A method and system for processing calibration information of an array antenna based on a navigation signal

Abstract

The present application relates to the technical field of array antenna calibration information processing, and provides an array antenna calibration information processing method and system based on a navigation signal, which comprises the following steps: based on platform motion information and environmental geometry information, a signal feature dictionary is constructed by real-time identification of the direct path feature and the reflected path feature of the digitized navigation signal; based on the signal feature dictionary, the direct signal component and the reflected signal component are identified; when the quality of the direct signal component meets a preset condition, the inter-channel phase difference is extracted from the direct signal component; when the quality of the direct signal component does not meet the preset condition, the reflected signal component is compensated to construct a reference signal to extract the inter-channel phase difference; based on the inter-channel phase difference, the amplitude and phase calibration parameters of each channel of the array antenna are solved, and the amplitude and phase calibration parameters are applied to the antenna control loop. The present application has the effect of improving the accuracy of array antenna calibration information processing.

Description

Technical Field

[0001] This invention relates to the technical field of array antenna calibration information processing, and specifically to a method and system for processing array antenna calibration information based on navigation signals. Background Technology

[0002] In modern radio communication, radar, and navigation systems, array antennas are widely used due to their beamforming, anti-interference, and signal enhancement capabilities. To ensure optimal performance, the amplitude and phase relationships between the channels of an array antenna must be precisely calibrated. Global Navigation Satellite System (GNSS) signals, due to their global coverage, high-precision timing, and positioning capabilities, are often used as an external reference source for array antenna calibration. Traditionally, navigation signal reception and tracking logic is designed primarily for open areas with stable signal transmission and minimal reflection interference. In such ideal environments, the direct signals from navigation satellites are clear and of high quality, providing a reliable reference for array antenna calibration.

[0003] However, in the complex electromagnetic environment of cities, especially when equipment carrying antenna systems moves at high speeds, the reception environment for navigation signals changes drastically. Tall buildings frequently block direct signals, causing a sharp drop in signal strength or even its disappearance. At the same time, a large number of reflected signals arrive at the antenna from the building surfaces in different directions with varying time delays. These reflected signals are strong and their characteristics change rapidly with the movement of the equipment.

[0004] In this harsh environment characterized by complex terrain, high-speed movement, and strong reflection interference, traditional methods for processing array antenna calibration information based on navigation signals become completely ineffective. Not only are they unable to handle input signals with degraded quality, but their processing speed also cannot keep up with the rate of error change. This results in the calculated calibration information being "outdated" by the time it is applied, causing significant deviations in antenna beam pointing and severely impacting the equipment's operational performance.

[0005] To address the aforementioned issues, existing technologies urgently need improvement. Summary of the Invention

[0006] This application discloses a method and system for processing array antenna calibration information based on navigation signals. It aims to solve the technical problem that traditional array antenna calibration methods fail due to unstable navigation signal reception, severe reflection interference, and slow calibration information processing speed in complex urban electromagnetic environments, which in turn affects the antenna beam pointing accuracy and equipment performance.

[0007] The technical solution of this application is as follows: In a first aspect, this application discloses a method for processing array antenna calibration information based on navigation signals, including: In the complex electromagnetic environment of the city, navigation signals are received and digitally processed to obtain digitally processed navigation signals. Acquire platform motion information and environmental geometry information; Based on platform motion information and environmental geometry information, a signal feature dictionary is constructed by identifying the direct path features and reflected path features corresponding to the digitally processed navigation signals in real time. Based on the signal feature dictionary, the digitized navigation signal is decomposed to identify the direct signal component and the reflected signal component; When the quality of the direct signal component meets the preset conditions, the inter-channel phase difference is extracted from the direct signal component; when the quality of the direct signal component does not meet the preset conditions, the reflected signal component is compensated according to the characteristics of the reflected signal component and the platform motion information to construct a reference signal, and the inter-channel phase difference is extracted from the reference signal. Based on the phase difference between channels, the amplitude and phase calibration parameters of each channel of the array antenna are calculated, and the amplitude and phase calibration parameters are applied to the antenna control loop to realize the array antenna calibration information processing.

[0008] This technical solution effectively addresses the challenges of receiving navigation signals in complex urban electromagnetic environments. By intelligently identifying and processing direct and reflected signals, it ensures the extraction of reliable inter-channel phase differences under different signal quality conditions, thereby achieving precise calibration of the array antenna and significantly improving the system's adaptability and calibration accuracy in complex environments.

[0009] Furthermore, when the quality of the direct signal component does not meet the preset conditions, the reflected signal component is compensated based on its characteristics and platform motion information to construct a reference signal, and the steps for extracting the inter-channel phase difference from the reference signal include: The amplitude, phase, and interrelationship of the received signal of each antenna element are extracted from the digital signal stream of each channel to form a spatial response feature vector; Read the platform motion information and associate it with the spatial response feature vector; Based on the correlation between platform motion information and spatial response feature vector, the conversion relationship from spatial response feature vector to phase difference between each channel of array antenna is established and updated; Based on the transformation relationship, the inter-channel phase difference of the array antenna is determined according to the current spatial response feature vector and platform motion information.

[0010] With this technical solution, when the quality of the direct signal is poor, this application can utilize the reflected signal and platform motion information to extract reliable inter-channel phase differences from complex multipath environments by constructing spatial response feature vectors and dynamically updating transformation relationships. This effectively compensates for the calibration difficulties caused by the absence or poor quality of the direct signal and improves the robustness of the system under extreme conditions.

[0011] Based on this, and taking into account the phase difference between channels, the amplitude and phase calibration parameters of each channel of the array antenna are calculated, and these parameters are then applied to the antenna control loop. The steps for processing array antenna calibration information include: Real-time monitoring of the physical status of each channel of the array antenna; Identify dynamic phase error patterns associated with changes in physical state from inter-channel phase differences; Based on the changes in physical state and the dynamic mode of phase error, determine the instantaneous change in the inherent amplitude and phase error of each channel; The instantaneous change is fused with the amplitude and phase calibration parameters to obtain the fused calibration parameters; The fused calibration parameters are applied to the antenna control loop to compensate for the received signal, thereby enabling the processing of array antenna calibration information.

[0012] This technical solution, by introducing real-time monitoring of the physical state of the array antenna and identification of dynamic phase error patterns, can incorporate the instantaneous error caused by changes in physical state into the calibration parameters, thereby realizing dynamic and adaptive adjustment of the array antenna calibration parameters. This effectively solves the problem of "expired" traditional calibration parameters and significantly improves the real-time performance and accuracy of calibration.

[0013] Furthermore, the steps of applying the fused calibration parameters to the antenna control loop to compensate for the received signal, thereby realizing the array antenna calibration information processing, include: Real-time monitoring of the internal signal transmission delay and update rate of the antenna control loop; The deviation between the actual effective time of the detection fusion calibration parameters and the instantaneous error state of the current antenna channel is obtained to obtain the effective time deviation; Based on the deviation in effective time, the fusion calibration parameters are proactively adjusted; The forward-adjusted fusion calibration parameters are applied to the antenna control loop to compensate for the received signal, thereby enabling the processing of array antenna calibration information.

[0014] This technical solution effectively solves the problem of time lag in the application of calibration parameters by monitoring the internal delay and update rate of the antenna control loop and making forward-looking adjustments to the fusion calibration parameters based on the effective time deviation. This ensures that the calibration parameters can accurately match the instantaneous error state of the antenna channel when they actually take effect, further improving the real-time performance and effectiveness of the calibration.

[0015] In some preferred embodiments, the step of real-time monitoring of the physical state of each channel of the array antenna includes: The pre-deployed micro sensors are activated to acquire temperature, vibration, and stress parameters of each channel of the array antenna, thus obtaining the raw data from each micro sensor. Based on the dynamic changes in the environment where the array antenna is located, the sampling frequency and data resolution of each micro-sensor are adjusted, and the raw data is denoised and calibrated to obtain denoised and calibrated sensor data. By combining the material properties, structural design, and thermodynamic model of the array antenna, and using denoised and calibrated sensor data, the physical state distribution of each channel of the array antenna is identified.

[0016] Through this technical solution, this application can accurately and in real time acquire the physical state parameters of each channel of the array antenna by deploying micro sensors and dynamically adjusting the sampling strategy according to the environment. Combined with the antenna's own characteristics, it can perform in-depth analysis, providing high-quality and high-precision physical state data support for subsequent dynamic pattern recognition of phase error.

