AI ​​correction system to fully consider the various influencing factors of the radar antenna array in the actual operating environment, and to fully understand and master these influencing factors by obtaining the global influencing information set and the mechanical and electronic influencing information set; to deeply analyze and reveal the interactions and relationships between these influencing factors by obtaining the global influencing factor relationship vector network set and the mechanical and electronic influencing factor relationship vector network set; and finally, to ensure that the radar antenna array is always in the best working state by obtaining the radar antenna array calibration label and performing dynamic calibration.

[0172] In some other independent embodiments, the calibration execution matching sub-algorithm included in the calibration control algorithm based on the global influencing factor relationship vector network set and the mechanical and electronic influencing factor relationship vector network set obtains the radar antenna array calibration label corresponding to the global influencing information set, including: based on the global influencing factor relationship vector network set, obtaining Q first space-time signal feature arrays through the space-time feature extraction module included in the calibration control algorithm, wherein each first space-time signal feature array corresponds to a global influencing factor relationship vector network; based on the mechanical and electronic influencing factor relationship vector network set, obtaining Q first space-time signal feature arrays through the space-time feature extraction module included in the calibration control algorithm The second spatial attention network obtains Q second space-time signal feature arrays, wherein each second space-time signal feature array corresponds to a mechanical and electronic influencing factor relationship vector network; the Q first space-time signal feature arrays and the Q second space-time signal feature arrays are integrated to obtain Q target space-time signal feature arrays, wherein each target space-time signal feature array includes a first space-time signal feature array and a second space-time signal feature array; based on the Q target space-time signal feature arrays, the radar antenna array calibration label corresponding to the global impact information set is obtained through the calibration execution matching sub-algorithm included in the calibration control algorithm.

[0173] It can be understood that this embodiment involves how to obtain the radar antenna array calibration label corresponding to the global impact information set through the calibration execution matching sub-algorithm in the calibration control algorithm based on the global impact factor relationship vector mesh set and the mechanical and electronic impact factor relationship vector mesh set. This process involves the integration of multiple key steps and technologies, aiming to ensure the calibration accuracy and efficiency of the radar antenna array.

[0174] First, the AI ​​correction system uses the space-time feature extraction module in the calibration control algorithm to obtain Q first space-time signal feature arrays based on the global influencing factor relationship vector network set. The core of this step is that each global influencing factor relationship vector network contains rich space-time information, which is crucial for understanding the behavior pattern of the radar antenna array on a global scale. The space-time feature extraction module can deeply explore the space-time features in these relationship vector networks and convert them into the first space-time signal feature array, providing strong support for subsequent processing.

[0175] Next, the AI ​​correction system turns to the set of mechanical and electronic influencing factors’ relationship vector networks, and obtains Q second space-time signal feature arrays by calibrating the second spatial attention network in the control algorithm. Mechanical and electronic influencing factors are closely related to the physical characteristics and operating status of the radar antenna array, so the analysis of their relationship vector networks is equally important. The second spatial attention network focuses on capturing the spatial features in these relationship vector networks and combining them with the temporal features to form the second space-time signal feature array, providing more refined information for the calibration of the radar antenna array.

[0176] After obtaining Q first space-time signal feature arrays and Q second space-time signal feature arrays, the AI ​​correction system performs integrated processing to obtain Q target space-time signal feature arrays. The purpose of this step is to fuse the space-time characteristics of global influencing factors and mechanical and electronic influencing factors to form a more comprehensive and accurate space-time signal feature representation. Each target space-time signal feature array contains a first space-time signal feature array and a second space-time signal feature array, which together form the basis for radar antenna array calibration.

[0177] Finally, based on the Q target space-time signal feature arrays, the AI ​​correction system obtains the radar antenna array calibration label corresponding to the global impact information set through the calibration execution matching sub-algorithm in the calibration control algorithm. This step is the core of the entire calibration process, which requires the AI ​​correction system to convert the space-time signal features obtained in the previous steps into specific calibration instructions or parameters. The calibration execution matching sub-algorithm deeply analyzes the information in the target space-time signal feature array, matches it with the preset calibration standard, and finally generates a calibration label suitable for the radar antenna array. These calibration labels contain all the necessary calibration information and can guide the radar antenna array to perform precise calibration operations in actual operation.

