Fast mvdr beamforming method and device based on LDLH decomposition

Abstract

The application belongs to the technical field of adaptive beam forming, and discloses a fast MVDR beam forming method and device based on LDLH decomposition. The method comprises the following steps: acquiring array noise and interference data and calculating a covariance matrix, performing LDLH decomposition on the covariance matrix to obtain a unit lower triangular matrix L and a real diagonal matrix D, solving the inverse matrices of L and D respectively, combining the inverse matrices to obtain the inverse of the covariance matrix, calculating the MVDR weight coefficient in combination with the expected signal steering vector to complete the beam forming. The above method utilizes the characteristic of the Hermitian matrix of the covariance matrix R, adopts LDLH decomposition for inversion, converts the inversion process into the operation on the triangular matrix and the diagonal matrix, naturally has better numerical stability, provides a regular calculation structure for hardware optimization, greatly reduces the operation complexity of the inversion of the covariance matrix, significantly improves the operation speed, has low hardware implementation resource occupation, and is especially suitable for array signal processing scenes with high real-time requirements.

Description

Technical Field

[0001] This invention relates to the field of adaptive beamforming technology, and in particular to a fast MVDR beamforming method and apparatus based on LDLH decomposition. Background Technology

[0002] MVDR (Minimum Variance Distortionless Response) beamforming is a core technology in array signal processing for suppressing interference and improving the signal-to-noise ratio of the desired signal. Its core step requires inverting the covariance matrix. Traditional MVDR beamforming typically uses Cholesky decomposition, singular value decomposition, or direct matrix inversion to calculate the inverse of the covariance matrix, resulting in a computational complexity as high as O(N³). In embedded / hardware implementations with large array sizes and high real-time requirements, this approach suffers from high computational latency and high hardware resource consumption, making it difficult to meet the application requirements of low power consumption and high real-time performance.

[0003] The above content is only used to help understand the technical solution of the present invention and does not represent an admission that the above content is prior art. Summary of the Invention

[0004] The main objective of this invention is to provide a fast MVDR beamforming method and apparatus based on LDLH decomposition, aiming to solve the technical problems of high complexity, poor real-time performance, and high hardware resource consumption in traditional MVDR beamforming.

[0005] To achieve the above objectives, the present invention provides a fast MVDR beamforming method based on LDLH decomposition, the fast MVDR beamforming method based on LDLH decomposition comprising the following steps: Obtain the column vectors of noise and interference data from the array, and calculate the covariance matrix R=XX based on the column vectors of noise and interference data. H Where X is the column vector of noise and interference data, H This indicates the conjugate transpose, and the covariance matrix R is a positive Hermitian definite matrix; Performing LDLH decomposition on the covariance matrix R yields a unit lower triangular matrix L and a real diagonal matrix D, where R = L × D × L H ; Solve for the inverse matrix L of the unit lower triangular matrix L. 1 and the inverse matrix D of the real diagonal matrix D 1 ; According to the inverse matrix L of the unit lower triangular matrix L 1 and the inverse matrix D of the real diagonal matrix D 1The inverse matrix R of the covariance matrix is ​​calculated. -1 ; The weighting coefficients of the MVDR beamformer are calculated based on the inverse of the covariance matrix and the steering vector of the desired signal direction, and the MVDR beamformer is completed based on the weighting coefficients.

[0006] In one embodiment, performing LDLH decomposition on the covariance matrix R includes: Iterate through j from 1 to n column by column, where n is the matrix dimension; Calculate the diagonal elements, ensuring that D[j,j] is a real number. The calculation formula is as follows: D[j,j]=REAL(R[j,j] And ensure that D[j,j] is a real number; Calculate the elements in the lower triangular portion of the j-th column of L, for each row i∈[j+1,n] 1]: L[i,j]= .