[0017] As an alternative, the steps for identifying dynamic phase error patterns associated with changes in physical state from inter-channel phase differences include: A multi-dimensional feature space is constructed to map the physical state of each channel of the array antenna and the phase difference between channels; Real-time monitoring of the rate of change and coupling degree of physical state, and adjustment of the dimension and feature weights of the multidimensional feature space based on the rate of change and coupling degree; Based on the adjusted dimensions and feature weights of the multidimensional feature space, an adaptive pattern recognition method is used to identify dynamic phase error patterns associated with changes in physical state.

[0018] Through this technical solution, this application constructs a multi-dimensional feature space and dynamically adjusts its dimensions and feature weights, which can more effectively capture the complex relationship between physical state changes and phase errors, realize adaptive recognition of dynamic phase error patterns, and significantly improve the accuracy and flexibility of error pattern recognition.

[0019] Based on the above, the steps of real-time monitoring of the rate of change and coupling degree of physical states, and adjusting the dimensions and feature weights of the multidimensional feature space according to the rate of change and coupling degree, include: Filtering is applied to the rate of change and the degree of coupling. Based on the rate of change and degree of coupling after filtering, the severity and correlation of the current physical state change are determined, and the state change judgment result is obtained. Based on the state change judgment results, select the dimension adjustment strategy and feature weight configuration, and adjust the dimension and feature weight of the multidimensional feature space according to the selected dimension adjustment strategy and feature weight configuration.

[0020] Through this technical solution, this application conducts a refined analysis of the rate of change and coupling degree of the physical state, and dynamically adjusts the dimension and feature weight of the multidimensional feature space accordingly, enabling the pattern recognition system to more accurately adapt to the dynamic changes of the physical state, and further optimizing the recognition effect of dynamic phase error patterns.

[0021] In one implementation, the step of identifying dynamic phase error patterns associated with changes in physical state using an adaptive pattern recognition method, based on the adjusted dimensions and feature weights of the multidimensional feature space, includes: Read and verify the dimensions and feature weights of the adjusted multidimensional feature space; Once the dimensions and feature weights of the adjusted multidimensional feature space are validated, the dynamic phase error patterns associated with changes in physical state are identified using an adaptive pattern recognition method based on the adjusted dimensions and feature weights of the multidimensional feature space.

[0022] This technical solution verifies the dimension and feature weights of the multidimensional feature space before pattern recognition, ensuring the validity and reliability of the input parameters for pattern recognition, thereby improving the accuracy and stability of dynamic pattern recognition with phase error.

[0023] In another implementation, after the adjusted dimensions and feature weights of the multidimensional feature space pass verification, the step of identifying the dynamic phase error pattern associated with changes in the physical state using an adaptive pattern recognition method based on the adjusted dimensions and feature weights of the multidimensional feature space includes: Continuously monitor the local density and distribution pattern of feature vectors in the adjusted multidimensional feature space; When the local density of the feature vectors is found to be higher than a preset threshold or the distribution patterns overlap, the local projection direction and projection weight of the adjusted multidimensional feature space are adjusted according to the severity of the change in physical state and the degree of coupling, so as to maximize the separation between similar patterns. By utilizing the local projection direction and projection weight of the adjusted multidimensional feature space, dynamic patterns of phase error can be distinguished and identified.

[0024] By using this technical solution, this application can effectively enhance the separation between similar phase error patterns by continuously monitoring the local density and distribution of feature vectors and dynamically adjusting the local projection direction and weight of the multidimensional feature space when necessary, thereby significantly improving the accuracy of phase error pattern differentiation and recognition in complex dynamic environments.

[0025] Secondly, this application also discloses an array antenna calibration information processing system based on navigation signals, used to perform array antenna calibration information processing based on navigation signals, including: The navigation signal processing module is used to receive navigation signals in the complex electromagnetic environment of the city and to digitize the navigation signals to obtain digitized navigation signals. The basic information acquisition module is used to acquire platform motion information and environmental geometry information; The feature dictionary construction module is used to construct a signal feature dictionary based on platform motion information and environmental geometry information by identifying the direct path features and reflected path features corresponding to the digitally processed navigation signals in real time. The signal component identification module is used to decompose the digitized navigation signal based on the signal feature dictionary and identify the direct signal component and the reflected signal component; The phase difference extraction module is used to extract the inter-channel phase difference from the direct signal component when the quality of the direct signal component meets the preset conditions; when the quality of the direct signal component does not meet the preset conditions, it compensates the reflected signal component according to the characteristics of the reflected signal component and the platform motion information to construct a reference signal and extract the inter-channel phase difference from the reference signal. The calibration information processing module is used to calculate the amplitude and phase calibration parameters of each channel of the array antenna based on the phase difference between channels, and apply the amplitude and phase calibration parameters to the antenna control loop to realize the calibration information processing of the array antenna.

[0026] This application provides a system-level solution through this technical solution. Through modular design, it realizes intelligent processing of navigation signals in complex urban electromagnetic environments, effective separation of direct and reflected signals, and reliable phase difference extraction under different signal quality conditions. Ultimately, it achieves precise calibration of the array antenna, providing solid hardware and software support for ensuring the performance of array antennas in complex environments.

[0027] Beneficial effects This application discloses a method for processing array antenna calibration information based on navigation signals. By receiving and digitally processing navigation signals in a complex urban electromagnetic environment, and combining platform motion information and environmental geometry information to construct a signal feature dictionary, it can identify and decompose direct and reflected signal components in real time. When the quality of the direct signal meets preset conditions, the inter-channel phase difference is directly extracted from the direct signal; when the quality of the direct signal is poor, the reflected signal is compensated according to the characteristics of the reflected signal and platform motion information, a reference signal is constructed, and the phase difference is extracted from it. Finally, based on the extracted inter-channel phase differences, the amplitude and phase calibration parameters of each channel of the array antenna are calculated and applied to the antenna control loop to realize the array antenna calibration information processing.

[0028] This method effectively solves the problems of unstable navigation signal reception and strong reflection interference causing traditional calibration methods to fail in complex urban electromagnetic environments. By intelligently distinguishing and processing direct and reflected signals, this application ensures that effective calibration can still be performed using reflected signals even when direct signals are blocked or degraded in quality, greatly improving the robustness and environmental adaptability of the calibration. Furthermore, this method provides stable and reliable calibration parameters, avoiding the calibration parameter jumps and "expirations" caused by poor signal quality in traditional methods, thereby significantly improving the beam pointing accuracy and mission performance of the array antenna in complex dynamic environments. Attached Figure Description

[0029] Figure 1 This is a flowchart of a method for processing array antenna calibration information based on navigation signals in one embodiment of the present invention; Figure 2 This is one of the flowcharts of an array antenna calibration information processing method based on navigation signals in another embodiment of the present invention; Figure 3 This is a second flowchart of a method for processing array antenna calibration information based on navigation signals in another embodiment of the present invention; Figure 4 This is a system block diagram of an array antenna calibration information processing system based on navigation signals according to another embodiment of the present invention; Explanation of reference numerals in the attached figures: 1. Navigation signal-based array antenna calibration information processing system; 11. Navigation signal processing module; 12. Basic information acquisition module; 13. Feature dictionary construction module; 14. Signal component identification module; 15. Phase difference extraction module; 16. Calibration information processing module. Detailed Implementation

[0030] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0031] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0032] This application proposes a method for processing array antenna calibration information based on navigation signals, combined with... Figure 1 As shown, it includes: S1 receives navigation signals in the complex electromagnetic environment of the city and digitizes the navigation signals to obtain digitized navigation signals. S2, acquires platform motion information and environmental geometry information; S3, based on platform motion information and environmental geometry information, constructs a signal feature dictionary by real-time identification of the direct path features and reflected path features corresponding to the digitally processed navigation signals; S4, based on the signal feature dictionary, decomposes the digitized navigation signal to identify the direct signal component and the reflected signal component; S5. When the quality of the direct signal component meets the preset conditions, the inter-channel phase difference is extracted from the direct signal component; when the quality of the direct signal component does not meet the preset conditions, the reflected signal component is compensated according to the characteristics of the reflected signal component and the platform motion information to construct a reference signal, and the inter-channel phase difference is extracted from the reference signal. S6, based on the phase difference between channels, calculates the amplitude and phase calibration parameters of each channel of the array antenna, and applies the amplitude and phase calibration parameters to the antenna control loop to realize the array antenna calibration information processing.

