Channel error correction method and device based on subarray correlation and global consistency

Abstract

The application provides a channel error correction method and device based on subarray correlation and global consistency, divides real-time array observation data of a radar to obtain a plurality of subarrays; extracts correlation features between channels in each subarray to obtain relative error estimation values of channels in each subarray relative to a local reference channel; constructs a global consistency constraint by using the relationship between different subarrays; solves global array channel error parameters according to the relative error estimation values and the global consistency constraint; and compensates the real-time array observation data of the radar on line according to the global array channel error parameters. In view of the problem that channel errors in the existing radar array system are difficult to be continuously, stably and low-pollution corrected on line, local correction information is obtained by using subarray internal correlation, and then all local results are unified and fused through the global consistency constraint, so that the robustness and scalability of error estimation are improved.

Description

Technical Field

[0001] This invention relates to the field of radar signal processing technology, and in particular to a channel error correction method and apparatus based on subarray correlation and global consistency. Background Technology

[0002] Millimeter-wave radar, with its advantages of all-weather operation, strong resistance to rain and fog, small size, and easy integration, has been widely used in fields such as intelligent driving, robot perception, security monitoring, industrial inspection, and environmental modeling. As millimeter-wave radar gradually evolves from traditional target detection to high-precision tasks such as high-resolution angle measurement, point cloud reconstruction, contour restoration, and spatial perception, the requirements for the amplitude and phase consistency of the array channels are constantly increasing.

[0003] For MIMO millimeter-wave radar, the system typically consists of a virtual array composed of multiple transmit channels and multiple receive channels. Ideally, each channel should have good amplitude consistency, phase consistency, and time delay consistency to ensure coherent superposition during array signal processing. However, in actual operation, factors such as temperature variations, device aging, power supply disturbances, RF link drift, packaging stress, chip differences, switching timing deviations, and sampling link mismatch often introduce varying degrees of channel errors into each transmit and receive channel.

[0004] The channel error typically manifests in the following forms:

[0005] 1. Inconsistent channel amplitude gain;

[0006] 2. Inconsistent channel phase offset;

[0007] 3. Differences in equivalent time delay between channels;

[0008] 4. Channel error drifts slowly over time;

[0009] 5. Some channels exhibit abnormal mutations in local time periods.

[0010] The aforementioned errors directly affect the beamforming, angle estimation, virtual array synthesis, multi-target resolution, and point cloud generation quality of millimeter-wave radar. Especially in high-resolution angle measurement and array imaging tasks, if channel errors are not effectively controlled, the following problems can easily occur:

[0011] 1. Beam main lobe shift, side lobe elevation;

[0012] 2. Increased estimation errors in azimuth and elevation angles;

[0013] 3. Array coherence gain decreases;

[0014] 4. Mismatch between the virtual array model and the real array model;

[0015] 5. Point cloud outline stretching and target shape distortion;

[0016] 6. Multi-frame coherent accumulation and tracking stability decrease.

[0017] In existing systems, most common channel calibration methods rely on the following means:

[0018] Firstly, a one-time offline calibration can be completed in a factory or laboratory environment using corner reflectors, standard targets, near-field probes, and loopback links.

[0019] Secondly, the initial amplitude and phase compensation parameters are obtained using the internal calibration module during the power-on initialization phase;

[0020] Third, periodic recalibration is performed under specific working conditions using static reference scenarios or manually deployed targets.

[0021] While the above methods can reduce channel errors to some extent, they have the following drawbacks:

[0022] First, offline calibration results are difficult to reflect the dynamic drift during system operation;

[0023] Secondly, online calibration methods that rely on dedicated reference targets or fixed scenarios have strong applicable conditions and are difficult to operate stably for a long time in complex dynamic scenarios.

[0024] Third, for large-scale MIMO arrays, cascaded arrays, or two-dimensional array systems, the number of channels is large, and if the calculation is entirely based on joint estimation of the entire array, the computational complexity is often high.

[0025] Fourth, the global unified calibration method is easily contaminated when local channel anomalies exist, leading to a decrease in overall estimation stability;

[0026] Fifth, many methods only focus on a single estimation and lack a continuous online correction mechanism for the long-term evolution of channel errors.

[0027] Further analysis of the array structure reveals that for most millimeter-wave radar arrays, the overall array can typically be naturally divided into several local subarrays. Each subarray contains a relatively small number of elements, with well-defined geometric relationships and more stable local signal correlation structures, making it easier to estimate relative channel errors. Simultaneously, the subarrays are not isolated from each other but are interconnected through shared targets, shared scene scattering, shared array geometric constraints, and shared global arrival structures. If local error estimations can be established within each subarray first, and then the results from each subarray are stitched together and unified through global consistency constraints, it is possible to achieve online correction of channel errors in complex array systems without heavily relying on external dedicated reference targets.

[0028] While some existing technologies utilize adjacent channel correlation, sub-aperture decomposition, or local phase consistency for auxiliary compensation, most remain at the level of local empirical correction. There is a lack of a systematic online channel error correction method that "uses subarray correlation as the basis for local observation, global consistency as the unified constraint framework, and online recursion as the operating mechanism."

