A radar beam forming method and system based on array signal processing

Abstract

The application provides a radar beam forming method and system based on array signal processing, comprising: calculating an environmental factor by processing a complex snapshot data matrix received by a current radar array; on the basis of the environmental factor, adaptively adjusting a diagonal loading factor, an iteration step and a convergence threshold; and realizing adaptive distribution of effective iteration depth through a convergence judgment mechanism. Specifically, by adaptively modifying the diagonal loading of the sampling covariance matrix, and combining the adaptive iteration step and the adaptive convergence threshold to update the beam weight vector, it is ensured that the redundant calculation is reduced when there are few interference sources, and more sufficient iteration optimization is provided in a complex interference environment to approximate the theoretical optimal beam weight vector. The application realizes efficient beam forming in different interference environments, and provides stronger real-time performance for a radar system.

Description

Technical Field

[0001] This invention relates to the field of wireless communication technology, and in particular to a radar beamforming method and system based on array signal processing. Background Technology

[0002] With the development of radar sensing technology and edge computing platforms, the application scenarios of array radar are extending from controlled environments to complex unstructured environments. Against this backdrop, radar systems not only face traditional interference such as ground clutter and multipath echoes, but also need to cope with the effects of multi-source active interference and complex electromagnetic environments. To achieve real-time and robust target detection on edge computing platforms with limited computing power, it is particularly necessary to study an efficient and adaptive radar beamforming method.

[0003] Existing radar beamforming methods based on array signal processing can be summarized as follows: First, acquire and preprocess multi-channel received data from the array antenna, and use the sampled data to estimate the covariance matrix of the received signal to sense the spatial electromagnetic environment; then, calculate the optimal weighting vector according to a specific optimization criterion (such as the minimum variance distortion-free response criterion), and perform complex weighted summation on the array element signals; finally, enhance the signal in the desired direction and effectively suppress the signal in the interference direction through spatial filtering.

[0004] For example, the Chinese invention patent with announcement number CN110161476B discloses a radar beamforming method based on the power-law iterative generalized Rayleigh quotient algorithm, which includes: 1. setting the echo signal of the radar receiving array; 2. calculating the sampling covariance matrix; 3. eigenvalue decomposition of the sampling covariance matrix; 4. calculating the optimal principal eigenvector of the radar beamformer; 5. calculating the weight vector of the radar beamformer; 6. determining whether the ratio of the radar beamformer weight vector of the current iteration to that of the previous iteration meets the specified condition. If yes, then the optimal weight of the radar beamformer is obtained and step 7 is executed; otherwise, step 4 is executed; 7. forming the radar beam.

[0005] However, in the process of implementing the inventive technical solution in the embodiments of this application, it was found that the above-mentioned technology has at least the following technical problems: Existing radar beamforming or beam weight vector iterative optimization methods typically employ fixed iterative structures and parameter configurations. However, the interference environment faced by real-world radars is highly dynamic, with the number of interference sources potentially fluctuating from a few to dozens. This mismatch between the fixed iterative structure and the dynamic environment leads to the following dilemmas: when there are few interference sources and a good condition number in the covariance matrix, the fixed iterative process results in unnecessary redundancy of computational power at the edge; while in complex environments with numerous interference sources and severe ill-conditioned matrices, the fixed number of iterations often fails to converge sufficiently, causing beamforming performance to degrade significantly with increasing interference numbers, making it difficult to approximate the theoretically optimal beam weight vector in real time. Summary of the Invention

[0006] This invention provides a radar beamforming method and system based on array signal processing, which solves the problem in the prior art where the fixed iterative structure and dynamic interference environment are mismatched in the process of optimizing the radar beamforming weight vector. As a result, the computing resources cannot be adaptively adjusted according to the interference complexity, leading to redundant computing power in simple scenarios and insufficient iteration to approach the optimal solution in complex scenarios.

[0007] In a first aspect, the present invention provides a radar beamforming method based on array signal processing, comprising: S1, obtain the complex snapshot data matrix received by the radar array at the current moment, calculate the sampling covariance matrix based on the complex snapshot data matrix, and obtain the environmental factors based on the sampling covariance matrix; S2, the sampling covariance matrix is ​​diagonally loaded and corrected using a diagonal loading factor to obtain the corrected covariance matrix. The corrected covariance matrix is ​​then inverted and combined with the target guidance vector to calculate the current beam weight vector. S3: Based on the sampling covariance matrix, the current beam weight vector, and the target steering vector, construct the minimum mean square error cost function for updating the beam weight vector, calculate the update gradient of the current beam weight vector, calculate the adaptive iteration step size according to the environmental factors, and iteratively update the current beam weight vector along the update gradient direction to obtain the updated beam weight vector. S4: Calculate the Euclidean distance between the current beam weight vector and the updated beam weight vector as the current iteration error. Generate an adaptive convergence threshold based on environmental factors. If the current iteration error is less than the adaptive convergence threshold, it is determined that convergence has been achieved and the iteration is stopped before proceeding to S5. Otherwise, the updated beam weight vector is used as the new current beam weight vector and returned to S3. S5 performs norm constraint processing on the beam weight vector output when the iteration stops, and outputs the final beam weight vector.