[0033] "Navigation signal" typically refers to Global Navigation Satellite System (GNSS) signals, such as those emitted by satellites like GPS, BeiDou, Galileo, or GLONASS. These signals carry high-precision timing and positioning information. "Digital processing" refers to converting the received analog navigation signals into digital signals using an analog-to-digital converter (ADC), followed by down-conversion, filtering, and sampling to facilitate subsequent digital signal processing. "Platform motion information" can include real-time data such as the position, velocity, acceleration, and attitude (e.g., pitch, roll, yaw) of the carrier equipped with the array antenna. This information can be acquired using sensors such as an inertial measurement unit (IMU), onboard odometer, and high-precision positioning module. "Environmental geometry information" refers to the geometric structure data of the environment surrounding the antenna, such as 3D models of buildings, terrain data, and obstacle distribution. This information can be acquired through pre-mapping, LiDAR scanning, or visual sensors. "Direct path characteristics" refers to the signal characteristics of the navigation signal propagating directly from the satellite to the antenna, such as direction of arrival (DOA), time of arrival (TOA), and signal strength. "Reflection path characteristics" refer to the signal characteristics of a navigation signal reaching the antenna path after one or more reflections. These characteristics differ from those of a direct path signal in terms of arrival direction, time delay, amplitude attenuation, and phase change. A "signal feature dictionary" is a database storing different signal path characteristics (including direct and reflected signals) used to match and identify received signal components. "Direct signal components" and "reflected signal components" refer to the signal portions corresponding to the direct and reflected paths, respectively, separated from the received mixed signal using signal decomposition techniques. "Inter-channel phase difference" refers to the relative difference in signal phase between different receiving channels of the array antenna; this is a key parameter for calibrating the array antenna. "Amplitude and phase calibration parameters" are a set of parameters used to correct inconsistencies in amplitude and phase differences between channels of the array antenna, typically expressed in complex numbers. "Antenna control loop" refers to the hardware and software system responsible for controlling and adjusting the operating state of each channel of the array antenna; calibration parameters are applied to this loop to compensate for the received signal.

[0034] The core of the array antenna calibration information processing method based on navigation signals proposed in this application lies in the intelligent identification and processing of navigation signals in complex electromagnetic environments, and on this basis, the realization of accurate array antenna calibration.

[0035] First, in the complex electromagnetic environment of an urban area, the array antenna receives navigation signals and digitizes them to obtain a digitally processed navigation signal. For example, each element of the antenna array amplifies the received radio frequency (RF) signal through a low-noise amplifier (LNA), then mixes it to an intermediate frequency (IF), and finally converts it into a digital signal stream through an analog-to-digital converter (ADC). These digital signal streams are then digitally down-converted, filtered, and sampled to remove noise and interference and extract the baseband signal.

[0036] Secondly, platform motion information and environmental geometry information are acquired. Platform motion information can be acquired in various ways. For example, an inertial measurement unit (IMU) can be used to measure the platform's angular velocity and acceleration in real time, and attitude and position changes can be obtained through integration. Simultaneously, the platform's position and velocity can be estimated by combining onboard odometer data and dead reckoning. Environmental geometry information can be obtained from a pre-established urban 3D map database, which contains information such as building height, location, and material properties. In some implementations, the surrounding environment can also be scanned in real time using onboard LiDAR or visual sensors to dynamically construct a 3D geometric model of the local environment.

[0037] Furthermore, based on platform motion information and environmental geometry information, a signal feature dictionary is constructed by real-time identification of the direct and reflected path characteristics of the digitized navigation signal. For example, platform motion information can be used to predict the line-of-sight direction of navigation satellites, and combined with environmental geometry information, the propagation path of the navigation signal in the current environment can be simulated using a ray tracing algorithm, including the direct path and possible reflected paths. For each path, features such as direction of arrival, time delay, Doppler shift, amplitude attenuation, and phase change can be calculated. These simulated feature data are organized into a signal feature dictionary, which contains the expected characteristics of the direct and reflected paths of the navigation signal under different platform positions and attitudes.

[0038] Subsequently, based on the signal feature dictionary, the digitized navigation signal is decomposed to identify the direct and reflected signal components. For example, multipath decomposition algorithms, such as subspace-based methods (e.g., MUSIC or ESPRIT) or sparse representation methods, can be used to decompose the received mixed signal into multiple independent signal components. By matching the features of these decomposed signal components (e.g., direction of arrival, time delay, etc.) with entries in the signal feature dictionary, it is possible to identify which components correspond to direct signals and which correspond to reflected signals.

[0039] Next, when the quality of the direct signal component meets preset conditions, the inter-channel phase difference is extracted from the direct signal component. When the quality of the direct signal component does not meet the preset conditions, the reflected signal component is compensated based on its characteristics and platform motion information to construct a reference signal, and the inter-channel phase difference is extracted from the reference signal. For example, preset conditions may include a carrier-to-noise ratio (CNR) of the direct signal higher than a certain threshold, stable signal strength, and a Doppler shift rate within the normal range. When the quality of the direct signal meets the conditions, cross-correlation processing can be directly performed on the direct signal components to calculate the phase difference between each channel. When the quality of the direct signal is poor (e.g., blocked or with low intensity), the identified reflected signal component can be used. By combining platform motion information and environmental geometry information, the arrival direction, time delay, and amplitude attenuation of the reflected signal can be compensated, thereby "restoring" approximate direct signal characteristics and constructing a reference signal. For example, the original incident direction of the reflected signal can be deduced by using the arrival direction of the reflected signal and platform attitude information, combined with the position of the reflection point, and the phase delay and amplitude attenuation on its propagation path can be compensated. Then, cross-correlation processing is performed on this compensated reference signal to extract the phase difference between channels.

[0040] Finally, based on the phase differences between channels, the amplitude and phase calibration parameters for each channel of the array antenna are calculated and applied to the antenna control loop to achieve array antenna calibration information processing. For example, the least squares method or other optimization algorithms can be used to calculate the amplitude and phase errors of each antenna element relative to the reference element based on the extracted phase differences between channels. These error values ​​constitute the amplitude and phase calibration parameters. These calibration parameters are then loaded into the antenna control loop, for example, by using digital attenuators and digital phase shifters to compensate the received signals of each channel in real time, thereby eliminating amplitude and phase inconsistencies between channels and ensuring the accuracy of array antenna beamforming and signal processing.

[0041] Optional, combined Figure 2 As shown, in step S5, when the quality of the direct signal component does not meet the preset conditions, the reflected signal component is compensated according to its characteristics and platform motion information to construct a reference signal, and the inter-channel phase difference is extracted from the reference signal. S51 extracts the amplitude, phase, and correlation of the received signal of each antenna element from the digital signal stream of each channel to form a spatial response feature vector; S52, reads platform motion information and associates platform motion information with spatial response feature vector; S53, based on the correlation between platform motion information and spatial response feature vector, establishes and updates the conversion relationship from spatial response feature vector to phase difference between each channel of array antenna; S54, based on the transformation relationship, determines the inter-channel phase difference of the array antenna according to the current spatial response feature vector and platform motion information.

[0042] Specifically, extracting the amplitude, phase, and interrelationships of the received signals from each antenna element from the digital signal stream of each channel to form a spatial response feature vector refers to further analysis of the digitized navigation signal. The amplitude and phase of the signal received by each antenna element are affected by a combination of multipath effects, antenna characteristics, and environmental factors. By monitoring and extracting these parameters in real time, a vector reflecting the current spatial distribution characteristics of the signal can be constructed. For example, signal processing techniques such as Fourier transform and wavelet analysis can be used to accurately separate the instantaneous amplitude and phase information of each antenna element from the digital signal stream. Furthermore, the interrelationships between the received signals of each antenna element, such as the cross-correlation function or covariance matrix, can further reveal the spatial coherence of the signal, which is crucial for subsequent identification of the characteristics of reflected signals. The purpose is to provide comprehensive spatial feature inputs for subsequent compensation and phase difference extraction.

[0043] The process of reading platform motion information and correlating it with spatial response feature vectors can be understood as synchronizing the dynamic behavior of the platform mounted on the array antenna with the characteristics of the received signal. Platform motion information can include, but is not limited to, the platform's velocity, acceleration, attitude (such as pitch, roll, and yaw angles), and position changes. This information can be acquired through an inertial measurement unit (IMU), a Global Navigation Satellite System (GNSS) receiver, or other vehicle-mounted sensors. Correlating this motion information with the real-time extracted spatial response feature vectors means recording and analyzing the corresponding spatial response characteristics under different platform motion states. The purpose is to establish a temporal and spatial correspondence between platform motion and reflected signal characteristics, providing a dynamic background for subsequent conversion relationships.