[0178] In this way, through key steps such as spatiotemporal feature extraction, integrated processing, and calibration execution matching, a comprehensive and accurate calibration of the radar antenna array is achieved. This method not only improves the calibration efficiency of the radar antenna array, but also provides a strong guarantee for its stability and reliability in practical applications.

[0179] In some other independent embodiments, the method of obtaining Q first spatiotemporal signal feature arrays based on the global influencing factor relationship vector network set through the spatiotemporal feature extraction module included in the calibration control algorithm includes: for each global influencing factor relationship vector network in the global influencing factor relationship vector network set, obtaining a first encoded influencing factor relationship vector network through the encoding layer included in the spatiotemporal feature extraction module, wherein the spatiotemporal feature extraction module belongs to the calibration control algorithm; for each global influencing factor relationship vector network in the global influencing factor relationship vector network set, obtaining a first decoded influencing factor relationship vector network through the decoding layer included in the spatiotemporal feature extraction module; for each global influencing factor relationship vector network in the global influencing factor relationship vector network set, based on the first encoding influencing factor relationship vector network and the first decoding influencing factor relationship vector network, obtaining a first fused influencing factor relationship vector network through the feature fusion layer included in the spatiotemporal feature extraction module; for each global influencing factor relationship vector network in the global influencing factor relationship vector network set, based on the first fused influencing factor relationship vector network and the global influencing factor relationship vector network, obtaining a first spatiotemporal signal feature array through the first decoding layer included in the spatiotemporal feature extraction module.

[0180] In the actual application process, based on the global influencing factor relationship vector network set, the technical solution of obtaining Q first space-time signal feature arrays through the space-time feature extraction module in the calibration control algorithm involves the integration of multiple key steps and technologies, aiming to comprehensively and accurately extract the space-time features in the global influencing factor relationship vector network set, and provide strong support for the subsequent calibration of the radar antenna array.

[0181] First, the AI ​​correction system obtains the first coded influencing factor relationship vector network for each global influencing factor relationship vector network in the global influencing factor relationship vector network set through the coding layer in the spatiotemporal feature extraction module. The coding layer is an important component of the spatiotemporal feature extraction module. It can perform preliminary processing and conversion on the global influencing factor relationship vector network, encode the key information and features therein, and form the first coded influencing factor relationship vector network. This step provides the basis for subsequent decoding and feature fusion.

[0182] Secondly, the AI ​​correction system also obtains the first decoded influencing factor relationship vector network for each global influencing factor relationship vector network in the global influencing factor relationship vector network set through the decoding layer in the spatiotemporal feature extraction module. The decoding layer complements the encoding layer, and it can further decode and process the encoded influencing factor relationship vector network, restore the spatiotemporal features and information therein, and form the first decoded influencing factor relationship vector network. This step provides the necessary decoding information for subsequent feature fusion.

[0183] Then, the AI ​​correction system again targets each global influencing factor relationship vector network in the global influencing factor relationship vector network set, based on the first encoded influencing factor relationship vector network and the first decoded influencing factor relationship vector network, and obtains the first fused influencing factor relationship vector network through the feature fusion layer in the spatiotemporal feature extraction module. The feature fusion layer is the core part of the spatiotemporal feature extraction module. It can effectively fuse the encoded and decoded influencing factor relationship vector networks, extract the key spatiotemporal features and information therein, and form the first fused influencing factor relationship vector network. This step provides important feature support for obtaining the first spatiotemporal signal feature array.

[0184] Finally, the AI ​​correction system still targets each global influencing factor relationship vector network in the global influencing factor relationship vector network set, and based on the first fused influencing factor relationship vector network and the global influencing factor relationship vector network, obtains the first space-time signal feature array through the first decoding layer in the space-time feature extraction module. The first decoding layer plays a key role at this time. It can finally decode and process the fused influencing factor relationship vector network, extract the space-time signal features therein, and form the first space-time signal feature array. This step is the core output of the entire technical solution, and provides important space-time signal feature support for the subsequent calibration of the radar antenna array.

[0185] In this way, through the integration of multiple key steps and technologies such as the encoding layer, decoding layer, feature fusion layer and the first decoding layer, the comprehensive and accurate extraction of the spatiotemporal features of the global influencing factor relationship vector network is achieved. This not only improves the accuracy and efficiency of radar antenna array calibration, but also provides a strong technical guarantee for its stability and reliability in practical applications.