[0008] According to the present invention, a radar beamforming method based on array signal processing is provided, which obtains environmental factors based on the sampling covariance matrix, including: The ratio of the largest eigenvalue to the smallest eigenvalue in the sampling covariance matrix is ​​calculated to obtain the eigenvalue diffusion. The eigenvalue diffusion is nonlinearly compressed using a logarithmic function and normalized by combining it with a preset benchmark eigenvalue diffusion to obtain the environmental factor.

[0009] According to the radar beamforming method based on array signal processing provided by the present invention, after obtaining the environmental factors based on the sampling covariance matrix, the method further includes: Calculate the fluctuation deviation between the environmental factor at the current moment and the environmental factor updated at the previous moment, and obtain the dynamic smoothing coefficient corresponding to the fluctuation deviation based on the first preset mapping relationship, wherein the preset mapping relationship is configured such that the dynamic smoothing coefficient increases monotonically with the increase of the fluctuation deviation. The environmental factors are updated by weighting and coupling the current environmental factors and the updated environmental factors from the previous time step using a dynamic smoothing coefficient.

[0010] According to the radar beamforming method based on array signal processing provided by the present invention, a second preset mapping relationship between a diagonal loading factor and an environmental factor is constructed, and a diagonal loading factor adaptively adjusted according to the interference environment is calculated based on the mapping relationship, including: Obtain the effective value range corresponding to the defined diagonal loading factor, and construct a nonlinear mapping function with environmental factor as independent variable and diagonal loading factor as dependent variable based on the effective value range. The nonlinear mapping function is configured to make the diagonal loading factor monotonically increase with the increase of environmental factor, so as to characterize the positive correlation between the complexity of the disturbance environment and the loading intensity. The environmental factors at the current moment are input into the nonlinear mapping function for calculation. The calculation results are restricted to the effective value range to obtain the diagonal loading factor that adaptively adjusts with the disturbance environment.

[0011] According to the present invention, a radar beamforming method based on array signal processing is provided, which involves diagonally loading and correcting the sampled covariance matrix using a diagonal loading factor to obtain a corrected covariance matrix, inverting the corrected covariance matrix, and calculating the current beam weight vector by combining it with the target guidance vector, including: The sampling covariance matrix is ​​diagonally loaded and corrected using a diagonal loading factor to obtain the corrected covariance matrix. Then, the corrected covariance matrix is ​​inverted to obtain the corrected covariance inverse matrix. Based on the minimum variance distortionless response criterion, the matrix product of the modified covariance inverse matrix and the target steering vector is calculated, and the matrix product result is normalized using the conjugate transpose of the target steering vector to obtain the current beam weight vector.

[0012] According to the present invention, a radar beamforming method based on array signal processing is provided, which calculates an adaptive iteration step size based on environmental factors and iteratively updates the current beam weight vector along the update gradient direction to obtain an updated beam weight vector, including: A nonlinear mapping relationship between environmental factors and iteration step size is constructed, and an adaptive iteration step size is calculated based on the environmental factors at the current moment. Based on the minimum mean square error criterion, the product of the sampling covariance matrix and the current beam weight vector is calculated, and the difference between the product and the target steering vector is used to obtain the instantaneous gradient of the current beam weight vector. The instantaneous gradient is weighted using an adaptive iteration step size. The difference between the current beam weight vector and the weighted instantaneous gradient is then calculated to obtain the updated beam weight vector.

[0013] According to the present invention, a radar beamforming method based on array signal processing generates an adaptive convergence threshold according to environmental factors, comprising: Construct a positive correlation mapping function between environmental factors and convergence threshold, and calculate the adaptive convergence threshold based on the environmental factors at the current time. The positive correlation mapping function is configured to adaptively increase the convergence threshold as the complexity of the interference environment increases, so as to match the steady-state fluctuation amplitude of the weight vector under complex interference environment.

[0014] According to the radar beamforming method based on array signal processing provided by the present invention, the beam weight vector output at the stopping iteration is subjected to norm constraint processing to output the final beam weight vector, including: Calculate the Euclidean norm of the beam weight vector output when the iteration stops, and use the Euclidean norm to normalize the beam weight vector output when the iteration stops, so as to obtain the final beam weight vector after norm constraint.

[0015] Secondly, the present invention also provides a radar beamforming system based on array signal processing, comprising: The environmental factor acquisition module is used to acquire the complex snapshot data matrix received by the radar array at the current moment, calculate the sampling covariance matrix based on the complex snapshot data matrix, and obtain the environmental factors based on the sampling covariance matrix. The diagonal loading correction module is used to perform diagonal loading correction on the sampling covariance matrix using a diagonal loading factor to obtain a corrected covariance matrix. The corrected covariance matrix is ​​then inverted, and the current beam weight vector is calculated in combination with the target guidance vector. The step size iterative update module is used to construct a minimum mean square error cost function for updating the beam weight vector based on the sampling covariance matrix, the current beam weight vector and the target steering vector, calculate the update gradient of the current beam weight vector, calculate the adaptive iterative step size according to environmental factors, and iteratively update the current beam weight vector along the update gradient direction to obtain the updated beam weight vector. The convergence comparison decision module is used to calculate the Euclidean distance between the current beam weight vector and the updated beam weight vector as the current iteration error. An adaptive convergence threshold is generated based on environmental factors. If the current iteration error is less than the adaptive convergence threshold, it is determined that convergence has been achieved and the iteration is stopped. Then, the operation of the beam weight vector output module is executed. Otherwise, the updated beam weight vector is used as the new current beam weight vector and the operation of the step size iteration update module is re-executed. The beam weight vector output module is used to perform norm constraint processing on the beam weight vector output when the iteration stops, and output the final beam weight vector.