[0044] In practical applications, based on the correlation between platform motion information and spatial response feature vectors, a conversion relationship from the spatial response feature vector to the phase differences between channels of the array antenna is established and updated. Specifically, a dynamic model is constructed using algorithms such as machine learning, deep learning, or adaptive filtering. This model can learn and map the intrinsic relationship between a specific spatial response feature vector and the actual phase differences between channels of the array antenna under different platform motion states. For example, a neural network model can be used, taking platform motion information and spatial response feature vectors as input and outputting the predicted inter-channel phase differences. As new data is continuously input, the model will be continuously trained and updated to adapt to the dynamic changes in the environment and platform motion, thereby improving the accuracy and robustness of the conversion relationship. The goal is to indirectly derive accurate inter-channel phase differences using known reflected signal characteristics and platform motion states when direct signals are unavailable.

[0045] Furthermore, based on the transformation relationship, and according to the current spatial response feature vector and platform motion information, the inter-channel phase difference of the array antenna is determined. This means that during actual operation, when the quality of the direct signal is poor, the system uses the real-time acquired spatial response feature vector and platform motion information, through a previously established and continuously updated transformation relationship model, to directly calculate or predict the phase difference between each channel of the array antenna at the current moment. This process is real-time, ensuring that even in complex electromagnetic environments where the direct signal is blocked, accurate inter-channel phase difference data can be continuously provided to support the calibration of the array antenna. Its purpose is to achieve effective utilization of the reflected signal components, compensating for the calibration information gaps when the direct signal is missing or of poor quality.

[0046] In some preferred embodiments, suppose a drone equipped with an array antenna is flying between tall buildings in a city. Due to the obstruction of these buildings, the direct path of the navigation signal is often blocked, causing the quality of the direct signal component to fail to meet the preset calibration conditions. In this case, the system on the drone will activate this scheme. First, the amplitude, phase, and cross-correlation of the navigation signal received by each antenna element are extracted in real time from the digital signal stream of each channel of the array antenna, forming a spatial response feature vector. For example, when the drone approaches a tall building, the amplitude of the reflected signal received by a certain antenna element may suddenly increase, and the phase may also change drastically; this information is encoded into the feature vector. Simultaneously, the drone's inertial measurement unit (IMU) and GNSS receiver continuously provide accurate platform motion information, including the drone's instantaneous position, velocity, and attitude angle. This platform motion information is correlated with the real-time generated spatial response feature vector; for example, the spatial response characteristics of the reflected signal at a specific position and attitude are recorded. The system utilizes historically accumulated data and a pre-trained neural network model to continuously update the conversion relationship from this correlated data to the phase differences between the channels of the array antenna. When the quality of the direct signal is poor, the system utilizes the current real-time spatial response feature vector and platform motion information to predict and output the phase difference between the channels of the array antenna using this updated neural network model. For example, the model may learn that when a drone passes through a certain area at a specific speed and attitude, a certain reflected signal pattern typically corresponds to a specific phase difference between channels. In this way, even if the direct signal is unavailable, the system can continuously and accurately acquire the phase difference between channels based on the reflected signal and platform motion information, thereby achieving effective calibration of the array antenna.

[0047] Optionally, based on the phase difference between channels, the amplitude and phase calibration parameters of each channel of the array antenna are calculated, and the amplitude and phase calibration parameters are applied to the antenna control loop to realize the array antenna calibration information processing steps include: Real-time monitoring of the physical status of each channel of the array antenna; Identify dynamic phase error patterns associated with changes in physical state from inter-channel phase differences; Based on the changes in physical state and the dynamic mode of phase error, determine the instantaneous change in the inherent amplitude and phase error of each channel; The instantaneous change is fused with the amplitude and phase calibration parameters to obtain the fused calibration parameters; The fused calibration parameters are applied to the antenna control loop to compensate for the received signal, thereby enabling the processing of array antenna calibration information.

[0048] Specifically, real-time monitoring of the physical state of each channel of the array antenna refers to continuously acquiring and analyzing physical parameters related to antenna performance, such as temperature, vibration, stress, and humidity, through sensors deployed on each channel of the array antenna. The purpose is to obtain real-time physical environment information of the array antenna under different operating conditions, providing fundamental data for subsequent error analysis.

[0049] Identifying dynamic phase error patterns associated with changes in the physical states from the inter-channel phase differences can be understood as establishing a mapping relationship between physical state changes and inter-channel phase errors through data analysis and pattern recognition techniques. Specifically, machine learning algorithms or statistical models can be used to analyze the correlation between physical state changes and phase errors in historical data, thereby identifying dynamic patterns of phase errors that may occur under specific physical state changes. The aim is to reveal the intrinsic laws governing the impact of physical state changes on array antenna performance.

[0050] In practical applications, based on the aforementioned changes in physical state and the dynamic phase error patterns, the instantaneous change in the inherent amplitude and phase error of each channel is determined. Specifically, this refers to predicting or calculating the additional amplitude and phase error of each channel at the current moment due to changes in physical state, based on real-time monitored physical state data and identified dynamic phase error patterns. For example, when the temperature rises, the resulting phase drift can be determined according to a preset model or learned pattern. The purpose is to quantify the instantaneous error caused by changes in physical state for accurate compensation.

[0051] Furthermore, fusing the aforementioned instantaneous changes with the aforementioned amplitude and phase calibration parameters to obtain fused calibration parameters refers to combining the amplitude and phase calibration parameters obtained through the basic method with the instantaneous error changes determined by changes in physical state. For example, weighted averaging, Kalman filtering, or other data fusion algorithms can be used to effectively integrate the two, forming a more comprehensive and real-time calibration parameter. The aim is to generate a comprehensive calibration parameter that can simultaneously consider both static and dynamic errors.

[0052] Therefore, the aforementioned fused calibration parameters are applied to the antenna control loop to compensate for the received signal, thereby enabling array antenna calibration information processing. Specifically, this involves loading the fused calibration parameters into the array antenna control system to adjust the amplitude and phase of each antenna element in real time, thus offsetting or reducing errors in the received signal. The aim is to ensure that the array antenna maintains high-precision beamforming and signal processing capabilities even in dynamic environments.

[0053] In some preferred embodiments, a specific example is given below. Assume an array antenna is deployed on a high-speed moving platform, where the operating environment's temperature and vibration levels continuously change. First, by installing miniature temperature sensors and accelerometers on each channel of the array antenna, the temperature and vibration parameters of each channel are acquired in real time. This data constitutes the physical state of each channel of the array antenna. Simultaneously, the system continuously extracts the phase difference between channels. Next, using a pre-trained neural network model, which learns the complex relationship between temperature, vibration, and phase difference between channels from historical data, the model can identify the dynamic phase error pattern corresponding to the current physical state change in real time. For example, when the temperature rises from 20 degrees Celsius to 30 degrees Celsius, and the vibration frequency is within a specific range, the model will identify a specific phase drift pattern. Based on the identified dynamic pattern and the current physical state, the system can accurately calculate the instantaneous amplitude phase error of each channel caused by temperature and vibration changes. For example, it may calculate that the phase of a certain channel will drift by an additional 0.5 degrees. Finally, this instantaneous error is fused with the fundamental amplitude and phase calibration parameters extracted from the direct or reflected signals, for example, through optimal estimation using a Kalman filter, to obtain a fused calibration parameter. This fused calibration parameter is then applied to the antenna control loop to adjust the excitation of each antenna element in real time, thereby effectively compensating for errors caused by dynamic environmental changes and ensuring high-precision calibration of the array antenna under high-speed movement and temperature variations.

[0054] Optional, combined Figure 3 As shown, the steps for applying fused calibration parameters to the antenna control loop to compensate for the received signal and thus process array antenna calibration information include: A1, real-time monitoring of the internal signal transmission delay and update rate of the antenna control loop; A2, detect the deviation between the actual effective time of the fusion calibration parameters and the instantaneous error state of the current antenna channel to obtain the effective time deviation; A3, Based on the deviation in effective time, make forward-looking adjustments to the fusion calibration parameters; A4 applies the forward-looking adjusted fusion calibration parameters to the antenna control loop to compensate for the received signal, thereby realizing the processing of array antenna calibration information.