[0007] In one embodiment, the step of solving for the inverse matrix L of the unit lower triangular matrix L is... 1 include: Initialize L 1 It is the identity matrix; Traverse the matrix column by column. For each column j, iterate through each row below column j, i∈[j+1,n]. 1], Calculate: L 1 [i,j]= , where n is the matrix dimension.

[0008] In one embodiment, the inverse matrix D of the real diagonal matrix D 1 D is obtained by reversing the diagonal elements. -1 [i,i]=1 / D[i,i], and the off-diagonal elements are kept to be 0.

[0009] In one embodiment, the inverse matrix R of the covariance matrix -1 The calculation formula is: R - ¹ = (L - ¹) × D - ¹× L - ¹.

[0010] In one embodiment, the method further includes: For applications with fixed array sizes, all loop operations are directly expanded into sequentially executed operation steps, and loop control logic is removed to reduce computational resource consumption.

[0011] In one embodiment, the fixed dimension is 4.

[0012] Furthermore, to achieve the above objectives, the present invention also proposes a fast MVDR beamforming apparatus based on LDLH decomposition, wherein the fast MVDR beamforming apparatus based on LDLH decomposition is applied to the fast MVDR beamforming method based on LDLH decomposition described above, and the apparatus includes: The acquisition module is used to acquire the noise and interference data column vectors of the array, and calculate the covariance matrix R=XX based on the noise and interference data column vectors. H Where X is the column vector of noise and interference data, H This indicates the conjugate transpose, and the covariance matrix R is a positive Hermitian definite matrix; The computation module is used to perform LDLH decomposition on the covariance matrix R to obtain a unit lower triangular matrix L and a real diagonal matrix D, where R = L × D × L H ; The computation module is used to solve for the inverse matrix L of the unit lower triangular matrix L. 1 and the inverse matrix D of the real diagonal matrix D 1 ; The calculation module is used to calculate the inverse matrix L of the unit lower triangular matrix L. 1 and the inverse matrix D of the real diagonal matrix D 1 The inverse matrix R of the covariance matrix is ​​calculated. -1 ; The output module is used to calculate the weighting coefficients of the MVDR beamformer based on the inverse matrix of the covariance matrix and the steering vector of the desired signal direction, and to complete the MVDR beamforming based on the weighting coefficients.

[0013] Furthermore, to achieve the above objectives, the present invention also proposes a fast MVDR beamforming device based on LDLH decomposition. The fast MVDR beamforming device based on LDLH decomposition includes: a memory, a processor, and a fast MVDR beamforming program based on LDLH decomposition stored in the memory and executable on the processor. The fast MVDR beamforming program based on LDLH decomposition is configured to implement the steps of the fast MVDR beamforming method based on LDLH decomposition as described above.

[0014] Furthermore, to achieve the above objectives, the present invention also proposes a storage medium storing a fast MVDR beamforming program based on LDLH decomposition, wherein when the fast MVDR beamforming program based on LDLH decomposition is executed by a processor, the program implements the steps of the fast MVDR beamforming method based on LDLH decomposition as described above.

[0015] This invention acquires array noise and interference data and calculates the covariance matrix. The covariance matrix is ​​then decomposed using LDLH to obtain a unit lower triangular matrix L and a real diagonal matrix D. The inverses of L and D are solved separately and combined to obtain the inverse of the covariance matrix. Finally, the MVDR weight coefficients are calculated using the desired signal steering vector to complete beamforming. This method utilizes the Hermitian nature of the covariance matrix R, employing LDLH decomposition for inversion. This transforms the inversion process into operations on the triangular and diagonal matrices, naturally possessing better numerical stability and providing a regular computational structure for hardware optimization. This significantly reduces the computational complexity of covariance matrix inversion, resulting in a significant speed improvement and low hardware resource consumption, making it particularly suitable for array signal processing scenarios with high real-time requirements. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the first embodiment of the fast MVDR beamforming method based on LDLH decomposition of the present invention. Figure 2 This is a structural block diagram of the first embodiment of the fast MVDR beamforming device based on LDLH decomposition of the present invention.