[0029] Therefore, there is an urgent need for a channel error correction method and apparatus based on subarray correlation and global consistency to improve the above problems. Summary of the Invention

[0030] The purpose of this invention is to provide a channel error correction method and apparatus based on subarray correlation and global consistency, which can improve the robustness and scalability of error estimation.

[0031] In a first aspect, the present invention provides a channel error correction method based on subarray correlation and global consistency, comprising the steps of: dividing the real-time array observation data of a radar into several subarrays; extracting the correlation features between channels within each subarray to obtain relative error estimates of each channel within the subarray relative to a local reference channel; constructing global consistency constraints using the relationships between different subarrays; solving for the full array channel error parameters based on the relative error estimates and the global consistency constraints; and performing online compensation on the real-time array observation data of the radar based on the full array channel error parameters.

[0032] Optionally, before dividing the radar real-time array observation data, the following may be included: establishing an array observation model containing channel error terms, defining the mapping relationship between channel complex error parameters and covariance matrix; and / or dividing the radar real-time array observation data in one of the following ways: (1) dividing by physically adjacent array elements; (2) dividing by transmitting subarray and receiving subarray; (3) dividing by continuous equally spaced array elements in the virtual array; (4) dividing by rows, columns or local blocks in the two-dimensional array; (5) grouping by the channel to which the cascaded chip belongs; and (6) constructing multiple overlapping subarrays by overlapping sliding window method.

[0033] Optionally, extracting the correlation features between channels within each subarray to obtain the relative error estimate of each channel within the subarray relative to the local reference channel includes: extracting the correlation features between channels within each subarray and establishing a relative error estimation model within the subarray based on the correlation features; obtaining the relative error value of each channel within the subarray relative to the local reference channel by minimizing the error estimation model and the reference structure matrix of the subarray; and / or obtaining the reference structure matrix by one of the following methods: (1) generation of an ideal array model; (2) estimation results of historical stable periods; (3) recovery results of low-rank approximation or principal eigenvalue decomposition within the subarray; (4) estimation results of neighboring subarray fusion.

[0034] Optionally, the correlation features include one or more of the following: adjacent channel multiple correlation coefficient, subarray principal eigenvector phase structure, subarray steering vector fitting residual, subarray normalized cross-spectral ratio, and relative phase and relative amplitude between the reference channel and other channels within the subarray.

[0035] Optionally, constructing global consistency constraints using the relationships between different subarrays includes: obtaining the full array error vector by utilizing the shared channels, geometric continuity, and common scene structure relationships between different subarrays; constructing global consistency constraints that satisfy the prior structure based on the error vector; and / or the prior structure is one or more of the following: an approximate Toeplitz structure, a low-rank signal subspace structure, a block Toeplitz structure corresponding to a two-dimensional array, a conjugate symmetric structure corresponding to a symmetric array, and a consistency structure of a historical reference model.

[0036] Optionally, the method for obtaining the full array channel error parameters is one of step-by-step stitching solution, least squares joint solution, or graph optimization solution; and / or online compensation of the radar's real-time array observation data based on the full array channel error parameters includes: continuously correcting the full array channel error parameters using a sliding window or recursive update method to obtain error estimates; converting the error estimates into correction factors, constructing a correction matrix based on the correction factors; and performing direct data domain compensation or covariance domain compensation of the radar's real-time array observation data based on the correction matrix.

[0037] Secondly, the present invention provides a channel error correction device based on subarray correlation and global consistency, the device comprising modules / units for performing any of the possible design methods described in the first aspect above. These modules / units can be implemented in hardware or by hardware executing corresponding software.

[0038] Thirdly, the present invention provides an electronic device including a memory and a processor, wherein the memory stores a program executable on the processor, and when the program is executed by the processor, the electronic device implements a method for performing any of the possible designs described above.

[0039] Fourthly, the present invention provides a readable storage medium storing a program, which, when executed, implements a method of any possible design of any of the above aspects.

[0040] Fifthly, the present invention provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0041] The beneficial effects of the method of this invention are as follows: Real-time radar array observation data is divided into several subarrays; correlation features between channels within each subarray are extracted to obtain relative error estimates of each subarray's channels relative to a local reference channel; global consistency constraints are constructed using the relationships between different subarrays; the full array channel error parameters are obtained based on the relative error estimates and the global consistency constraints; and online compensation is performed on the real-time radar array observation data based on the full array channel error parameters. Addressing the problem that channel errors in existing radar array systems are difficult to continuously, stably, and with low pollution during operation, this invention first utilizes the correlation within subarrays to obtain local correction information, and then unifies and fuses all local results through global consistency constraints. This improves the robustness and scalability of error estimation, enabling continuous updates of channel error parameters under conditions of temperature drift, device aging, local anomalies, and dynamic scene changes. The online recursive correction mechanism provides a stable basis for subsequent beamforming, angle measurement, imaging, and point cloud generation. Attached Figure Description

[0042] Figure 1 A flowchart illustrating a channel error correction method based on subarray correlation and global consistency provided in an embodiment of the present invention;

[0043] Figure 2 A schematic diagram of a channel error correction device based on subarray correlation and global consistency provided in an embodiment of the present invention;

[0044] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0045] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions in the embodiments 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 inventive effort are within the scope of protection of this invention. Unless otherwise defined, the technical or scientific terms used herein should have the ordinary meaning understood by those skilled in the art. The terms "comprising" and similar expressions used herein mean that the element or object preceding the word covers the element or object listed following the word and its equivalents, but do not exclude other elements or objects.