[0055] Specifically, real-time monitoring of the internal signal transmission delay and update rate of the antenna control loop refers to continuously acquiring the time required for a signal to travel from input to output within the antenna control loop, as well as the frequency at which calibration parameters are processed and updated within the loop. For example, this can be achieved by deploying timestamp recorders at key nodes or using hardware timers to precisely measure signal propagation time and record the parameter update cycle. The aim is to quantify the inherent time characteristics of the system, providing fundamental data for subsequent deviation detection and adjustment.

[0056] The effective time deviation is obtained by detecting the deviation between the actual effective time of the fused calibration parameters and the instantaneous error state of the current antenna channel. This can be understood as comparing the time when the calibration parameters are calculated and prepared for application with the time when the parameters actually begin to affect the signal in the antenna control loop, while considering the actual error state of the current antenna channel. Specifically, this can be achieved by comparing the generation timestamp of the calibration parameters with their actual application timestamp, and combining this with the real-time acquired antenna channel error model to calculate the time difference between the two, i.e., the effective time deviation. Its purpose is to identify the lag in the application of calibration parameters caused by system latency.

[0057] In practical applications, the fusion calibration parameters are proactively adjusted based on the effective time deviation. Specifically, this involves predicting the antenna channel error state at a future point in time based on the calculated effective time deviation, and then correcting the fusion calibration parameters accordingly. For example, prediction algorithms (such as Kalman filtering or linear prediction models) can be used in conjunction with historical error data and the effective time deviation to estimate the potential errors in the antenna channel when the calibration parameters actually take effect. This allows for advance adjustment of the calibration parameters, enabling them to more accurately compensate for the errors at the time of effective calibration. The aim is to offset the adverse effects of system latency on the calibration effect, ensuring the real-time performance and accuracy of the calibration.

[0058] Furthermore, applying the forward-adjusted fused calibration parameters to the antenna control loop to compensate for the received signal, thereby achieving array antenna calibration information processing, refers to precisely loading the calibration parameters, which have undergone time deviation compensation and forward-adjustment, into the antenna control loop so that it can immediately compensate for the received signal. The aim is to ensure that the calibration parameters function at the most appropriate time and in the most accurate form, thereby maximizing the calibration effect of the array antenna.

[0059] In some preferred embodiments, this application is implemented as follows: Assume an array antenna system operates on a high-speed mobile platform (e.g., an aircraft), and its antenna control loop has a fixed signal transmission delay of approximately 50 milliseconds. At a certain moment, a fusion calibration parameter is calculated by monitoring the physical state of each channel of the array antenna and the phase difference between channels. If this parameter is directly applied to the antenna control loop, due to the 50-millisecond delay, the parameter will only take effect after 50 milliseconds. However, within these 50 milliseconds, due to factors such as platform motion and temperature changes, the instantaneous error state of the antenna channel may have changed significantly. To solve this problem, this application first monitors the internal signal transmission delay of the antenna control loop in real time (e.g., by integrating a high-precision timer in an FPGA or DSP to measure the round-trip time of data packets from the processing unit to the antenna drive unit and recording the update cycle of the calibration parameter). Then, the system detects a 50-millisecond deviation between the calculation completion time of the fusion calibration parameter and its actual effective time. Based on this deviation, the system uses a predictive model (e.g., a linear extrapolation model based on historical error trends and platform motion information) to predict the instantaneous error state of the antenna channel after 50 milliseconds. For example, if it is predicted that the phase error of a certain channel will increase by 2 degrees within 50 milliseconds, the phase compensation value of the corresponding channel in the fused calibration parameters will be proactively adjusted before applying the fused calibration parameters to the antenna control loop to preemptively offset this expected 2-degree error. Finally, the proactively adjusted fused calibration parameters are applied to the antenna control loop to ensure that when the parameters actually take effect, they can accurately compensate for the antenna channel error at that time, thereby achieving more accurate and real-time array antenna calibration.

[0060] Optionally, the steps for real-time monitoring of the physical status of each channel of the array antenna may include the following: The pre-deployed micro-sensors are activated to obtain the temperature, vibration, and stress parameters of each channel of the array antenna, thereby obtaining the raw data of each micro-sensor.

[0061] The miniature sensors are designed to be integrated into the array antenna structure and monitor the local physical environment of the antenna elements in real time. Temperature parameters reflect the thermal load on the antenna materials and electronic components; vibration parameters indicate the mechanical stress or external vibration interference that may exist in the antenna structure; and stress parameters directly quantify the deformation or load borne by the antenna structure. The raw data of these parameters form the basis for subsequent analysis.

[0062] Based on the dynamic changes in the environment in which the array antenna is located, the sampling frequency and data resolution of each micro-sensor can be adjusted, and the raw data can be denoised and calibrated to obtain denoised and calibrated sensor data.

[0063] For example, when the array antenna is in an environment of violent motion or rapid temperature changes, the sensor's sampling frequency and data resolution can be increased to capture transient changes; while in a stable environment, the sampling frequency and resolution can be reduced to save power consumption and processing resources. Noise reduction processing aims to eliminate inherent noise or external interference from the sensor, while calibration is used to correct systematic errors in the sensor, ensuring the accuracy and reliability of the data.

[0064] By combining the material properties, structural design, and thermodynamic model of the array antenna, and using denoised and calibrated sensor data, the physical state distribution of each channel of the array antenna is identified.

[0065] Material properties include, but are not limited to, coefficient of thermal expansion and elastic modulus; structural design encompasses the layout of antenna elements, support structures, and connection methods; and thermodynamic models describe the transfer and distribution of heat within the antenna structure. By combining these engineering models with high-precision sensor data, real-time physical state maps of each channel of the array antenna can be constructed, such as temperature field distribution, stress concentration regions, or minute deformation modes.

[0066] Optionally, the step of identifying dynamic phase error patterns associated with changes in the aforementioned physical states from inter-channel phase differences includes: A multi-dimensional feature space is constructed to map the physical state of each channel of the array antenna and the phase difference between channels; Real-time monitoring of the rate of change and coupling degree of physical state, and adjustment of the dimension and feature weights of the multidimensional feature space based on the rate of change and coupling degree; Based on the adjusted dimensions and feature weights of the multidimensional feature space, an adaptive pattern recognition method is used to identify dynamic phase error patterns associated with changes in physical state.

[0067] Specifically, constructing a multi-dimensional feature space to map the physical states of each channel of the array antenna and the phase differences between channels refers to uniformly representing the physical state parameters of each channel of the array antenna (such as temperature, vibration, stress, etc.) and the phase differences between channels extracted from the navigation signal in a high-dimensional mathematical space. This mapping process aims to transform heterogeneous data into feature vectors that can be processed by pattern recognition algorithms, thereby capturing the potential correlation between physical states and phase errors more comprehensively and effectively. Here, physical states can be understood as various environmental or intrinsic parameters that affect the performance of the array antenna, with the aim of providing a quantitative description of the external and internal factors affecting phase errors.

[0068] The real-time monitoring of the rate of change and coupling degree of physical states, and the adjustment of the dimensions and feature weights of the multidimensional feature space based on these factors, refers to the system continuously tracking the speed of change (rate of change) of various physical state parameters over time and the strength of the interaction (coupling degree) between different physical parameters. Based on these real-time monitoring results, the representation of the multidimensional feature space is dynamically optimized. For example, when a physical parameter (such as temperature) changes drastically, its corresponding dimension in the feature space may be assigned a higher weight to highlight its importance for phase error pattern recognition; when multiple physical parameters (such as vibration and stress) exhibit strong coupling, their combined features may be given priority or higher weight. Dimension adjustment can include increasing or decreasing the number of features to adapt to complexity or simplify the model; feature weight adjustment is used to highlight features that contribute more to phase error pattern recognition, aiming to enable the feature space to adaptively adapt to dynamic environmental changes and ensure the effectiveness and relevance of feature representation.

[0069] In practical applications, based on the adjusted dimensions and feature weights of a multidimensional feature space, adaptive pattern recognition is used to identify dynamic phase error patterns associated with changes in physical states. This refers to using algorithms that can automatically adjust recognition strategies according to data characteristics and environmental changes within a dynamically optimized multidimensional feature space to identify phase error patterns. For example, adaptive algorithms such as Support Vector Machines (SVM), neural networks, clustering algorithms, or deep learning models can be employed. These algorithms can learn and distinguish phase error patterns corresponding to different physical states, thereby improving the accuracy and robustness of recognition. The goal is to accurately extract the correlation patterns between physical states and phase errors from a dynamically adjusted feature space using intelligent algorithms.