[0017] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0018] It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the invention.

[0019] This invention provides a fast MVDR beamforming method based on LDLH decomposition, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of a fast MVDR beamforming method based on LDLH decomposition according to the present invention.

[0020] In this embodiment, the fast MVDR beamforming method based on LDLH decomposition includes the following steps: Step S10: Obtain the noise and interference data column vectors of the array, and calculate the covariance matrix R=XX based on the noise and interference data column vectors. H .

[0021] In this embodiment, the executing entity is a fast MVDR beamforming device based on LDLH decomposition. This fast MVDR beamforming device based on LDLH decomposition has functions such as data processing, data communication, and program execution. The fast MVDR beamforming device based on LDLH decomposition can be a computer terminal device or other network device, or other devices with similar functions. This embodiment does not limit the scope of such devices.

[0022] It should be noted that MVDR (Minimum Variance Distortionless Response) beamforming is a core technology in array signal processing for suppressing interference and improving the signal-to-noise ratio of the desired signal. Its core step requires inverting the covariance matrix. Traditional MVDR beamforming typically uses Cholesky decomposition, singular value decomposition, or direct matrix inversion to calculate the inverse of the covariance matrix, resulting in a computational complexity as high as O(N³). In embedded / hardware implementations with large array sizes and high real-time requirements, this approach suffers from high computational latency and high hardware resource consumption, making it difficult to meet the application requirements of low power consumption and high real-time performance.

[0023] To address the aforementioned technical issues, this embodiment acquires array noise and interference data and calculates the covariance matrix. LDLH decomposition of the covariance matrix yields a unit lower triangular matrix L and a real diagonal matrix D. The inverses of L and D are then calculated and combined to obtain the inverse of the covariance matrix. Finally, the MVDR weight coefficients are calculated using the desired signal steering vector to complete beamforming. This method leverages the Hermitian nature of the covariance matrix R, employing LDLH decomposition for inversion. This transforms the inversion process into operations on the triangular and diagonal matrices, naturally possessing better numerical stability and providing a regular computational structure for hardware optimization. This significantly reduces the computational complexity of covariance matrix inversion, resulting in a substantial increase in computational speed and low hardware resource consumption, making it particularly suitable for array signal processing scenarios with high real-time requirements.

[0024] It's important to explain the principle of MVDR beamformers: The core idea of ​​MVDR beamformers is to minimize the total output power (i.e., variance) of the beamformer under the constraint of a distortion-free response in the direction of the desired signal. Since the desired signal comes from a specific direction, minimizing the output power is equivalent to maximally suppressing interference and noise from other directions. Intuitively, this means forming a "main lobe" in the direction of the desired signal, with a gain of 1 (i.e., no distortion), while simultaneously forming a "null" in the direction of the interference source, with a gain of 0 (i.e., minimizing interference). Mathematically, this can be expressed as:

[0025] in, The weighting coefficients can be considered as a filter. Indicates conjugate transpose; The covariance matrix representing the noise and interference signals is typically estimated using multiple frames of signal data. ,in The data representing noise and interference is a column vector; express The guide vector corresponding to the direction, Generally, this refers to the desired direction. Solving the above equation using the Lagrange multiplier method yields:

[0026] It is not difficult to see that the entire process only requires estimating the covariance matrix R and solving for its inverse matrix R. -1 Then, the weighting coefficients can be obtained according to the above formula. Taking four microphones as an example, the covariance matrix R is a 4-dimensional matrix. Using conventional inversion methods, the real and imaginary parts need to be expanded into an 8-dimensional matrix, and then elementary transformations are used to solve for the inverse matrix. Considering that the covariance matrix is ​​equal to a column vector multiplied by its conjugate transpose, R = R0. H Therefore, it must be a positive Hermitian definite matrix. Thus, LDLH decomposition can be used to accelerate the solution of its inverse.