[0046] The technical solutions of the embodiments of the present invention will be described below with reference to the accompanying drawings. In the description of the embodiments of the present invention, the terminology used in the following embodiments is for the purpose of describing specific embodiments only and is not intended to limit the present invention. The singular expressions “a,” “the,” “the,” and “this” are intended to also include expressions such as “one or more,” unless the context clearly indicates otherwise. It should also be understood that in the following embodiments of the present invention, “at least one” and “one or more” refer to one or more (including two). The term “and / or” is used to describe the relationship between related objects, indicating that three relationships can exist; for example, A and / or B can represent: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. The character “ / ” generally indicates that the preceding and following related objects are in an “or” relationship.

[0047] References to "one embodiment" or "some embodiments" in this specification mean that a particular feature, structure, or characteristic described in connection with that embodiment is included in one or more embodiments of the invention. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," and "in still other embodiments" appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including, but not limited to," unless otherwise specifically emphasized. The term "connection" includes both direct and indirect connections, unless otherwise stated. "First" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated.

[0048] In embodiments of the present invention, "exemplarily" or "for example" are used to indicate that they are examples, illustrations, or descriptions. Any embodiment or design described as "exemplarily" or "for example" in embodiments of the present invention should not be construed as being more preferred or advantageous than other embodiments or design solutions. Rather, the use of "exemplarily" or "for example" is intended to present the relevant concepts in a specific manner.

[0049] like Figure 1 As shown, this invention provides a channel error correction method based on subarray correlation and global consistency, including the following steps:

[0050] S101 divides the real-time radar array observation data into several subarrays.

[0051] In some embodiments, before dividing the real-time radar array observation data, the method further includes: establishing an array observation model containing channel error terms and defining the mapping relationship between channel complex error parameters and covariance matrices.

[0052] In other embodiments, the radar real-time array observation data is divided in one of the following ways: (1) by physically adjacent array elements; (2) by transmitting subarray and receiving subarray; (3) by continuous equally spaced array elements in the virtual array; (4) by rows, columns or local blocks in the two-dimensional array; (5) by grouping according to the channel to which the cascaded chip belongs; (6) by constructing multiple overlapping subarrays in an overlapping sliding window manner.

[0053] S102, extract the correlation features between channels within each subarray to obtain the relative error estimate of each channel within the subarray relative to the local reference channel.

[0054] In some embodiments, extracting the correlation features between channels within each subarray to obtain the relative error estimate of each channel within the subarray relative to the local reference channel includes: extracting the correlation features between channels within each subarray and establishing a relative error estimation model within the subarray based on the correlation features; and obtaining the relative error value of each channel within the subarray relative to the local reference channel by minimizing the error estimation model and the reference structure matrix of the subarray.

[0055] In some specific embodiments, the reference structure matrix is ​​obtained by one of the following methods: (1) generation of an ideal array model; (2) estimation results of historical stable periods; (3) recovery results of low-rank approximation or principal eigenvalue decomposition within the subarray; (4) estimation results of neighboring subarray fusion.

[0056] In other embodiments, the correlation features include one or more of the following: adjacent channel complex correlation coefficient, subarray principal feature vector phase structure, subarray steering vector fitting residual, subarray normalized cross-spectral ratio, and relative phase and relative amplitude between the reference channel and other channels within the subarray.

[0057] S103 utilizes the relationships between different subarrays to construct global consistency constraints.

[0058] In some embodiments, constructing global consistency constraints by utilizing the relationships between different subarrays includes: obtaining a full array error vector by utilizing the shared channels, geometric continuity, and common scene structure relationships between different subarrays; and constructing global consistency constraints that satisfy the prior structure based on the error vector.

[0059] In other embodiments, the prior structure is one or more of the following: an approximate Toeplitz structure, a low-rank signal subspace structure, a block Toeplitz structure corresponding to a two-dimensional array, a conjugate symmetric structure corresponding to a symmetric array, and a consistency structure of a historical reference model.

[0060] S104. Based on the relative error estimate and global consistency constraints, the full array channel error parameters are obtained.

[0061] In some embodiments, the full array channel error parameters are obtained by one of the following methods: step-by-step splicing solution, least squares joint solution, or graph optimization solution.

[0062] S105, perform online compensation on the real-time array observation data of the radar according to the full array channel error parameters.