[0016] Thirdly, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the radar beamforming method based on array signal processing as described above.

[0017] The radar beamforming method and system based on array signal processing provided by this invention constructs an environmental perception mechanism through extracted environmental factors. This method can adjust the computation process in real time according to fluctuations in the number of interference sources: in simple scenarios with few interference sources and good matrix condition numbers, it terminates the iteration early through the adaptive convergence threshold mechanism of S4, avoiding unnecessary redundancy of edge-side computing power caused by the fixed iteration process in existing technologies; in complex scenarios with many interference sources and severe matrix ill-conditioning, it automatically increases the number of effective iterations. This dynamic iterative control method effectively solves the technical problem of mismatch between the fixed iterative structure and the dynamic environment in existing technologies.

[0018] Furthermore, addressing the shortcomings of existing technologies where beamforming performance significantly degrades and fails to approximate the theoretical optimal solution in real time due to insufficient fixed iterations when there are numerous interference sources, this invention utilizes environmental factors to adaptively adjust the diagonal loading factor in S2, effectively correcting the ill-conditioned covariance matrix and providing a stable numerical basis for subsequent optimization. Simultaneously, S3 introduces an adaptive iteration step size related to environmental factors, improving convergence speed in simple environments and reducing the step size in complex environments to suppress oscillations and enhance convergence stability. Compared to the performance bottleneck caused by rigid iterative control in existing technologies, this invention ensures full convergence of the beam weight vector in complex environments, significantly improving the radar's anti-interference capability and signal fidelity in harsh electromagnetic environments.

[0019] Meanwhile, existing technologies often over-compute in simple scenarios, wasting valuable processing time and power. This invention calculates the current iteration error using S4 and compares it with an adaptive convergence threshold, achieving intelligent control of the iteration process. Compared to traditional methods with a fixed number of iterations, this invention significantly reduces redundant computational steps in simple scenarios while maintaining performance, effectively reducing the processing latency of the radar system and improving its real-time response speed. This has significant engineering practical value for edge radar platforms with limited computing power.

[0020] Compared to traditional beamforming methods that are prone to performance degradation under low signal-to-noise ratio or small snapshot counts, this invention provides dual protection from two dimensions: the numerical stability of matrix inversion and the stability of weight vector amplitudes, through adaptive diagonal loading correction in S2 and norm constraint processing in S5. This method effectively suppresses the divergence of noise eigenvalues ​​and eliminates random fluctuations in the weight vector. Compared to existing technologies, it can maintain stable and reliable beamforming output even under conditions of limited data volume or poor signal-to-noise ratio.

[0021] In summary, this invention, by introducing an environment-factor-driven adaptive diagonal loading, adaptive step size, and adaptive convergence control mechanism, achieves a transformation from fixed parameters and fixed iterative processes to environment-aware dynamic iterative control, effectively solving the problem of mismatch between fixed iterative structures and dynamic interference environments in existing technologies. It reduces computational redundancy in simple interference scenarios and enhances convergence sufficiency in complex scenarios, significantly improving the robustness, resource utilization efficiency, and ability to approximate the optimal solution of the radar beamforming algorithm in dynamic interference environments. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0023] Figure 1 This is a flowchart illustrating the radar beamforming method based on array signal processing provided by the present invention. Figure 2 This is the logic diagram of the convergence comparison and decision module provided by the present invention; Figure 3 This is a schematic diagram of the radar beamforming system based on array signal processing provided by the present invention; Figure 4 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

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

[0025] In modern radar systems, beamforming technology is one of the core technologies for achieving high-precision target detection and interference suppression. With the diversification of radar application scenarios, the interference environment faced by radar systems is becoming increasingly complex, and the number, location, and characteristics of interference sources can change drastically in time and space. Traditional beamforming methods often rely on fixed iterative structures and parameter configurations to solve for beam weight vectors. However, this fixed iterative structure suffers from a significant mismatch with the dynamically changing interference environment, making it inefficient in responding to environmental changes in practical applications.

[0026] Therefore, existing radar beamforming methods using fixed iterative structures often suffer from problems such as inability to adaptively adjust computational resources and failure to achieve sufficient convergence when facing dynamic and complex interference environments, leading to wasted computational resources and performance degradation. How to dynamically adjust the iterative process, optimize the computational flow, and ensure beamforming performance under different interference environments is a core issue that urgently needs to be addressed in current radar beamforming technology.

[0027] Figure 1 This is a flowchart illustrating the radar beamforming method based on array signal processing provided in an embodiment of the present invention, as shown below. Figure 1 As shown, it includes: S1, obtain the complex snapshot data matrix received by the radar array at the current moment, calculate the sampling covariance matrix based on the complex snapshot data matrix, and obtain the environmental factors based on the sampling covariance matrix; S2, the sampling covariance matrix is ​​diagonally loaded and corrected using a diagonal loading factor to obtain the corrected covariance matrix. The corrected covariance matrix is ​​then inverted and combined with the target guidance vector to calculate the current beam weight vector. S3: Based on the sampling covariance matrix, the current beam weight vector, and the target steering vector, construct the minimum mean square error cost function for updating the beam weight vector, calculate the update gradient of the current beam weight vector, calculate the adaptive iteration step size according to the environmental factors, and iteratively update the current beam weight vector along the update gradient direction to obtain the updated beam weight vector. S4: Calculate the Euclidean distance between the current beam weight vector and the updated beam weight vector as the current iteration error. Generate an adaptive convergence threshold based on environmental factors. If the current iteration error is less than the adaptive convergence threshold, it is determined that convergence has been achieved and the iteration is stopped before proceeding to S5. Otherwise, the updated beam weight vector is used as the new current beam weight vector and returned to S3. S5 performs norm constraint processing on the beam weight vector output when the iteration stops, and outputs the final beam weight vector.