[0027] In this embodiment, R=XX H Where X is the column vector of noise and interference data, H The term denotes the conjugate transpose, and the covariance matrix R is a positive Hermitian definite matrix.

[0028] Step S20: Perform LDLH decomposition on the covariance matrix R to obtain the unit lower triangular matrix L and the real diagonal matrix D.

[0029] In practical implementation, the covariance matrix R is also known as the Hermitian positive definite matrix. The covariance matrix R can be decomposed as follows: R = L × D × L H， Where L is a unit lower triangular matrix, that is, the diagonal elements are all 1s and the upper triangular elements are all 0s; D is a real diagonal matrix, that is, the diagonal elements are real numbers and the other elements are all 0s.

[0030] In one embodiment, the covariance matrix R is decomposed by LDLH, specifically by traversing j from 1 to n column by column, where n is the matrix dimension; Calculate the diagonal elements, ensuring that D[j,j] is a real number. The calculation formula is as follows: D[j,j]=REAL(R[j,j] And ensure that D[j,j] is a real number; Calculate the elements in the lower triangular portion of the j-th column of L, for each row i∈[j+1,n] 1]: L[i,j]= .

[0031] In practical applications, the pseudocode for the LDLH decomposition process is shown below: FUNCTION LDLH_DECOMPOSITION(R) n = ROWS(A) / / Matrix dimension, 4 as an example L = EYE(n) / / Initialize L as the identity matrix D = ZEROS(n,n) / / Initialize D as a zero matrix FOR j FROM 0 TO n-1 DO: / / Iterate through columns / / --- Step 1: Calculate the diagonal elements D[j,j] --- sum_diag = 0 FOR k FROM 0 TO j-1 DO: / / Sum of the first j elements sum_diag=sum_diag+L[j,k]×D[k,k]× CONJUGATE(L[j,k]) END FOR D[j,j] = REAL(A[j,j]- sum_diag) / / Ensure D is a real number / / --- Step 2: Calculate the j-th column of L --- FOR i FROM j+1 TO n-1 DO: / / Traverse the rows below column j sum_col = 0 FOR k FROM 0 TO j-1 DO: / / Sum of the first j elements sum_col=sum_col+L[i,k]×D[k,k]×CONJUGATE(L[j,k]) END L[i,j] = (A[i,j]- sum_col) / D[j,j] END END RETURN L, D END The pseudocode performs matrix initialization, calculates the diagonal elements of D, and calculates each column of L from top to bottom. In actual use, since the FOR loop can be removed for a given array, all operations can be directly expanded. Although this increases the number of lines of code, the required computing resources will be further reduced.

[0032] Step S30: Solve for the inverse matrix L of the unit lower triangular matrix L. 1 and the inverse matrix D of the real diagonal matrix D 1 .

[0033] In the specific implementation, the inverse matrix L of the unit lower triangular matrix L is Linverse matrix. 1 The solution process is to initialize L 1 It is the identity matrix; Traverse the matrix column by column. For each column j, iterate through each row below column j, i∈[j+1,n]. 1], Calculate: L 1 [i,j]= , where n is the matrix dimension.

[0034] Furthermore, the inverse matrix D of the real diagonal matrix D 1 D is obtained by reversing the diagonal elements. -1 [i,i]=1 / D[i,i], and the off-diagonal elements are kept to be 0.

[0035] In practical applications, the pseudocode for the above implementation process is shown below: FUNCTION LDL_INVERSE(L, D) n = ROWS(L) / / --- Step 1: Calculate the inverse of D --- D_inv = ZEROS(n,n) FOR i FROM 0 TO n-1 DO: D_inv[i,i] = 1.0 / D[i,i] / / Finding the inverse of a diagonal matrix is ​​very simple END / / --- Step 2: Calculate the inverse of L --- L_inv = EYE(n) / / The inverse of a unit lower triangular matrix is ​​also a unit lower triangular matrix. FOR j FROM 0 TO n-1 DO: FOR i FROM j+1 TO n-1 DO: sum_inv = 0 FOR k FROM j TO i-1 DO: sum_inv = sum_inv + L[i,k] × L_inv[k,j] END L_inv[i,j] = -sum_inv / / Because L[i,i] = 1 END END Step S40: Based on the inverse matrix L of the unit lower triangular matrix L 1 and the inverse matrix D of the real diagonal matrix D 1 The inverse matrix R of the covariance matrix is ​​calculated. -1 .