[0186] In summary, in an exemplary application scenario, the AI ​​correction system acquires the following sensor measurement status data in real time through the built-in sensors and measuring devices of the radar system: pointing angle of 352.6 degrees, signal strength of -45dBm, phase difference of 2.3 degrees, ambient temperature of 25 degrees Celsius, humidity of 40%RH, and mechanical vibration amplitude of 0.05mm. These data are organized into a multidimensional state attribute vector set, expressed as: [352.6, -45, 2.3, 25, 40, 0.05], which provides a basis for subsequent analysis and processing.

[0187] After analyzing the sensor measurement status data, the AI ​​correction system determined the following target performance feature identification results: the feature weight of signal transmission efficiency is 0.8, the feature weight of phase consistency is 0.7, and the feature weight of pointing accuracy is 0.9. These feature weights indicate their importance in the overall performance evaluation. For example, the feature weight of pointing accuracy is the highest, indicating that it has a crucial impact on the performance of X-band radar.

[0188] Next, the AI ​​correction system further analyzed the data and determined the following target correction feature recognition results: the feature weight of the correction decision feature corresponding to signal transmission efficiency is 0.6, the feature weight of the correction decision feature corresponding to phase consistency is 0.5, and the feature weight of the correction decision feature corresponding to pointing accuracy is 0.8. These feature weights reflect the effectiveness and applicability of the corresponding correction decision features.

[0189] After comprehensive evaluation, the AI ​​correction system determined that the key X-band performance characteristic was pointing accuracy, and determined that the desired characteristic of the required horizontal correction was adjusting the pointing angle to reduce deviation.

[0190] The AI ​​correction system further analyzed the distribution information and deviation information of the pointing angle in the sensor measurement status data. It found that the current pointing angle was 352.6 degrees, while the expected pointing angle should be 355 degrees. Therefore, the AI ​​correction system automatically performed the calibration procedure and adjusted the control parameters of the antenna array, such as the phase control code and phase shifter settings, to achieve precise horizontal correction. After the calibration was completed, the AI ​​correction system verified the correction results and confirmed that the pointing angle had been successfully adjusted to 355 degrees, and the performance of the antenna array had been effectively improved.

[0191] Further, Figure 2 FIG. 2 is a schematic diagram of the structure of an AI correction system 200 provided in an embodiment of the present application. Figure 2 The AI ​​correction system 200 shown includes a processor 210, which can call and run a computer program from a memory to implement the method in the embodiment of the present application.

[0192] Alternatively, if Figure 2 As shown, the AI ​​correction system 200 may further include a memory 230. The processor 210 may call and run a computer program from the memory 230 to implement the method in the embodiment of the present application.

[0193] The memory 230 may be a separate device independent of the processor 210 , or may be integrated into the processor 210 .

[0194] Alternatively, if Figure 2 As shown, the AI ​​correction system 200 may further include a transceiver 220, and the processor 210 may control the transceiver 220 to interact with other devices. Specifically, it may send information or data to other devices, or receive information or data sent by other devices.

[0195] Optionally, the AI ​​correction system 200 can implement the corresponding processes corresponding to the storage engine or components in the storage engine (such as a processing module) or a device deployed with a storage engine in each method of the embodiments of the present application. For the sake of brevity, they are not repeated here.

[0196] It should be understood that the processor of the embodiment of the present application may be an integrated circuit chip with signal processing capabilities.

[0197] It is understood that the memory in the embodiments of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memory. It should be noted that the memory of the systems and methods described herein is intended to include but is not limited to suitable types of memory.

[0198] Based on the above, a readable storage medium is provided, on which a program or instruction is stored, and when the program or instruction is executed by a processor, the steps of the above method are implemented.

[0199] It should be noted that, in this article, the terms "comprise", "include" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise one..." do not exclude the presence of other identical elements in the process, method, article or device including the element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved, for example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0200] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a computer software product, which is stored in a storage medium (such as ROM / RAM, a disk, or an optical disk), and includes a number of instructions for a terminal (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0201] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the present application, all of which are within the protection of the present application.