[0070] In some preferred embodiments, it is assumed that the array antenna operates on an aircraft, and its physical state is affected by a variety of factors such as flight altitude, speed, attitude changes, and engine vibration.

[0071] First, a multi-dimensional feature space can be constructed, which includes the temperature, vibration frequency, stress value, and inter-channel phase difference extracted from the navigation signal for each channel of the array antenna. For example, temperature parameters can be obtained using miniature sensors deployed on the antenna surface, vibration parameters can be obtained using accelerometers, and stress parameters can be obtained using strain gauges. These physical state parameters, together with the inter-channel phase difference, constitute the feature vector.

[0072] Secondly, the system monitors the rate of change of these physical state parameters and the degree of coupling between them in real time. For example, when the aircraft makes a sharp turn, the rate of change of stress and vibration parameters increases significantly, and they may exhibit a stronger coupling relationship. At this time, the system adjusts the dimensions and feature weights of the multidimensional feature space according to these dynamic changes. Specifically, it can increase the feature dimensions related to stress and vibration and increase their weights, while reducing the weights of features that do not change significantly or have low correlation in the current scenario.

[0073] Finally, based on the adjusted dimensions and feature weights of the multidimensional feature space, an adaptive pattern recognition method, such as a deep learning-based recurrent neural network (RNN) model, is used to identify dynamic phase error patterns. This RNN model can learn the dependencies in time-series data and, based on dynamically adjusted feature inputs, identify specific dynamic patterns of phase errors between array antenna channels under different flight conditions (such as level flight, climb, dive, and turn). For example, during high-speed turns, a specific phase error pattern caused by structural deformation and airflow impact may be identified, while in high-altitude, low-temperature environments, another pattern caused by thermal expansion and contraction of materials may be identified. In this way, the system can accurately understand and predict the phase errors that the array antenna will generate under specific physical conditions, thus providing an accurate basis for subsequent calibration.

[0074] Optionally, the steps of real-time monitoring of the rate of change and coupling degree of physical states, and adjusting the dimensions and feature weights of the multidimensional feature space based on the rate of change and coupling degree, include: Filtering is applied to the rate of change and the degree of coupling. Based on the rate of change and degree of coupling after filtering, the severity and correlation of the current physical state change are determined, and the state change judgment result is obtained. Based on the state change judgment results, select the dimension adjustment strategy and feature weight configuration, and adjust the dimension and feature weight of the multidimensional feature space according to the selected dimension adjustment strategy and feature weight configuration.

[0075] Specifically, filtering the rate of change and coupling degree refers to applying appropriate digital filters, such as Kalman filters, moving average filters, or median filters, to the real-time monitored data on the rate of change and coupling degree of the physical state. This removes random noise and instantaneous fluctuations from the data, ensuring the accuracy and stability of subsequent judgments. The aim is to provide smoother, more reliable input data, avoiding misjudgments or over-adjustments caused by noise.

[0076] The process involves determining the severity and correlation of the current physical state change based on the filtered rate of change and the degree of coupling, thus obtaining a state change judgment result. This can be understood as evaluating whether the filtered rate of change exceeds a specific limit and whether the degree of coupling reaches a significant level by setting preset thresholds or employing statistical analysis methods. For example, multiple severity levels (e.g., slight, moderate, severe) and correlation levels (e.g., weak correlation, moderate correlation, strong correlation) can be defined, and a comprehensive state change judgment result can be generated based on these levels. The purpose is to quantify the dynamic characteristics of physical state changes, providing a decision-making basis for subsequent adaptive adjustments.

[0077] In practical applications, based on the judgment results of state changes, a dimensional adjustment strategy and feature weight configuration are selected. Then, according to the selected strategy and weight configuration, the dimensions and feature weights of the multidimensional feature space are adjusted. Specifically, based on the judgment results, the most suitable dimensional adjustment scheme and feature weight allocation scheme for the current physical state change are selected from a predefined strategy library. For example, when the judgment results show that the physical state change is drastic and highly correlated, the dimensions of the multidimensional feature space might be increased to capture more details, and the weights of features related to the change might be increased. Conversely, when the change is minor, a lower dimension might be maintained to reduce computational complexity, and the weights adjusted to highlight stable features. The aim is to enable the multidimensional feature space to dynamically adapt to changes in the physical state, thereby more accurately mapping and identifying dynamic patterns of phase error.

[0078] Optionally, based on the adjusted dimensions and feature weights of the multidimensional feature space, the steps for identifying dynamic phase error patterns associated with changes in physical state using an adaptive pattern recognition approach include: Read and verify the dimensions and feature weights of the adjusted multidimensional feature space; Once the dimensions and feature weights of the adjusted multidimensional feature space are validated, the dynamic phase error patterns associated with changes in physical state are identified using an adaptive pattern recognition method based on the adjusted dimensions and feature weights of the multidimensional feature space.

[0079] The step of "reading and verifying the adjusted dimensions and feature weights of the multidimensional feature space" refers to the system acquiring the dimension and feature weight parameters of the multidimensional feature space determined by the preceding steps and performing a series of validity, rationality, or reliability checks on them. For example, the verification process may include parameter range checks, consistency checks, comparative analysis with historical data, or significance tests using a pre-set statistical model. The purpose is to ensure that these adjusted parameters accurately reflect the current physical state changes and provide high-quality input for subsequent pattern recognition.

[0080] "Once the adjusted dimensions and feature weights of the multidimensional feature space are verified, the system uses an adaptive pattern recognition method to identify dynamic phase error patterns associated with changes in the physical state, based on these dimensions and feature weights." This means that the system applies these verified parameters to the adaptive pattern recognition algorithm only after the above verification process confirms the validity of the dimensions and feature weights. Adaptive pattern recognition can be understood as an algorithm that can dynamically adjust its recognition strategy according to the characteristics of the input data, such as methods based on machine learning (e.g., neural networks, support vector machines), fuzzy logic, or statistical learning. Its purpose is to efficiently and accurately identify dynamic phase error patterns closely related to changes in the physical state of the array antenna from the multidimensional feature space, while ensuring the reliability of the input parameters.

[0081] Optionally, after the adjusted dimensions and feature weights of the multidimensional feature space pass verification, the steps for identifying the dynamic phase error patterns associated with changes in the physical state using an adaptive pattern recognition method, based on the adjusted dimensions and feature weights of the multidimensional feature space, include: Continuously monitor the local density and distribution pattern of feature vectors in the adjusted multidimensional feature space; When the local density of the feature vectors is found to be higher than a preset threshold or the distribution patterns overlap, the local projection direction and projection weight of the adjusted multidimensional feature space are adjusted according to the severity of the change in physical state and the degree of coupling, so as to maximize the separation between similar patterns. By utilizing the local projection direction and projection weight of the adjusted multidimensional feature space, dynamic patterns of phase error can be distinguished and identified.

[0082] Specifically, continuously monitoring the local density and distribution pattern of eigenvectors in the adjusted multidimensional feature space refers to the system's real-time analysis of the clustering degree and spatial distribution pattern of each eigenvector within the multidimensional feature space. Local density can be understood as the number of other eigenvectors within a specific radius around a given eigenvector, while distribution pattern refers to the geometric arrangement trend of eigenvectors in space, such as whether they are clustered, chain-like, or diffuse. The aim is to dynamically grasp the discriminative power of different phase error modes in the feature space and promptly identify potential mode confusion regions.

[0083] When the local density of feature vectors exceeds a preset threshold or their distribution patterns overlap, it indicates the existence of multiple phase error patterns in the multidimensional feature space that exhibit high similarity or interweaving in their features, potentially leading to ambiguity in pattern recognition. In this case, the local projection direction and projection weights of the adjusted multidimensional feature space are adjusted based on the severity and coupling degree of the physical state changes to maximize the separation between similar patterns. Specifically, the severity of the physical state changes reflects the intensity of the current disturbance to the system, while the coupling degree indicates the close interaction between different physical parameters. Based on this information, the local projection direction of the feature space can be dynamically adjusted, for example, by using methods such as Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA), or nonlinear manifold learning to find the projection axis that best distinguishes these similar patterns. Simultaneously, the projection weights are adjusted, assigning higher weights to features that contribute more to pattern differentiation. This allows similar patterns that were previously difficult to distinguish to be effectively separated in the new projection space, achieving maximum separation.