[0036] In a concrete implementation, the inverse matrix R of the covariance matrix... -1 The calculation formula is: R - ¹ = (L - ¹) × D - ¹ × L - ¹.

[0037] In practical applications, the pseudocode for the above implementation process is shown below: / / --- Step 3: Calculate the inverse of R --- / / R -1 = (L -1 ) ×D -1 ×L -1 R_inv←CONJUGATE_TRANSPOSE(L_inv)×D_inv×L_inv RETURN R_inv END FUNCTION Step S50: Calculate the weighting coefficients of the MVDR beamformer based on the inverse of the covariance matrix and the steering vector of the desired signal direction, and complete the MVDR beamforming based on the weighting coefficients.

[0038] In practical implementation, let the guiding vector be a(θ0), and the formula for calculating the weight coefficient is: w

[0039] Weight coefficients w=[w1,w2,...,w N ] H It is a complex vector of length N, corresponding to the weighting coefficient of each element. Then, the column vector X(t) = [x1(t)] is composed of the received signals of all elements at the current time. x2(t),...,x N [t)]T, where x i (t) represents the time-domain / frequency-domain signal received by the i-th array element. The weighting coefficients are then subjected to an inner product operation: y(t)=wH·X(t)=w1 * x1(t)+w2 * x2(t)+...+w N * x N (t) The obtained y(t) is the final waveform output by the beamformer, where, * This indicates taking the conjugate.

[0040] Furthermore, in one embodiment, for applications with a fixed array size, all loop operations are directly expanded into sequentially executed operation steps, and the loop control logic is removed to reduce computational resource consumption. The fixed dimension is, for example, 4. Actual testing shows that the above method is approximately 7000 times faster than direct expansion of complex matrices plus elementary matrix transformations when the matrix dimension is 4.

[0041] In this embodiment, array noise and interference data are acquired and the covariance matrix is ​​calculated. LDLH decomposition of the covariance matrix yields a unit lower triangular matrix L and a real diagonal matrix D. The inverses of L and D are then calculated and combined to obtain the inverse of the covariance matrix. Finally, the MVDR weight coefficients are calculated using the desired signal steering vector to complete beamforming. This method leverages the Hermitian nature of the covariance matrix R, employing LDLH decomposition for inversion. This transforms the inversion process into operations on the triangular and diagonal matrices, naturally possessing better numerical stability and providing a regular computational structure for hardware optimization. This significantly reduces the computational complexity of covariance matrix inversion, resulting in a substantial increase in computational speed and low hardware resource consumption, making it particularly suitable for array signal processing scenarios with high real-time requirements.

[0042] Furthermore, this embodiment of the invention also proposes a storage medium storing a fast MVDR beamforming program based on LDLH decomposition. When the fast MVDR beamforming program based on LDLH decomposition is executed by a processor, it implements the steps of the fast MVDR beamforming method based on LDLH decomposition as described above.

[0043] Reference Figure 2 , Figure 2 This is a structural block diagram of the first embodiment of the fast MVDR beamforming device based on LDLH decomposition of the present invention.