[0063] In some embodiments, online compensation of the radar's real-time array observation data based on the full array channel error parameters includes: continuously correcting the full array channel error parameters using a sliding window or recursive update method to obtain an error estimate; converting the error estimate into a correction factor, constructing a correction matrix based on the correction factor; and performing direct data domain compensation or covariance domain compensation on the radar's real-time array observation data based on the correction matrix.

[0064] The advantages of this invention are that, for a multi-channel radar system, the overall array is first divided into several local subarrays based on the array geometry or virtual array structure. Within each subarray, a relative error model is established using the correlation, coherence, or local covariance structure between channels. Then, by constructing global consistency constraints, the local error estimates in multiple subarrays are aligned, stitched together, and uniformly solved to obtain the channel error parameters of the entire array. Finally, online correction is achieved through a sliding window or recursive update method.

[0065] To facilitate understanding, this embodiment further elaborates on the specific implementation process of the above method in conjunction with a specific application scenario. Taking MIMO millimeter-wave radar as an example, the specific steps include:

[0066] Step 1: Establish an array observation model including channel error terms

[0067] A set of A millimeter-wave radar array system with multiple channels, wherein the array can be a receiving array, a transmitting array, a virtual array, or an equivalent array of any combination thereof. Let the... The equivalent complex error of each channel is expressed as:

[0068]

[0069] in, For the first Amplitude error of each channel; For the first Phase error of each channel.

[0070] If frequency-dependent errors or time delay errors are further considered, it can be expanded to:

[0071]

[0072] in, For the first The equivalent delay error of each channel.

[0073] Suppose the ideal array observation vector of the system at a certain distance cell, a certain Doppler cell, a certain snapshot, or a certain spatial frequency is:

[0074]

[0075] in, For the target angle parameter set The guiding matrix formed; Let the target complex envelope vector be denoted as . This is the noise vector.

[0076] After considering channel errors, the actual observations are as follows:

[0077]

[0078] in:

[0079]

[0080] This is the channel error diagonal matrix.

[0081] The corresponding covariance matrix is ​​expressed as:

[0082]

[0083] in, It is the ideal covariance matrix.

[0084] The objective of this invention is based on observation. and its local subarray structure, online estimation The error parameters for each channel included in it.

[0085] Step 2: Divide the overall array into subarrays.

[0086] To improve the observability and robustness of error estimation, this invention first divides the real-time radar array observation data into several subarrays. Assume the overall array contains... Each channel is divided into Each unit, remember the first The set of subarray channel indices is:

[0087]

[0088] in, For the first The number of channels contained in each subarray and It is a positive integer.

[0089] The subarray partitioning method can be determined according to the specific structure of the system, including but not limited to the following methods:

[0090] (1) Divided according to physically adjacent array elements;

[0091] (2) Divided into transmitting subarrays and receiving subarrays;

[0092] (3) Divide according to the continuous equally spaced array elements in the virtual array;

[0093] (4) Divide according to rows, columns or local blocks in a two-dimensional array;

[0094] (5) Group the cascaded chips according to their respective channels;

[0095] (6) Construct multiple overlapping subarrays using an overlapping sliding window method.

[0096] Preferably, the present invention employs an overlapping subarray partitioning method, enabling partial shared channels between different subarrays, thereby facilitating the subsequent establishment of global consistency constraints. For example, the following overlapping subarrays can be constructed for a linear array:

[0097]

[0098]

[0099]

[0100]

[0101] This partitioning method naturally creates a bridging relationship between local estimation results, which is beneficial for piecing together local relative errors into a global error solution.

[0102] Step 3: Extract the correlation features between channels within each subarray

[0103] For each subarray From the overall observation vector Extract the corresponding subarray observation vector:

[0104]

[0105] in, Choose a matrix for the submatrix.

[0106] Therefore, the first... The sample covariance matrix of each subarray:

[0107]

[0108] in, This represents the number of samples within the sliding window.

[0109] Under ideal conditions, the phase relationship, amplitude relationship, and correlation structure between different channels within a subarray are mainly determined by the target incident direction, array geometry, and propagation environment. If the channel errors within a subarray are small or the local geometry is stable, the correlation matrix within the subarray will exhibit strong structural regularity. If errors exist in some channels, this structural regularity will deviate.

[0110] Let the first The ideal covariance matrix of each subarray is:

[0111]

[0112] The actual submatrix covariance matrix satisfies:

[0113]

[0114] in, For the sub-array Error parameter vectors for each channel; This is the corresponding diagonal error matrix.

[0115] To enhance observability, this invention can extract the following features from the submatrix covariance matrix:

[0116] 1. Correlation coefficient between adjacent channels;

[0117] 2. Phase structure of the principal eigenvectors of the subarray;

[0118] 3. Subarray steering vector fitting residuals;

[0119] 4. Normalized cross-spectral ratio within the subarray;

[0120] 5. The relative phase and relative amplitude of the reference channel and the other channels within the subarray.