[0084] Therefore, by utilizing the local projection direction and projection weights of the adjusted multidimensional feature space, dynamic phase error patterns can be distinguished and identified. This means that in the feature space after optimized projection and weighting, different dynamic phase error patterns will exhibit clearer boundaries and more significant differences, thus enabling them to be more accurately identified and classified by pattern recognition algorithms.

[0085] In some preferred embodiments, it is assumed that the array antenna may be subjected to two different physical disturbances during operation: a slight temperature gradient change and a small accumulation of mechanical stress. These two disturbances may produce very similar dynamic patterns of phase error in the multidimensional feature space, resulting in high local density and overlapping distribution patterns of their eigenvectors in space.

[0086] Specifically, the system first continuously monitors the local density and distribution of feature vectors in the adjusted multidimensional feature space. When the local density of feature vectors in a certain region is detected to be higher than a preset threshold, and the distribution of these feature vectors shows obvious overlap, the system will determine that there is a risk of pattern confusion. For example, if the phase error patterns caused by temperature gradients and mechanical stress appear as two closely adjacent clusters in a certain feature subspace, conventional classifiers may have difficulty distinguishing them effectively.

[0087] At this point, the system adjusts the local projection direction and projection weights of the multidimensional feature space based on current physical state change information, such as the rate of temperature change and the degree of stress accumulation (i.e., the severity and coupling of physical state changes). For example, the system can employ a nonlinear dimensionality reduction technique, such as t-SNE (t-distributed Stochastic Neighbor Embedding) or UMAP (Uniform Manifold Approximation and Projection), to find an optimal projection direction within a local region, maximizing the separation of the two similar clusters in the new projection space. Simultaneously, the system can dynamically adjust feature weights based on the influence of physical state on phase error. For instance, if the current temperature change is determined to be the primary factor, temperature-related features are given higher weights, thereby further widening the gap between modes during the projection process.

[0088] Ultimately, by utilizing the projection direction and projection weights of the multidimensional feature space after this local optimization adjustment, the pattern recognition algorithm can more clearly identify and distinguish between phase error patterns caused by temperature gradients and those caused by accumulated mechanical stress. In this way, even when faced with highly similar error patterns, the proposed solution can achieve high-precision differentiation and identification, ensuring the accuracy of array antenna calibration.

[0089] This application also discloses a navigation signal-based array antenna calibration information processing system, used to perform navigation signal-based array antenna calibration information processing, combined with... Figure 4 As shown, the array antenna calibration information processing system 1 based on navigation signals includes: The navigation signal processing module 11 is used to receive navigation signals in the complex electromagnetic environment of the city and to digitize the navigation signals to obtain digitized navigation signals. The basic information acquisition module 12 is used to acquire platform motion information and environmental geometry information; The feature dictionary construction module 13 is used to construct a signal feature dictionary based on platform motion information and environmental geometry information by real-time identification of the direct path features and reflected path features corresponding to the digitally processed navigation signals; The signal component identification module 14 is used to decompose the digitized navigation signal based on the signal feature dictionary and identify the direct signal component and the reflected signal component; The phase difference extraction module 15 is used to extract the inter-channel phase difference from the direct signal component when the quality of the direct signal component meets the preset conditions; when the quality of the direct signal component does not meet the preset conditions, it compensates the reflected signal component according to the characteristics of the reflected signal component and the platform motion information to construct a reference signal and extract the inter-channel phase difference from the reference signal. The calibration information processing module 16 is used to calculate the amplitude and phase calibration parameters of each channel of the array antenna based on the phase difference between channels, and apply the amplitude and phase calibration parameters to the antenna control loop to realize the calibration information processing of the array antenna.

[0090] Specifically, the navigation signal processing module is used to receive navigation signals in the complex electromagnetic environment of an urban area and digitize them to obtain a digitally processed navigation signal. The reception and digitization of navigation signals have already been described in the above embodiments and will not be repeated here. It is important to emphasize that this navigation signal processing module can be configured to include a radio frequency (RF) front-end, an analog-to-digital converter (ADC), and a digital signal processor (DSP) or a field-programmable gate array (FPGA). The RF front-end is responsible for receiving navigation signals and performing low-noise amplification and down-conversion. The ADC converts the analog signal into a digital signal. The DSP or FPGA performs digital down-conversion, filtering, and sampling on the digital signal. As a preferred implementation, this module can be implemented using a high-performance general-purpose processor in conjunction with a software-defined radio (SDR) architecture; however, this implementation may face computational resource bottlenecks when processing complex signals with extremely high real-time requirements.

[0091] The basic information acquisition module is used to acquire platform motion information and environmental geometry information. The acquisition of platform motion information and environmental geometry information has already been described in the above embodiments and will not be repeated here. It should be emphasized that this basic information acquisition module may include an inertial measurement unit (IMU), a Global Navigation Satellite System (GNSS) receiver, an onboard odometer, and environmental sensors (such as lidar or cameras). The inertial measurement unit provides the platform's attitude, angular velocity, and acceleration information. The GNSS receiver provides high-precision position and velocity information. The onboard odometer provides the vehicle's travel distance information. The environmental sensors acquire three-dimensional geometric data of the surrounding environment. For example, this module may rely solely on the inertial measurement unit and the onboard odometer for dead reckoning, but this approach may result in significant cumulative position errors after long-term operation.

[0092] The feature dictionary construction module is used to construct a signal feature dictionary based on platform motion information and environmental geometry information by real-time identification of the direct path features and reflected path features corresponding to the digitized navigation signal. The construction of the signal feature dictionary has already been described in the above embodiments and will not be repeated here. It is important to emphasize that this feature dictionary construction module can be configured to include a ray tracing engine and a feature database. The ray tracing engine uses platform motion information and environmental geometry information to simulate the propagation path of the navigation signal in the current environment and calculates the arrival direction, time delay, Doppler shift, and other features of each path. The feature database is used to store these simulated feature data, forming the signal feature dictionary. In some embodiments, this module can use a pre-computed static environment model to construct the feature dictionary, but this approach may not adapt to dynamic changes in the environment, leading to a decrease in the accuracy of the dictionary.

[0093] The signal component identification module decomposes the digitized navigation signal based on a signal feature dictionary to identify direct and reflected signal components. The signal component identification process has already been described in the above embodiments and will not be repeated here. It is important to emphasize that this signal component identification module can be configured to include a multipath decomposition unit and a feature matching unit. The multipath decomposition unit uses a multipath decomposition algorithm (such as a subspace-based method or a sparse representation method) to decompose the received mixed signal into multiple independent signal components. The feature matching unit then matches the features of these decomposed signal components with entries in the signal feature dictionary to identify direct and reflected signal components. For example, this module can use a simple threshold-based matching strategy, but this strategy may lead to misidentification when faced with highly similar direct and reflected features.

[0094] The phase difference extraction module is used to extract the inter-channel phase difference from the direct signal component when the quality of the direct signal component meets preset conditions; when the quality of the direct signal component does not meet preset conditions, it compensates the reflected signal component according to the characteristics of the reflected signal component and platform motion information to construct a reference signal, and extracts the inter-channel phase difference from the reference signal. The extraction of phase differences has already been described in the above embodiments and will not be repeated here. It should be emphasized that this phase difference extraction module can be configured to include a signal quality evaluation unit, a direct signal processing unit, and a reflected signal compensation unit. The signal quality evaluation unit is used to evaluate the quality of the direct signal component. The direct signal processing unit directly extracts the inter-channel phase difference from the direct signal component when the direct signal quality meets preset conditions. The reflected signal compensation unit compensates the reflected signal component according to the characteristics of the reflected signal component and platform motion information when the direct signal quality is poor, to construct a reference signal, and extracts the inter-channel phase difference from the reference signal. As a specific implementation, the reflected signal compensation unit can use a simple compensation algorithm based on a geometric model, but this algorithm may not completely eliminate errors caused by complex multipath effects.

[0095] The calibration information processing module is used to calculate the amplitude and phase calibration parameters of each channel of the array antenna based on the phase difference between channels, and apply the amplitude and phase calibration parameters to the antenna control loop to realize the calibration information processing of the array antenna. The calculation and application of calibration parameters have been described in the above embodiments and will not be repeated here. It should be emphasized that the calibration information processing module can be configured to include a parameter calculation unit and a calibration parameter application unit. The parameter calculation unit uses the least squares method or other optimization algorithms to calculate the amplitude error and phase error of each antenna element relative to the reference element based on the extracted phase difference between channels, forming the amplitude and phase calibration parameters. The calibration parameter application unit then loads these calibration parameters into the antenna control loop to compensate the received signal of each channel in real time. For example, this module can use a fixed calibration parameter update frequency, but this method may not be able to respond in time to rapid changes in the physical state of the antenna, causing the calibration parameters to become "outdated".