[0044] like Figure 2 As shown, the fast MVDR beamforming device based on LDLH decomposition proposed in this embodiment of the invention includes: Acquisition module 10 is used to acquire the noise and interference data column vectors of the array, and calculate the covariance matrix R=XX based on the noise and interference data column vectors. H Where X is the column vector of noise and interference data, H This indicates the conjugate transpose, and the covariance matrix R is a positive Hermitian definite matrix; The computation module 20 is used to perform LDLH decomposition on the covariance matrix R to obtain a unit lower triangular matrix L and a real diagonal matrix D, where R = L × D × L H ; The computation module 20 is used to solve for the inverse matrix L of the unit lower triangular matrix L. 1 and the inverse matrix D of the real diagonal matrix D 1 ; The calculation module 20 is used to calculate the inverse matrix L of the unit lower triangular matrix L. 1 and the inverse matrix D of the real diagonal matrix D 1 The inverse matrix R of the covariance matrix is ​​calculated. -1 ; The output module 30 is used to calculate the weight coefficients of the MVDR beamformer based on the inverse matrix of the covariance matrix and the steering vector of the desired signal direction, and to complete the MVDR beamforming based on the weight coefficients.

[0045] In this embodiment, array noise and interference data are acquired and the covariance matrix is ​​calculated. LDLH decomposition of the covariance matrix yields a unit lower triangular matrix L and a real diagonal matrix D. The inverses of L and D are then calculated and combined to obtain the inverse of the covariance matrix. Finally, the MVDR weight coefficients are calculated using the desired signal steering vector to complete beamforming. This method leverages the Hermitian nature of the covariance matrix R, employing LDLH decomposition for inversion. This transforms the inversion process into operations on the triangular and diagonal matrices, naturally possessing better numerical stability and providing a regular computational structure for hardware optimization. This significantly reduces the computational complexity of covariance matrix inversion, resulting in a substantial increase in computational speed and low hardware resource consumption, making it particularly suitable for array signal processing scenarios with high real-time requirements.

[0046] This application embodiment also provides a fast MVDR beamforming device based on LDLH decomposition, including a processor, a communication interface, a memory, and a communication bus. The processor, communication interface, and memory communicate with each other through the communication bus. The memory is used to store the fast MVDR beamforming program based on LDLH decomposition. When the processor executes the program stored in the memory, it implements the above-mentioned fast MVDR beamforming method based on LDLH decomposition.

[0047] The communication bus mentioned in the aforementioned fast MVDR beamforming device based on LDLH decomposition can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into address bus, data bus, control bus, etc.

[0048] The communication interface is used for communication between the aforementioned fast MVDR beamforming device based on LDLH decomposition and other devices.

[0049] The memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0050] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be 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, or discrete hardware components.

[0051] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid state disk (SSD)).

[0052] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0053] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0054] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. 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. Such 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.

[0055] It should be understood that the above are merely illustrative examples and do not constitute any limitation on the technical solutions of the present invention. In specific applications, those skilled in the art can make settings as needed, and the present invention does not impose any restrictions on this.

[0056] It should be noted that the workflow described above is merely illustrative and does not limit the scope of protection of this invention. In practical applications, those skilled in the art can select some or all of the workflow to achieve the purpose of this embodiment according to actual needs, and no restrictions are imposed here.

[0057] In addition, for technical details not described in detail in this embodiment, please refer to the fast MVDR beamforming method based on LDLH decomposition provided in any embodiment of the present invention, which will not be repeated here.

[0058] Furthermore, it should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0059] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0060] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as read-only memory (ROM) / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0061] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

[0062] It is understood that the system provided in the embodiments of the present invention corresponds to the method provided in the embodiments of the present invention, and the explanation, examples and beneficial effects of the relevant content can be referred to the corresponding parts of the above methods.

Claims

1. A fast MVDR beamforming method based on LDLH decomposition, characterized in that, The fast MVDR beamforming method based on LDLH decomposition includes: Obtain the column vectors of noise and interference data from the array, and calculate the covariance matrix R=XX based on the column vectors of noise and interference data. H Where X is the column vector of noise and interference data, H This represents the conjugate transpose, and the covariance matrix R is a positive Hermitian definite matrix; Performing LDLH decomposition on the covariance matrix R yields a unit lower triangular matrix L and a real diagonal matrix D, where R = L × D × L H ; Solve for the inverse matrix L of the unit lower triangular matrix L. 1 and the inverse matrix D of the real diagonal matrix D 1 ; According to the inverse matrix L of the unit lower triangular matrix L 1 and the inverse matrix D of the real diagonal matrix D 1 The inverse matrix R of the covariance matrix is ​​calculated. -1 ; The weighting coefficients of the MVDR beamformer are calculated based on the inverse of the covariance matrix and the steering vector of the desired signal direction, and the MVDR beamformer is completed based on the weighting coefficients.