[0028] It is understood that the method of this invention includes several core steps: environmental factor acquisition, diagonal loading correction, step size iterative update, convergence comparison decision, and beam weight vector output, as detailed below: The specific steps for obtaining environmental factors in S1 are as follows: S11, calculate the ratio of the largest eigenvalue to the smallest eigenvalue of the sampling covariance matrix to obtain the eigenvalue diffusion. This eigenvalue diffusion directly reflects the numerical characteristics of the received data covariance matrix: when the interference environment is simple and the signal-to-noise ratio is high, the eigenvalue distribution is relatively concentrated, and the eigenvalue diffusion is small; when there are many interference sources or strong interference, the gap between large and small eigenvalues ​​widens, the ill-conditioned nature of the matrix intensifies, and the eigenvalue diffusion will increase significantly.

[0029] S12, using a logarithmic function to nonlinearly compress the eigenvalue diffusion and then normalizing it based on a preset benchmark eigenvalue diffusion, yields the environmental factor. Specifically, the constraint expression for the environmental factor is: , In the formula, Indicates environmental factors, Indicates the eigenvalue diffusion. Represents the largest eigenvalue. Represents the smallest eigenvalue. This represents the preset baseline feature value dispersion, which is usually set according to the nominal operating environment or empirical value of the radar system (for example, it can be set to the feature value ratio corresponding to a medium-complexity interference environment).

[0030] Through the above processing, the environmental factor becomes a dimensionless scalar. When the complexity of the actual disturbance environment is the same as the baseline environment, the environmental factor equals 1; when the environment is more complex than the baseline environment (the matrix ill-conditioned is more severe), the environmental factor is greater than 1; when the environment is simpler than the baseline environment, the environmental factor is less than 1. This environmental factor can accurately and quantitatively characterize the relative complexity of the current disturbance environment, providing a precise input basis for dynamically adjusting the diagonal loading factor, iteration step size, and convergence threshold in subsequent steps.

[0031] As a further approach, in radar systems, the interference environment is typically dynamic, and environmental factors reflect the complexity of this change. In some cases, environmental factors may fluctuate significantly, especially when interference sources suddenly increase or change drastically. Such fluctuations can cause the system's beamforming algorithm to overreact or exhibit unstable behavior. Particularly when facing time-varying interference, the instantaneous changes in environmental factors lead to frequent updates of the beam weight vector, which can cause unnecessary computational overhead and affect system stability. Therefore, adaptive temporal smoothing of environmental factors can smooth out these instantaneous fluctuations, making their changes more stable, thus providing a more stable and consistent input for the subsequent beamforming process, improving system performance and efficiency. Therefore, after obtaining the environmental factors, the process also includes adaptive temporal smoothing of the environmental factors, with the following specific steps: S121, calculate the fluctuation deviation between the environmental factors at the current time and the environmental factors updated at the previous time, i.e., the absolute difference between the two. Based on a preset mapping relationship, find the dynamic smoothing coefficient corresponding to the fluctuation deviation. The mapping relationship is configured such that the dynamic smoothing coefficient monotonically increases as the fluctuation deviation increases. Specifically, a preset mapping table or mapping function stores the correspondence between fluctuation deviation and dynamic smoothing coefficient. This mapping relationship is configured such that the dynamic smoothing coefficient monotonically increases as the fluctuation deviation increases. The dynamic smoothing coefficient is typically set within the range [0,1], and the mapping relationship can be implemented using a piecewise linear function, a sigmoid function, or a lookup table.

[0032] For example, in specific implementation, the following mapping logic is constructed: the calculated fluctuation deviation is compared with the preset first deviation threshold and second deviation threshold, wherein the first deviation threshold is less than the second deviation threshold, both of which are set by professionals according to the standards in the field. The dynamic smoothing coefficient is determined based on the comparison result: if the fluctuation deviation is less than or equal to the first deviation threshold, it is determined that the environment is in a stable state, and the dynamic smoothing coefficient is set to the preset minimum smoothing coefficient. At this time, the smoothing process focuses on using historical data to suppress noise jitter.

[0033] If the fluctuation deviation is greater than or equal to the second deviation threshold, the environment is determined to be in a sudden change state, and the dynamic smoothing coefficient is set to the preset maximum smoothing coefficient. At this time, the smoothing process focuses on quickly responding to the sudden changes in the current data.

[0034] If the fluctuation deviation is between the first deviation threshold and the second deviation threshold, the environment is determined to be in a transitional state. The dynamic smoothing coefficient is calculated according to the linear interpolation rule, and its value increases linearly with the increase of the fluctuation deviation to ensure a smooth transition of the smoothing intensity.