[0096] The array antenna calibration information processing system based on navigation signals proposed in this application demonstrates significant technological progress and innovation in solving the problem of array antenna calibration in complex urban electromagnetic environments. In urban environments, traditional systems often suffer from unstable signal reception due to frequent obstruction of direct signals and strong reflection interference, making it difficult to accurately distinguish between direct and reflected signals, thus causing calibration parameter calculations to become invalid or "expired".

[0097] The above are merely embodiments of this application and are not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for processing array antenna calibration information based on navigation signals, characterized in that, include: In the complex electromagnetic environment of the city, navigation signals are received and digitally processed to obtain digitally processed navigation signals. Acquire platform motion information and environmental geometry information; Based on the platform motion information and environmental geometry information, a signal feature dictionary is constructed by real-time identification of the direct path features and reflected path features corresponding to the digitally processed navigation signals. Based on the signal feature dictionary, the digitized navigation signal is decomposed to identify the direct signal component and the reflected signal component; When the quality of the direct signal component meets the preset conditions, the inter-channel phase difference is extracted from the direct signal component; when the quality of the direct signal component does not meet the preset conditions, the reflected signal component is compensated according to the characteristics of the reflected signal component and the platform motion information to construct a reference signal, and the inter-channel phase difference is extracted from the reference signal. Based on the phase difference between the channels, the amplitude and phase calibration parameters of each channel of the array antenna are calculated, and the amplitude and phase calibration parameters are applied to the antenna control loop to realize the array antenna calibration information processing.

2. The array antenna calibration information processing method based on navigation signals according to claim 1, characterized in that, The step of compensating the reflected signal component based on its characteristics and platform motion information when the quality of the direct signal component does not meet the preset conditions, in order to construct a reference signal, and extracting the inter-channel phase difference from the reference signal, includes: The amplitude, phase, and interrelationship of the received signal of each antenna element are extracted from the digital signal stream of each channel to form a spatial response feature vector; Read the platform motion information and associate the platform motion information with the spatial response feature vector; Based on the correlation between the platform motion information and the spatial response feature vector, a conversion relationship from the spatial response feature vector to the phase difference between each channel of the array antenna is established and updated. Based on the transformation relationship, the inter-channel phase difference of the array antenna is determined according to the current spatial response feature vector and the platform motion information.

3. The array antenna calibration information processing method based on navigation signals according to claim 2, characterized in that, The steps of calculating the amplitude and phase calibration parameters of each channel of the array antenna based on the phase difference between the channels, and applying the amplitude and phase calibration parameters to the antenna control loop to realize the array antenna calibration information processing include: Real-time monitoring of the physical status of each channel of the array antenna; From the phase differences between the channels, identify dynamic patterns of phase errors that are associated with changes in the physical state; Based on the changes in the physical state and the phase error dynamic mode, determine the instantaneous change in the inherent amplitude phase error of each channel; The instantaneous change is fused with the amplitude and phase calibration parameters to obtain fused calibration parameters; The fused calibration parameters are applied to the antenna control loop to compensate the received signal, thereby enabling the processing of array antenna calibration information.

4. The array antenna calibration information processing method based on navigation signals according to claim 3, characterized in that, The step of applying the fused calibration parameters to the antenna control loop to compensate the received signal and thus process the array antenna calibration information includes: Real-time monitoring of the internal signal transmission delay and update rate of the antenna control loop; The deviation between the actual effective time of the fusion calibration parameters and the instantaneous error state of the current antenna channel is detected to obtain the effective time deviation; Based on the effective time deviation, the fusion calibration parameters are proactively adjusted; The forward-adjusted fusion calibration parameters are applied to the antenna control loop to compensate for the received signal, thereby enabling the processing of array antenna calibration information.

5. The array antenna calibration information processing method based on navigation signals according to claim 3, characterized in that, The steps for real-time monitoring of the physical state of each channel of the array antenna include: The pre-deployed micro sensors are activated to acquire temperature, vibration, and stress parameters of each channel of the array antenna, thus obtaining the raw data from each micro sensor. Based on the dynamic changes in the environment where the array antenna is located, the sampling frequency and data resolution of each micro-sensor are adjusted, and the raw data is denoised and calibrated to obtain denoised and calibrated sensor data. By combining the material properties, structural design, and thermodynamic model of the array antenna, and using denoised and calibrated sensor data, the physical state distribution of each channel of the array antenna is identified.

6. The array antenna calibration information processing method based on navigation signals according to claim 3, characterized in that, The step of identifying dynamic patterns of phase error associated with changes in the physical state from the inter-channel phase differences includes: A multi-dimensional feature space is constructed to map the physical state of each channel of the array antenna to the phase difference between the channels; The rate of change and coupling degree of the physical state are monitored in real time, and the dimensions and feature weights of the multidimensional feature space are adjusted according to the rate of change and coupling degree. Based on the adjusted dimensions and feature weights of the multidimensional feature space, an adaptive pattern recognition method is used to identify the dynamic phase error patterns associated with changes in the physical state.

7. The array antenna calibration information processing method based on navigation signals according to claim 6, characterized in that, The step of real-time monitoring of the rate of change and coupling degree of the physical state, and adjusting the dimension and feature weights of the multidimensional feature space according to the rate of change and coupling degree, includes: The rate of change and the degree of coupling are filtered. Based on the rate of change and degree of coupling after filtering, the severity and correlation of the current change in the physical state are determined, and the state change judgment result is obtained. Based on the state change judgment result, a dimension adjustment strategy and feature weight configuration are selected, and the dimensions and feature weights of the multidimensional feature space are adjusted according to the selected dimension adjustment strategy and feature weight configuration.

8. The array antenna calibration information processing method based on navigation signals according to claim 7, characterized in that, The step of identifying the dynamic phase error pattern associated with the change of the physical state based on the adjusted dimensions and feature weights of the multidimensional feature space and using an adaptive pattern recognition method includes: Read and verify the dimensions and feature weights of the adjusted multidimensional feature space; Once the dimensions and feature weights of the adjusted multidimensional feature space are verified, the dynamic phase error pattern associated with the change in the physical state is identified using an adaptive pattern recognition method based on the adjusted dimensions and feature weights of the multidimensional feature space.

9. The array antenna calibration information processing method based on navigation signals according to claim 8, characterized in that, Once the adjusted multidimensional feature space's dimensions and feature weights have passed verification, the step of identifying the dynamic phase error pattern associated with the change in the physical state using an adaptive pattern recognition method, based on the adjusted multidimensional feature space's dimensions and feature weights, includes: Continuously monitor the local density and distribution pattern of feature vectors in the adjusted multidimensional feature space; When the local density of the feature vector is found to be higher than a preset threshold or the distribution pattern overlaps, the local projection direction and projection weight of the adjusted multidimensional feature space are adjusted according to the severity of the change in the physical state and the degree of coupling, so as to maximize the separation between similar patterns. By utilizing the local projection direction and projection weight of the adjusted multidimensional feature space, dynamic patterns of phase error can be distinguished and identified.

10. A navigation signal-based array antenna calibration information processing system, used to perform navigation signal-based array antenna calibration information processing, characterized in that, include: The navigation signal processing module is used to receive navigation signals in a complex urban electromagnetic environment and to digitize the navigation signals to obtain digitized navigation signals. The basic information acquisition module is used to acquire platform motion information and environmental geometry information; The feature dictionary construction module is used to construct a signal feature dictionary based on the platform motion information and environmental geometry information by real-time identification of the direct path features and reflected path features corresponding to the digitally processed navigation signals; The signal component identification module is used to decompose the digitized navigation signal based on the signal feature dictionary and identify the direct signal component and the reflected signal component. The phase difference extraction module is used to extract the inter-channel phase difference from the direct signal component when the quality of the direct signal component meets the preset conditions; when the quality of the direct signal component does not meet the preset conditions, it compensates the reflected signal component according to the characteristics of the reflected signal component and the platform motion information to construct a reference signal and extracts the inter-channel phase difference from the reference signal. The calibration information processing module is used to calculate the amplitude and phase calibration parameters of each channel of the array antenna based on the phase difference between the channels, and apply the amplitude and phase calibration parameters to the antenna control loop to realize the calibration information processing of the array antenna.

Images