[0121] For example, if the first channel within the subarray is selected as a local reference, a relative correlation feature can be defined:

[0122]

[0123] in, Indicates the first The first subarray sample covariance matrix Line number Column elements. This quantity reflects the complex correlation between the reference channel and the other channels within the subarray, and can serve as the basis for estimating local relative errors.

[0124] Step 4: Establish a relative error estimation model within the subarray based on the aforementioned correlation characteristics.

[0125] Within each subarray, this invention does not directly solve for the global absolute error, but instead prioritizes estimating the relative error values ​​of each channel within the subarray relative to the local reference channel.

[0126] Set up a sub-array The local reference channel index is Then any channel in the subarray The relative error is defined as:

[0127]

[0128] Within a subarray, if the dominant scattering of the scene is relatively stable within a short time window, then the observations of each channel in the subarray can be considered to satisfy an approximately common incident structure. In this case, the relative error can be estimated from the correlation ratio. For example, it can be constructed based on a reference channel:

[0129]

[0130] in, It is a compensation factor obtained by correction of array geometry, local guidance relationship or local ideal structure.

[0131] In a more general case, we can construct an optimization problem within the subarray:

[0132]

[0133] in, For the first The reference structure matrix of each subarray can be obtained in the following way:

[0134] (1) Generation of ideal array model;

[0135] (2) Estimation results of historical stable periods;

[0136] (3) The results of low-rank approximation or principal eigenvalue decomposition recovery within the subarray;

[0137] (4) Estimation results of neighboring subarray fusion.

[0138] By minimizing This allows us to obtain the relative error values ​​of each subarray's internal channels relative to the local reference channel:

[0139]

[0140] It should be emphasized that what is obtained at this point is mainly the local relative solution of the subarray. There may still be differences in the overall phase offset, the overall amplitude scale, or the reference normalization between different subarrays. Therefore, it is necessary to unify them through a global consistency step.

[0141] Step 5: Construct global consistency constraints using the relationships between different subarrays.

[0142] Since different subarrays often share channels, geometric continuity, and common scene structures, these relationships can be used to unify the local error results of multiple subarrays into the same global coordinate system.

[0143] Let the full array error vector be:

[0144]

[0145] For any subarray Its local error vector should satisfy:

[0146]

[0147] If the first Sub-arrays and the first There is a shared set of channels among the subarrays:

[0148]

[0149] Ideally, the error values ​​estimated by the shared channel in both subarrays should be consistent. Therefore, a cross-subarray consistency constraint can be constructed:

[0150]

[0151] Accordingly, a global consistency cost function is defined:

[0152]

[0153] in This is the consistency weighting coefficient.

[0154] In further implementation, global array structural consistency constraints can be introduced. For example, for the corrected full array covariance matrix:

[0155]

[0156] It is required to satisfy some kind of a priori structure as much as possible, such as:

[0157] 1. Approximates Toeplitz structure;

[0158] 2. Low-rank signal subspace structure;

[0159] 3. Block Toeplitz structure in two-dimensional array;

[0160] 4. Conjugate symmetric structures under symmetric arrays;

[0161] 5. Consistent structure of historical reference model.

[0162] For example, a global structural consistency term can be defined:

[0163]

[0164] in, This represents a mapping operator that projects a matrix onto a target structure set.

[0165] Finally, the global joint optimization objective can be written as:

[0166]

[0167] in, These are the weight parameters.

[0168] Step 6: Jointly solve for the error parameters of the entire array channels

[0169] After constructing the local subarray error model and global consistency constraints, this invention jointly solves the error parameters of the entire array channels.

[0170] Method 1: Solving by step-by-step assembly

[0171] First, obtain the relative solutions of local errors within each subarray, and then recursively stitch them together using shared channels:

[0172]

[0173] In this way, local relative errors can be gradually propagated to the entire array, and finally a reference channel is selected for normalization.

[0174] Method 2: Joint Least Squares Solution

[0175] By rewriting all local constraints and global consistency constraints of the submatrices into a unified system of equations, the following least squares problem is constructed:

[0176]

[0177] This problem can be solved using alternating least squares, conjugate gradient, Gauss-Newton, ADMM, or recursive least squares.

[0178] Method 3: Graph Optimization Solution

[0179] Treating each channel as a node in a graph, and the relative error relationships within subarrays and the shared consistency relationships across subarrays as edge constraints, a channel error graph model can be constructed. The global consistency error can then be solved using graph optimization methods.

[0180]

[0181] in, It is the set of graph edges; This represents the relative error ratio obtained from the subarray correlation estimation. The corresponding edge confidence weights.

[0182] This graph optimization method is particularly suitable for robust solutions in scenarios with overlapping subarrays and when local abnormal channels exist.

[0183] Step 7: Construct an online recursive update and anomaly suppression mechanism

[0184] To achieve online calibration, this invention uses a sliding window or recursive update method to continuously correct the channel error parameters.

[0185] Set time The channel error estimation result is Then, the following recursive update can be used:

[0186]

[0187] in, The error parameters are the newly estimated parameters within the current window; To update the step size or forgetting factor.