Claims

1. An X-band radar antenna intelligent horizontal correction method, characterized in that: The method is applied to an AI correction system, and the method comprises: Acquire sensor measurement state data of the target radar antenna array, and determine a multidimensional state attribute vector set of the sensor measurement state data; Determining a target performance feature identification result based on the multidimensional state attribute vector set, wherein the target performance feature identification result includes a plurality of first antenna performance feature units, and a feature weight of each first antenna performance feature unit indicates a confidence level that the first antenna performance feature unit is an X-band performance feature; Determining a target proofreading feature recognition result based on the multi-dimensional state attribute vector set, wherein the target proofreading feature recognition result includes a plurality of second antenna performance feature units, each second antenna performance feature unit corresponds to one of the first antenna performance feature units, and a feature weight of each second antenna performance feature unit indicates a correction decision feature matched by the first antenna performance feature unit corresponding to the second antenna performance feature unit; Determine X-band performance characteristics and horizontal correction expected characteristics based on the target performance characteristic recognition result and the target calibration characteristic recognition result; The distribution information and deviation information of the data to be adjusted are determined in the sensor measurement state data according to the X-band performance characteristics and the expected characteristics of the horizontal correction, and the target radar antenna array is calibrated based on the distribution information and the deviation information.

2. The method according to claim 1, characterized in that Determining the target performance feature recognition result based on the multidimensional state attribute vector set includes: Performing a feature embedding operation on the multidimensional state attribute vector set to determine a multidimensional state attribute vector set that has completed feature embedding, wherein the multidimensional state attribute vector set that has completed feature embedding includes a plurality of third antenna performance feature units; For each third antenna performance characteristic unit in the multidimensional state attribute vector set that has completed feature embedding, determine the target pointing angle characteristic variable of the multidimensional state attribute vector set that has completed feature embedding in the pointing dimension, determine the target phase angle characteristic variable of the multidimensional state attribute vector set that has completed feature embedding in the phase dimension, and determine the characteristic weight of the first antenna performance characteristic unit in the target performance characteristic identification result based on the target pointing angle characteristic variable and the target phase angle characteristic variable; Wherein, determining the target pointing angle characteristic variable of the multidimensional state attribute vector set in the pointing dimension for completing feature embedding includes: traversing the characteristic weights of all third antenna performance characteristic units between the third antenna performance characteristic unit and the first pointing characteristic tag of the multidimensional state attribute vector set based on the pointing dimension, and determining the maximum value among the traversed characteristic weights as the first target pointing angle characteristic variable, and / or traversing the feature weights of all third antenna performance feature units between the third antenna performance feature unit and the second pointing feature tag of the multidimensional state attribute vector set based on the pointing dimension, and determining the maximum value among the traversed feature weights as the second target pointing angle feature variable, determining the target phase angle feature variable of the multidimensional state attribute vector set in the phase dimension for completing feature embedding includes: traversing the feature weights of all third antenna performance feature units between the third antenna performance feature unit and the first phase feature tag of the multidimensional state attribute vector set based on the phase dimension, and determining the maximum value among the traversed feature weights as the first target phase angle feature variable, And / or traversing the characteristic weights of all third antenna performance characteristic units between the third antenna performance characteristic unit and the second phase characteristic label of the multidimensional state attribute vector set based on the phase dimension, and determining the maximum value among the traversed characteristic weights as the second target phase angle characteristic variable, and determining the characteristic weight of the first antenna performance characteristic unit in the target performance characteristic identification result based on the target pointing angle characteristic variable and the target phase angle characteristic variable, including: weighting the first target pointing angle characteristic variable and / or the second target pointing angle characteristic variable with the first target phase angle characteristic variable and / or the second target phase angle characteristic variable, and determining the characteristic weight of the first antenna performance characteristic unit corresponding to the third antenna performance characteristic unit in the target performance characteristic identification result.

3. The method according to claim 1, characterized in that The target performance feature identification result includes at least one performance attention indicator, and the feature weight of the first antenna performance feature unit in each performance attention indicator indicates the confidence that the first antenna performance feature unit is an X-band performance feature of the target of the corresponding performance category.

4. The method according to claim 1, characterized in that Determining the target proofreading feature recognition result based on the multi-dimensional state attribute vector set includes: A feature embedding operation is performed on the multidimensional state attribute vector set to obtain the target calibration feature identification result, wherein for the X-band performance characteristics of each performance type, the target calibration feature identification result includes at least one performance attention indicator, and the feature weight indication of the second antenna performance characteristic unit of each performance attention indicator corresponds to the first antenna performance characteristic unit of the second antenna performance characteristic unit as a performance correction reference under the set correction rule matched by the X-band performance characteristics of the performance type.