[0035] S122 updates the environmental factors by applying a weighted coupling process using a dynamic smoothing coefficient to the current environmental factors and the updated environmental factors from the previous time step. The calculation formula for the weighted coupling process is as follows: In the formula, This represents the updated environmental factors. Indicates the dynamic smoothing coefficient. Represents the environmental factors at the current moment. This represents the environmental factors updated at the previous moment.

[0036] Through adaptive temporal smoothing, environmental factors can remain stable in complex interference environments, avoiding performance degradation caused by instantaneous fluctuations in environmental factors and improving the system's computational efficiency, stability, and robustness. This step provides a more reliable and consistent input for subsequent beamforming optimization, enhancing the radar system's adaptability to dynamically changing environments.

[0037] S2. A mapping relationship between the diagonal loading factor and environmental factors is established. Based on this mapping relationship, the diagonal loading factor is calculated to adaptively adjust with the interference environment. The diagonal loading factor is used to correct and invert the sampling covariance matrix to obtain the corrected covariance matrix. This corrected covariance matrix is ​​then calculated in conjunction with the target steering vector to determine the current beam weight vector. By establishing a mapping relationship between the diagonal loading factor and environmental factors and dynamically adjusting the diagonal loading factor based on this relationship, this method can effectively correct the sampling covariance matrix, especially in complex interference environments or when the matrix condition number is poor. Adaptive correction of the covariance matrix not only improves the numerical stability of the matrix but also reduces the impact of ill-conditioned matrices on the beamforming process, ensuring the accuracy and stability of the beam weight vector optimization process. This step significantly improves the system's beamforming capability in complex environments, providing strong support for subsequent target signal enhancement and interference suppression.

[0038] In radar beamforming systems, the diagonal loading factor is a key parameter used to correct the sampling covariance matrix, improve numerical stability, and optimize the beam weight vector. To enable the system to adapt to complex and ever-changing interference environments, a mechanism must be designed to dynamically adjust the diagonal loading factor based on environmental factors. By constructing a nonlinear mapping relationship between environmental factors and the diagonal loading factor, the strength of the diagonal loading factor can be automatically adjusted under different interference environments, thereby improving the accuracy and robustness of beamforming. It should be noted that constructing the mapping relationship between the diagonal loading factor and environmental factors, and calculating the diagonal loading factor adaptively adjusted according to the interference environment based on this mapping relationship, specifically includes: S21, Obtain the valid value range of the defined diagonal loading factor. The valid value range is the closed interval corresponding to the preset lower limit of diagonal loading to the upper limit of diagonal loading. Construct a nonlinear mapping function with environmental factors as independent variables and diagonal loading factors as dependent variables. The nonlinear mapping function is configured such that the diagonal loading factor monotonically increases with the increase of environmental factors to characterize the positive correlation between the complexity of the interference environment and the loading intensity. As a specific implementation method, the nonlinear mapping function can adopt an exponential function model.

[0039] S22, the environmental factors at the current moment are input into the nonlinear mapping function for calculation, and the calculation results are subjected to boundary constraint processing to limit the calculation results to within the effective value range, so as to obtain the diagonal loading factor that is adaptively adjusted with the disturbance environment.

[0040] By establishing a nonlinear mapping relationship between environmental factors and diagonal loading factors, the diagonal loading factors can be adaptively adjusted according to the complexity of the interference environment. Through this mechanism, the radar system can achieve efficient beamforming under different interference environments and maintain excellent interference suppression performance in dynamically changing environments. The adaptive adjustment method improves the system's flexibility, stability, and robustness, and effectively optimizes the use of computational resources, ensuring the overall performance and computational efficiency of the system under complex interference conditions.

[0041] As a further approach, the sampling covariance matrix is ​​corrected and inverted using a diagonal loading factor to obtain a corrected covariance matrix, which is then combined with the target steering vector to calculate the current beam weight vector. Specifically, this includes: First, construct an identity matrix with the same dimensions as the sampling covariance matrix. Then, perform matrix addition on the product of the diagonal loading factor and the identity matrix to obtain the corrected covariance matrix. Finally, perform matrix inversion on the corrected covariance matrix to obtain the corrected covariance inverse matrix.

[0042] Next, based on the minimum variance distortionless response criterion, the matrix product of the modified covariance inverse matrix and the target steering vector is calculated, and the matrix product result is normalized using the conjugate transpose of the target steering vector to obtain the current beam weight vector.

[0043] By correcting the covariance matrix using a diagonal loading factor and calculating the beam weight vector in conjunction with the target steering vector, the numerical stability of the radar system in complex interference environments is enhanced, and the accuracy and efficiency of beamforming are also improved. Through adaptive adjustment of the loading factor, the system can better adapt to dynamically changing environments, improve the reception quality of target signals, and effectively suppress interference. This process significantly optimizes beamforming performance, ensuring that the radar system can operate efficiently and stably under various interference conditions.

[0044] S3: Based on the sampling covariance matrix and the current beam weight vector, a minimum mean square error cost function is constructed for beam weight vector updating. The update gradient of the current beam weight vector is calculated, and an adaptive iteration step size related to environmental factors is introduced. The current beam weight vector is iteratively updated along the update gradient direction to obtain the updated beam weight vector. In this step, a weight vector update model under the minimum mean square error criterion is constructed based on the sampling covariance matrix and the current beam weight vector, and an adaptive iteration step size related to environmental factors is introduced to update the beam weight vector. By optimizing the beam weight vector, the reception quality of the target signal can be effectively improved and interference can be gradually suppressed. The introduction of the adaptive step size not only accelerates the convergence process but also enhances the stability of beam weight vector updating when the interference environment is complex. This method ensures the efficiency and stability of the algorithm under different interference intensities, ensuring that the beamforming approximates the theoretically optimal beam weight vector as closely as possible.