[0188] To prevent local anomalies from contaminating the global solution, this invention also constructs an anomaly suppression mechanism. A local residual is defined for each subarray:

[0189]

[0190] like If the threshold is exceeded, it indicates that the current estimate of the subarray may be affected by anomalous scattering, low signal-to-noise ratio, or local fault. In this case, the following measures can be taken:

[0191] 1. Reduce the weight of this submatrix in the global solution;

[0192] 2. Do not update the local solution corresponding to this subarray for now;

[0193] 3. Maintain the current correction using only historical stable solutions;

[0194] 4. Freeze or isolate suspected abnormal channels.

[0195] Furthermore, define the drift amount for a single channel:

[0196]

[0197] in, This is a historical benchmark or stable mean. If... If the threshold is exceeded, the corresponding channel drift alarm or abnormal alarm will be triggered.

[0198] Step 8: Output channel correction parameters and perform online compensation.

[0199] After obtaining the full array channel error estimation results, this invention outputs the correction factor for each channel:

[0200]

[0201] And construct the correction matrix:

[0202]

[0203] Perform online compensation on real-time observation data:

[0204]

[0205] Alternatively, compensation can be performed on the covariance matrix:

[0206]

[0207] The corrected results can be further used for:

[0208] Beamforming;

[0209] Azimuth and elevation angle estimation;

[0210] High-resolution array processing;

[0211] MIMO virtual array compensation;

[0212] Point cloud generation and fine imaging;

[0213] Multi-frame coherent accumulation.

[0214] In addition, the present invention can also output the following monitoring information:

[0215] Current amplitude and phase error estimates for each channel;

[0216] Local fitting residuals of each subarray;

[0217] Globally consistent residuals;

[0218] Abnormal subarray table;

[0219] List of abnormal channels;

[0220] It is recommended to update, freeze, or remove channel status information.

[0221] The key aspects of the embodiments of the present invention are as follows:

[0222] 1. Decompose the full array calibration problem into "local observability of subarrays + global unified constraints".

[0223] This invention does not directly perform a single global error solution on the entire array. Instead, it first extracts relative error information within a local subarray and then completes unified correction through global consistency constraints, thereby reducing the difficulty of the solution and improving robustness.

[0224] 2. Using subarray correlation as the basis for error observation

[0225] This invention utilizes the correlation, coherence, and local covariance structure between channels within the subarray as sources of error observation, without heavily relying on external dedicated reference targets, thus enabling the system to have stronger online availability in actual operating scenarios.

[0226] 3. Establish bridging relationships using overlapping subarrays and shared channels.

[0227] By dividing the array into overlapping subarrays, shared channels exist between different subarrays, allowing local estimation results to be progressively pieced together into a global error solution. This mechanism makes the method particularly suitable for large-scale arrays, virtual arrays, and cascaded array systems.

[0228] 4. Introduce global structural consistency constraints

[0229] This invention not only requires that the local estimates of each subarray be self-consistent, but also requires that the statistical structure of the corrected full array satisfy global priors such as Toeplitz property, low rank property, symmetry or historical consistency, thereby improving the physical rationality and stability of the global solution.

[0230] 5. Possesses online recursive update capability

[0231] This invention achieves continuous online correction of channel errors through sliding window, recursive update and forgetting factor mechanisms, and can adapt to slow-changing disturbances such as temperature drift, aging and changes in operating conditions.

[0232] 6. Possesses the ability to suppress anomalies and isolate localized failures.

[0233] This invention can identify abnormal subarrays and abnormal channels by monitoring subarray residuals and channel drift, thereby avoiding local failures from contaminating the global correction results and enhancing the long-term reliability of the system.

[0234] The advantages of the embodiments of the present invention are as follows:

[0235] 1. Improve the observability and stability of online calibration.

[0236] By introducing subarray partitioning, this invention transforms the complex full array error problem into multiple locally observable problems, and combines global consistency to achieve a unified solution, resulting in a more stable overall estimate.

[0237] 2. Reduce the difficulty of online calibration for large-scale array systems

[0238] Compared with methods that directly perform high-dimensional joint estimation across the entire array, this invention is more suitable for MIMO arrays, virtual arrays, and cascaded array systems with a large number of channels.

[0239] 3. Enhance adaptability to local anomalies and dynamic drift.

[0240] Because this invention employs a strategy that combines local estimation and global fusion, when a local subarray is affected by anomalies, its impact on the overall result can be mitigated through methods such as weight suppression and anomaly isolation.

[0241] 4. Improve the accuracy of subsequent angle measurement, imaging, and point cloud processing.

[0242] The array data corrected online by this invention has better amplitude and phase consistency, which is beneficial to improving the performance of beamforming, angle measurement, point cloud reconstruction and array imaging algorithms.

[0243] 5. Suitable for long-term online channel maintenance

[0244] This invention not only provides channel correction parameters, but also continuously outputs drift and residual information, making it suitable for health monitoring and predictive maintenance in long-term deployment systems.