5. The method according to claim 1, characterized in that Determining the X-band performance characteristics and the expected characteristics of horizontal correction based on the target performance characteristic recognition result and the target calibration characteristic recognition result includes: Determine, based on the feature weight of each first antenna performance feature unit in the target performance feature identification result, a discrimination score that the first antenna performance feature unit belongs to the X-band performance feature; The first antenna performance feature unit whose discrimination score is greater than the set discrimination score is determined as the X-band performance feature; and based on the distribution characteristics of the X-band performance feature in the target performance feature identification result, the horizontal correction expected feature corresponding to the X-band performance feature is determined in the target proofreading feature identification result.

6. The method according to claim 1, characterized in that Determining the distribution information and deviation information of the data to be adjusted in the sensor measurement state data according to the X-band performance characteristics and the expected characteristics of the horizontal correction includes: Determine a characteristic association description between the multidimensional state attribute vector set and the sensor measurement state data; Determining distribution information and initial noise information of the data to be adjusted in the sensor measurement state data according to the X-band performance characteristics, the expected characteristics of the horizontal correction, and the characteristic association description; Determining a noise disturbance for the initial noise information based on the multidimensional state attribute vector set; The deviation information of the data to be adjusted is determined in the sensor measurement state data according to the noise disturbance and the initial noise information.

7. The method according to claim 1, characterized in that The X-band performance characteristics and the horizontal correction desired characteristics include X-band performance characteristics of the data to be adjusted for at least one time series tag and horizontal correction desired characteristics of the data to be adjusted for the performance category.

8. The method according to claim 1, characterized in that The sensor measurement state data is radar antenna array state data collected by a built-in measurement device. Before determining the multidimensional state attribute vector set of the sensor measurement state data, the method further includes: adjusting the original characteristic value of each antenna performance characteristic unit in the sensor measurement state data.

9. The method according to claim 1, characterized in that Determining the multidimensional state attribute vector set of the sensor measurement state data includes: Processing the sensor measurement state data through at least one feature embedding branch to obtain sensor measurement state data with feature embedding completed; Performing at least one feature pooling and at least one feature derivation on the sensor measurement state data after feature embedding through at least one feature pooling branch and at least one feature derivation branch to determine a multidimensional state attribute vector set of the sensor measurement state data, wherein the feature granularity of the multidimensional state attribute vector set is smaller than that of the sensor measurement state data; The multidimensional state attribute vector set, the target performance feature recognition result, and the target calibration feature recognition result are generated by at least one long short-term memory model, and the at least one long short-term memory model is debugged by the following steps: Determine a debugging sample set, wherein the debugging sample set includes at least one sensor measurement state data sample, and the at least one sensor measurement state data sample carries prior distribution information and prior deviation information of prior data to be adjusted; For each sensor measurement status data sample: Determine a multi-dimensional state attribute vector set sample of the sensor measurement state data sample; Determine a target performance feature recognition result sample based on the multi-dimensional state attribute vector set sample; Determine a target proofreading feature recognition result sample based on the multi-dimensional state attribute vector set sample; Determine an X-band performance characteristic sample and a horizontal correction expected characteristic sample based on the target performance characteristic recognition result sample and the target calibration characteristic recognition result sample; Determining a distribution information sample and a deviation information sample of a data sample to be adjusted in the sensor measurement state data sample according to the X-band performance characteristic sample and the horizontal correction expected characteristic sample; Optimizing the model weights of the at least one long short-term memory model so that the difference between the distribution information sample and the deviation information sample of the data sample to be adjusted in the sensor measurement state data sample and the prior distribution information and the prior deviation information of the prior data to be adjusted converges; The difference includes at least one of the following comparison errors: a distribution comparison error between the distribution information sample and the prior distribution information; and a deviation comparison error between the deviation information sample and the prior deviation information.

10. An AI correction system, characterized in that: The method comprises at least one processor and a memory; the memory stores computer-executable instructions; the at least one processor executes the computer-executable instructions stored in the memory, so that the at least one processor executes the method according to any one of claims 1 to 9.

Images