[0045] It should be further explained that an adaptive iteration step size related to environmental factors is introduced, and the current beam weight vector is iteratively updated along the update gradient direction to obtain the updated beam weight vector, specifically including: S31. Construct a nonlinear mapping relationship between environmental factors and iteration step size. Calculate the adaptive iteration step size based on the environmental factors at the current time. Since environmental factors characterize the complexity of the interfering environment and the ill-conditioned nature of the matrix, when the environmental complexity is high, the error surface is usually steep or noisy. In this case, a smaller step size should be used to ensure convergence stability. When the environmental complexity is low, the error surface is relatively flat, and a larger step size can be used to accelerate convergence. As a specific implementation method, an exponential decay function can be used to calculate the adaptive iteration step size.

[0046] S32, based on the minimum mean square error criterion, calculate the product of the sampling covariance matrix and the current beam weight vector, and obtain the difference between the product and the target steering vector to get the instantaneous gradient of the current beam weight vector.

[0047] S33 uses an adaptive iteration step size to weight the instantaneous gradient, and performs a difference operation between the current beam weight vector and the weighted instantaneous gradient to obtain the updated beam weight vector.

[0048] By introducing an adaptive iterative step size related to environmental factors and iteratively updating the beam weight vector, the system can flexibly adjust the step size according to the complexity of the interference environment, ensuring the stability, efficiency, and accuracy of the beamforming process. The adaptive step size mechanism enhances the system's adaptability and robustness, especially in complex interference environments, enabling faster convergence and effective interference suppression. This process not only improves the target signal gain and reception quality but also optimizes the use of computational resources, ensuring the radar system maintains excellent performance in dynamically changing interference environments.

[0049] S4, calculate the Euclidean distance between the current beam weight vector and the updated beam weight vector as the current iteration error, generate an adaptive convergence threshold based on environmental factors, and compare the current iteration error with the adaptive convergence threshold, such as... Figure 2 As shown, if the current iteration error is less than the adaptive convergence threshold, convergence is determined, and iteration stops before executing S5. Otherwise, the updated beam weight vector is used as the new current beam weight vector, and the process returns to S3. By calculating the Euclidean distance between the current beam weight vector and the updated beam weight vector, and generating an adaptive convergence threshold based on environmental factors, the system can flexibly determine whether the convergence condition has been met. In simple interference environments, the system can stop iterations in a timely manner, avoiding unnecessary computational waste. In complex environments, the system ensures a sufficient number of iterations to achieve convergence and obtain accurate beam weight vectors. This mechanism not only improves computational efficiency and avoids resource waste but also ensures convergence accuracy in complex scenarios, significantly improving the quality of beamforming and the algorithm's adaptive capability.

[0050] It should be noted that in S4, the specific steps for generating the adaptive convergence threshold based on environmental factors include: A positive correlation mapping function between environmental factors and convergence threshold is constructed, and an adaptive convergence threshold is calculated based on the environmental factors at the current time. The positive correlation mapping function is configured to adaptively increase the convergence threshold as the complexity of the disturbance environment increases, so as to match the steady-state fluctuation amplitude of the weight vector under complex disturbance environment.

[0051] By dynamically adjusting the adaptive convergence threshold based on environmental factors, the system can maintain stable and efficient beamforming performance in various interference environments. As the complexity of the interference environment increases, the convergence threshold adaptively increases, enabling the system to better adapt to larger fluctuations and interference, thus improving convergence accuracy, stability, and efficiency. This mechanism enhances the system's adaptability, optimizes computational efficiency, and ensures high robustness and beamforming quality in complex environments.

[0052] S5 applies norm constraints to the beam weight vector output at the end of the iteration, outputting the final beam weight vector. By applying norm constraints to the converged beam weight vector, the physical realizability and amplitude limitations of the beam weight vector in practical radar applications are ensured. This process guarantees that the amplitude of the final beam weight vector meets actual engineering requirements, avoiding physical implementation problems caused by excessively large or small amplitudes, and ensuring the stability and reliability of the system. The introduction of norm constraints not only improves the practicality of the output beam weight vector but also maintains the signal gain and interference suppression performance during beamforming, enabling the system to maintain consistently excellent performance in complex radar environments.

[0053] It should be noted that the beam weight vector output at the end of the iteration is subjected to norm constraint processing, specifically including: Calculate the Euclidean norm of the beam weight vector output when the iteration stops, and use the Euclidean norm to normalize the beam weight vector output when the iteration stops, so as to obtain the final beam weight vector after norm constraint.

[0054] By applying norm constraints to the beam weight vector output at the end of the iteration, the system can ensure that the beam weight vector remains within a stable and consistent amplitude range. This is crucial for improving the accuracy, stability, and robustness of beamforming. Norm constraints not only enhance signal gain and interference suppression capabilities but also strengthen the system's adaptability to different interference environments and noise levels. They optimize convergence speed, reduce computational resource waste, and guarantee stable and efficient system operation under various conditions.