[0245] 6. It has good scalability.

[0246] This invention can be extended to linear arrays, area arrays, MIMO virtual arrays, multi-chip cascaded arrays, and other radar array systems that can be locally subdivided into subarrays, and has strong engineering promotion value and patent layout value.

[0247] The following are optional embodiments and extensions of the method of the present invention.

[0248] Online calibration based on receiver subarray

[0249] In some embodiments, only the receiving array is subdivided and online channel correction is performed, which is suitable for systems where receiving link error is the primary concern.

[0250] Online calibration based on virtual array

[0251] In some embodiments, the MIMO virtual array can be directly divided into subarrays and its consistency corrected, which is applicable to virtual channel error compensation in TDMA-MIMO systems.

[0252] Block subarray partitioning based on two-dimensional surface array

[0253] For two-dimensional area arrays, local block subarrays, row and column subarrays, or overlapping window subarrays can be used for local error modeling, thereby adapting to two-dimensional angle measurement and three-dimensional imaging scenarios.

[0254] Introducing a robust weight function

[0255] In the global joint solution, Huber weight function, Tukey weight function or other robust weighting mechanisms can be used for subarrays with large local residuals to improve the solution stability under abnormal conditions.

[0256] Linked with online compensation closed loop

[0257] The channel correction parameters output by this invention can be directly used in the array online compensation module, and combined with the correction residuals, they are fed back to the error estimation module to form a closed-loop structure of "estimation-compensation-verification-update".

[0258] like Figure 2 As shown, based on the above method, the present invention provides a channel error correction device based on subarray correlation and global consistency, comprising: a partitioning unit 201 for partitioning the real-time array observation data of the radar into several subarrays; an extraction unit 202 for extracting the correlation features between channels within each subarray to obtain relative error estimates of each channel within the subarray relative to a local reference channel; a construction unit 203 for constructing global consistency constraints using the relationships between different subarrays; a calculation unit 204 for solving for the full array channel error parameters based on the relative error estimates and the global consistency constraints; and a compensation unit 205 for performing online compensation on the real-time array observation data of the radar based on the full array channel error parameters.

[0259] It should be understood that all relevant content of each step involved in the above method embodiments can be referenced to the functional description of the corresponding functional module, and will not be repeated here. Furthermore, the use of suffixes such as "module," "component," or "unit" to represent elements is merely for the purpose of illustrative purposes and has no specific meaning in itself. Therefore, "module," "component," or "unit" can be used interchangeably. Terminals can be implemented in various forms. For example, the terminals described in this invention may include mobile terminals such as mobile phones, tablets, laptops, handheld computers, personal digital assistants (PDAs), portable media players (PMPs), navigation devices, wearable devices, smart bracelets, pedometers, etc., as well as fixed terminals such as digital TVs and desktop computers. The following description will use mobile terminals as examples; those skilled in the art will understand that, in addition to elements specifically designed for mobile purposes, the construction according to embodiments of the present invention can also be applied to fixed-type terminals.

[0260] In other embodiments of the present invention, an electronic device 300 is disclosed, such as... Figure 3 As shown, the device may include: one or more processors 301; memory 302; display 303; one or more application programs (not shown); and one or more computer programs 304. These devices can be connected via one or more communication buses 305. The one or more computer programs 304 are stored in the memory 302 and configured to be executed by the one or more processors 301. The one or more computer programs 304 include instructions that can be used to perform actions such as... Figure 1 Each step in the corresponding embodiment.

[0261] Processor 301 can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor.

[0262] The memory 302 can be an internal storage unit of the electronic device 300, such as a hard disk or RAM of the electronic device 300. The memory 302 can also be an external storage device of the electronic device 300, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, or FlashCard equipped on the electronic device 300. Furthermore, the memory 302 can include both internal and external storage units of the electronic device 300. The memory 302 is used to store computer programs and other programs and data required by the electronic device. The memory 302 can also be used to temporarily store data that has been output or will be output.

[0263] The computer program 304 can be divided into one or more modules / units. The one or more modules / units can be a series of computer program instruction segments that can perform a specific function. The instruction segments are used to describe the execution process of the computer program 304 in the electronic device 300.

[0264] In addition to the above-described structure, those skilled in the art will understand that Figure 3 This is merely an example of electronic device 300 and does not constitute a limitation on electronic device 300. Electronic device 300 may include more or fewer components than shown, or combine certain components, or different components. For example, electronic device may also include input / output devices, network access devices, buses, etc.

[0265] Those skilled in the art will understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the functions described above can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this invention. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0266] Based on the above embodiments, the present invention also discloses a computer-readable storage medium having at least one computer program stored thereon, wherein the computer program, when executed by a processor, implements the methods described in the foregoing embodiments.

[0267] Those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program instructing a processor. The program can be stored in a computer-readable storage medium, which is a non-transitory medium, such as random access memory, read-only memory, flash memory, hard disk, solid-state drive, magnetic tape, floppy disk, optical disk, and any combination thereof. The storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. This available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., digital video disc (DVD)), or a semiconductor medium (e.g., solid-state drive (SSD)).

[0268] The descriptions of the processes or structures corresponding to the above figures each have their own emphasis. For parts of a process or structure that are not described in detail, please refer to the relevant descriptions of other processes or structures.

[0269] Although the embodiments of the present invention have been described in detail above, it will be apparent to those skilled in the art that various modifications and variations can be made to these embodiments. The above descriptions are merely embodiments of the present invention and do not limit the patent scope of the present invention. However, it should be understood that such modifications and variations fall within the scope and spirit of the present invention. Moreover, the present invention described herein may have other embodiments and can be implemented or realized in various ways. All equivalent transformations made based on the description and drawings of the present invention, or direct or indirect applications in related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A channel error correction method based on subarray correlation and global consistency, characterized in that, Including the following steps: The real-time radar array observation data is divided into several subarrays; The correlation features between channels within each subarray are extracted to obtain the relative error estimate of each channel within the subarray relative to the local reference channel; Construct global consistency constraints by utilizing the relationships between different subarrays; Based on the relative error estimate and global consistency constraints, the full array channel error parameters are obtained. The real-time array observation data of the radar is compensated online based on the full array channel error parameters.

2. The method according to claim 1, characterized in that, Before dividing the radar real-time array observation data, the following steps are also included: Establish an array observation model with channel error terms, and define the mapping relationship between channel complex error parameters and covariance matrix; And / or the radar real-time array observation data can be divided in one of the following ways: (1) Divided according to physically adjacent array elements; (2) Divided into transmitting subarrays and receiving subarrays; (3) Divide according to the continuous equally spaced array elements in the virtual array; (4) Divide according to rows, columns or local blocks in a two-dimensional array; (5) Group the cascaded chips according to their respective channels; (6) Construct multiple overlapping subarrays using an overlapping sliding window method.

3. The method according to claim 1, characterized in that, The correlation features between channels within each subarray are extracted to obtain the relative error estimates of each subarray's channels relative to the local reference channel, including: Extract the correlation features between channels within each subarray, and establish a relative error estimation model within the subarray based on the correlation features; Based on the error estimation model and the reference structure matrix of the subarray, the relative error value of each subarray's internal channel relative to the local reference channel is obtained by minimization. And / or the reference structure matrix is ​​obtained in one of the following ways: (1) Generation of ideal array model; (2) Estimation results of historical stable periods; (3) The results of low-rank approximation or principal eigenvalue decomposition recovery within the subarray; (4) Estimation results of neighboring subarray fusion.

4. The method according to claim 1, characterized in that, The correlation features include one or more of the following: adjacent channel multiple correlation coefficient, subarray principal feature vector phase structure, subarray steering vector fitting residual, subarray normalized cross spectrum ratio, and relative phase and relative amplitude between the reference channel and other channels within the subarray.

5. The method according to claim 1, characterized in that, Constructing global consistency constraints using the relationships between different subarrays includes: By utilizing the shared channels, geometric continuity, and common scene structure relationships among different subarrays, the full array error vector is obtained; Construct a global consistency constraint that satisfies the prior structure based on the error vector; And / or the prior structure is one or more of the following: an approximate Toeplitz structure, a low-rank signal subspace structure, a block Toeplitz structure corresponding to a two-dimensional array, a conjugate symmetric structure corresponding to a symmetric array, and a consistency structure of a historical reference model.

6. The method according to any one of claims 1-5, characterized in that, The method to obtain the error parameters of the entire array channels is one of the following: step-by-step splicing solution, least squares joint solution, or graph optimization solution. And / or online compensation of the radar's real-time array observation data based on the full array channel error parameters includes: The error parameters of the entire array channel are continuously corrected using a sliding window or recursive update method to obtain the error estimate. The error estimate is converted into a correction factor, and a correction matrix is ​​constructed based on the correction factor. The real-time array observation data of the radar is directly compensated in the data domain or compensated in the covariance domain based on the correction matrix.

7. A channel error correction device based on subarray correlation and global consistency, used in the method of any one of claims 1-6, characterized in that, include: The division unit is used to divide the real-time radar array observation data into several subarrays; The extraction unit is used to extract the correlation features between channels within each subarray to obtain the relative error estimate of each channel within the subarray relative to the local reference channel. Building units are used to construct global consistency constraints by utilizing the relationships between different subarrays; The calculation unit is used to solve for the full array channel error parameters based on the relative error estimate and the global consistency constraint. The compensation unit is used to perform online compensation on the real-time array observation data of the radar based on the full array channel error parameters.

8. An electronic device, characterized in that, The device includes a memory and a processor, wherein the memory stores a program that runs on the processor, and when the program is executed by the processor, causes the electronic device to perform the method of any one of claims 1-6.

9. A readable storage medium storing a program, characterized in that, When the program is executed, it implements the method of any one of claims 1-6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1-6.

Images