[0055] The radar beamforming system based on array signal processing provided by the present invention is described below. The radar beamforming system based on array signal processing described below can be referred to in correspondence with the radar beamforming method based on array signal processing described above.

[0056] Figure 3 This is a schematic diagram of the radar beamforming system based on array signal processing provided in an embodiment of the present invention, as shown below. Figure 3 As shown, it includes: an environmental factor acquisition module 31, a diagonal loading correction module 32, a step size iterative update module 33, a convergence comparison and decision module 34, and a beam weight vector output module 35, wherein: The environmental factor acquisition module 31 acquires the complex snapshot data matrix received by the radar array at the current moment, calculates the sampling covariance matrix based on the complex snapshot data matrix, and obtains the environmental factors based on the sampling covariance matrix. The diagonal loading correction module 32 performs diagonal loading correction on the sampling covariance matrix using the diagonal loading factor to obtain the corrected covariance matrix, performs inversion on the corrected covariance matrix, and calculates the current beam weight vector in combination with the target guidance vector. The step size iteration update module 33 constructs the minimum mean square error cost function for beam weight vector update based on the sampling covariance matrix, the current beam weight vector, and the target guidance vector, calculates the update gradient of the current beam weight vector, and calculates the environmental factors. An adaptive iteration step size is used to iteratively update the current beam weight vector along the update gradient direction to obtain the updated beam weight vector. The convergence comparison decision module 34 is used to calculate the Euclidean distance between the current beam weight vector and the updated beam weight vector as the current iteration error. An adaptive convergence threshold is generated based on environmental factors. If the current iteration error is less than the adaptive convergence threshold, it is determined that convergence has been achieved and the iteration stops. Then, the operation of the beam weight vector output module is executed. Otherwise, the updated beam weight vector is used as the new current beam weight vector and the operation of the step size iteration update module is executed again. The beam weight vector output module 35 is used to perform norm constraint processing on the beam weight vector output when the iteration stops and output the final beam weight vector.

[0057] Figure 4 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 4As shown, the electronic device may include: a processor 410, a communication interface 420, a memory 430, and a communication bus 440. The processor 410, communication interface 420, and memory 430 communicate with each other via the communication bus 440. The processor 410 can call logical instructions in the memory 430 to execute a radar beamforming method based on array signal processing. This method includes: S1, acquiring the complex snapshot data matrix received by the radar array at the current moment, calculating the sampling covariance matrix based on the complex snapshot data matrix, and obtaining environmental factors based on the sampling covariance matrix; S2, performing diagonal loading correction on the sampling covariance matrix using a diagonal loading factor to obtain a corrected covariance matrix, performing an inversion operation on the corrected covariance matrix, and calculating the current beam weight vector in conjunction with the target steering vector; S3, constructing a minimum beam weight vector update based on the sampling covariance matrix, the current beam weight vector, and the target steering vector. S4 calculates the updated gradient of the current beam weight vector using the mean squared error cost function, calculates the adaptive iteration step size based on environmental factors, and iterates along the updated gradient direction to obtain the updated beam weight vector; S5 calculates the Euclidean distance between the current beam weight vector and the updated beam weight vector as the current iteration error, generates an adaptive convergence threshold based on environmental factors, and if the current iteration error is less than the adaptive convergence threshold, it is determined that convergence has been achieved and the iteration stops, then proceeds to S5; otherwise, the updated beam weight vector is used as the new current beam weight vector and returned to S3; S6 performs norm constraint processing on the beam weight vector output when the iteration stops, and outputs the final beam weight vector.

[0058] Furthermore, the logical instructions in the aforementioned memory 430 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0059] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0060] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0061] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A radar beamforming method based on array signal processing, characterized in that, include: S1, obtain the complex snapshot data matrix received by the radar array at the current moment, calculate the sampling covariance matrix based on the complex snapshot data matrix, and obtain the environmental factors based on the sampling covariance matrix; S2, the sampling covariance matrix is ​​diagonally loaded and corrected using a diagonal loading factor to obtain the corrected covariance matrix. The corrected covariance matrix is ​​then inverted and combined with the target guidance vector to calculate the current beam weight vector. S3: Based on the sampling covariance matrix, the current beam weight vector, and the target steering vector, construct the minimum mean square error cost function for updating the beam weight vector, calculate the update gradient of the current beam weight vector, calculate the adaptive iteration step size according to the environmental factors, and iteratively update the current beam weight vector along the update gradient direction to obtain the updated beam weight vector. S4: Calculate the Euclidean distance between the current beam weight vector and the updated beam weight vector as the current iteration error. Generate an adaptive convergence threshold based on environmental factors. If the current iteration error is less than the adaptive convergence threshold, it is determined that convergence has been achieved and the iteration is stopped before proceeding to S5. Otherwise, the updated beam weight vector is used as the new current beam weight vector and returned to S3. S5 performs norm constraint processing on the beam weight vector output when the iteration stops, and outputs the final beam weight vector.

2. The radar beamforming method based on array signal processing according to claim 1, characterized in that, Environmental factors are obtained based on the sampling covariance matrix, including: The ratio of the largest eigenvalue to the smallest eigenvalue in the sampling covariance matrix is ​​calculated to obtain the eigenvalue diffusion. The eigenvalue diffusion is nonlinearly compressed using a logarithmic function and normalized by combining it with a preset benchmark eigenvalue diffusion to obtain the environmental factor.

3. The radar beamforming method based on array signal processing according to claim 2, characterized in that, After obtaining the environmental factors based on the sampling covariance matrix, the following is also included: Calculate the fluctuation deviation between the environmental factor at the current moment and the environmental factor updated at the previous moment, and obtain the dynamic smoothing coefficient corresponding to the fluctuation deviation based on the first preset mapping relationship, wherein the preset mapping relationship is configured such that the dynamic smoothing coefficient increases monotonically with the increase of the fluctuation deviation. The environmental factors are updated by weighting and coupling the current environmental factors and the updated environmental factors from the previous time step using a dynamic smoothing coefficient.

4. The radar beamforming method based on array signal processing according to claim 1, characterized in that, A second pre-defined mapping relationship between the diagonal loading factor and environmental factors is constructed. Based on this mapping relationship, the diagonal loading factor is calculated to adaptively adjust with the disturbance environment, including: Obtain the effective value range corresponding to the defined diagonal loading factor, and construct a nonlinear mapping function with environmental factor as independent variable and diagonal loading factor as dependent variable based on the effective value range. The nonlinear mapping function is configured to make the diagonal loading factor monotonically increase with the increase of environmental factor, so as to characterize the positive correlation between the complexity of the disturbance environment and the loading intensity. The environmental factors at the current moment are input into the nonlinear mapping function for calculation. The calculation results are restricted to the effective value range to obtain the diagonal loading factor that adaptively adjusts with the disturbance environment.

5. The radar beamforming method based on array signal processing according to claim 1, characterized in that, The sampling covariance matrix is ​​diagonally modified using a diagonal loading factor to obtain the modified covariance matrix. The modified covariance matrix is ​​then inverted, and the current beam weight vector is calculated using the target steering vector, including: The sampling covariance matrix is ​​diagonally loaded and corrected using a diagonal loading factor to obtain the corrected covariance matrix. Then, the corrected covariance matrix is ​​inverted to obtain the corrected covariance inverse matrix. Based on the minimum variance distortionless response criterion, the matrix product of the modified covariance inverse matrix and the target steering vector is calculated, and the matrix product result is normalized using the conjugate transpose of the target steering vector to obtain the current beam weight vector.

6. The radar beamforming method based on array signal processing according to claim 1, characterized in that, The adaptive iteration step size is calculated based on environmental factors, and the current beam weight vector is iteratively updated along the update gradient direction to obtain the updated beam weight vector, including: A nonlinear mapping relationship between environmental factors and iteration step size is constructed, and an adaptive iteration step size is calculated based on the environmental factors at the current moment. Based on the minimum mean square error criterion, the product of the sampling covariance matrix and the current beam weight vector is calculated, and the difference between the product and the target steering vector is used to obtain the instantaneous gradient of the current beam weight vector. The instantaneous gradient is weighted using an adaptive iteration step size. The difference between the current beam weight vector and the weighted instantaneous gradient is then calculated to obtain the updated beam weight vector.

7. The radar beamforming method based on array signal processing according to claim 1, characterized in that, An adaptive convergence threshold is generated based on environmental factors, including: Construct a positive correlation mapping function between environmental factors and convergence threshold, and calculate the adaptive convergence threshold based on the environmental factors at the current time. The positive correlation mapping function is configured to adaptively increase the convergence threshold as the complexity of the interference environment increases, so as to match the steady-state fluctuation amplitude of the weight vector under complex interference environment.

8. The radar beamforming method based on array signal processing according to claim 1, characterized in that, The beam weight vector output at the end of the iteration is subjected to norm constraint processing to output the final beam weight vector, including: Calculate the Euclidean norm of the beam weight vector output when the iteration stops, and use the Euclidean norm to normalize the beam weight vector output when the iteration stops, so as to obtain the final beam weight vector after norm constraint.

9. A radar beamforming system based on array signal processing, characterized in that, include: The environmental factor acquisition module is used to acquire the complex snapshot data matrix received by the radar array at the current moment, calculate the sampling covariance matrix based on the complex snapshot data matrix, and obtain the environmental factors based on the sampling covariance matrix. The diagonal loading correction module is used to perform diagonal loading correction on the sampling covariance matrix using a diagonal loading factor to obtain a corrected covariance matrix. The corrected covariance matrix is ​​then inverted, and the current beam weight vector is calculated in combination with the target guidance vector. The step size iterative update module is used to construct a minimum mean square error cost function for updating the beam weight vector based on the sampling covariance matrix, the current beam weight vector and the target steering vector, calculate the update gradient of the current beam weight vector, calculate the adaptive iterative step size according to environmental factors, and iteratively update the current beam weight vector along the update gradient direction to obtain the updated beam weight vector. The convergence comparison decision module is used to calculate the Euclidean distance between the current beam weight vector and the updated beam weight vector as the current iteration error. An adaptive convergence threshold is generated based on environmental factors. If the current iteration error is less than the adaptive convergence threshold, it is determined that convergence has been achieved and the iteration is stopped. Then, the operation of the beam weight vector output module is executed. Otherwise, the updated beam weight vector is used as the new current beam weight vector and the operation of the step size iteration update module is re-executed. The beam weight vector output module is used to perform norm constraint processing on the beam weight vector output when the iteration stops, and output the final beam weight vector.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the radar beamforming method based on array signal processing as described in any one of claims 1 to 8.

Images