2. The fast MVDR beamforming method based on LDLH decomposition as described in claim 1, characterized in that, The LDLH decomposition of the covariance matrix R includes: Iterate through j from 1 to n column by column, where n is the dimension of the matrix; Calculate the diagonal elements, ensuring that D[j,j] is a real number. The calculation formula is as follows: D[j,j]=REAL(R[j,j] And ensure that D[j,j] is a real number; Calculate the elements in the lower triangular portion of the j-th column of L, for each row i∈[j+1,n] 1]: L[i,j]= 。 3. The fast MVDR beamforming method based on LDLH decomposition as described in claim 1, characterized in that, The inverse matrix L of the unit lower triangular matrix L is obtained by solving for L. 1 include: Initialize L 1 It is the identity matrix; Traverse the matrix column by column. For each column j, traverse each row below column j, i∈[j+1,n]. 1], Calculate: L 1 [i,j]= , where n is the matrix dimension.

4. The fast MVDR beamforming method based on LDLH decomposition as described in claim 1, characterized in that, The inverse matrix D of the real diagonal matrix D 1 D is obtained by reversing the diagonal elements. -1 [i,i]=1 / D[i,i], and the off-diagonal elements are kept to be 0.

5. The fast MVDR beamforming method based on LDLH decomposition as described in claim 1, characterized in that, The inverse matrix R of the covariance matrix -1 The calculation formula is: R - ¹ = (L - ¹) × D - ¹ × L - ¹.

6. The fast MVDR beamforming method based on LDLH decomposition as described in any one of claims 1 to 5, characterized in that, The method further includes: For applications with fixed array sizes, all loop operations are directly expanded into sequentially executed operation steps, and loop control logic is removed to reduce computational resource consumption.

7. The fast MVDR beamforming method based on LDLH decomposition as described in claim 6, characterized in that, The fixed dimension is 4.

8. A fast MVDR beamforming device based on LDLH decomposition, characterized in that, The fast MVDR beamforming apparatus based on LDLH decomposition is applied to the fast MVDR beamforming method based on LDLH decomposition as described in any one of claims 1 to 7, and the apparatus comprises: The acquisition module is used to acquire the noise and interference data column vectors of the array, and calculate the covariance matrix R=XX based on the noise and interference data column vectors. H Where X is the column vector of noise and interference data, H This represents the conjugate transpose, and the covariance matrix R is a positive Hermitian definite matrix; The computation module is used to perform LDLH decomposition on the covariance matrix R to obtain a unit lower triangular matrix L and a real diagonal matrix D, where R = L × D × L H ; The computation module is used to solve for the inverse matrix L of the unit lower triangular matrix L. 1 and the inverse matrix D of the real diagonal matrix D 1 ; The calculation module is used to calculate the inverse matrix L of the unit lower triangular matrix L. 1 and the inverse matrix D of the real diagonal matrix D 1 The inverse matrix R of the covariance matrix is ​​calculated. -1 ; The output module is used to calculate the weighting coefficients of the MVDR beamformer based on the inverse matrix of the covariance matrix and the steering vector of the desired signal direction, and to complete the MVDR beamforming based on the weighting coefficients.

9. A fast MVDR beamforming device based on LDLH decomposition, characterized in that, The fast MVDR beamforming apparatus based on LDLH decomposition includes: a memory, a processor, and a fast MVDR beamforming program based on LDLH decomposition stored in the memory and executable on the processor, the fast MVDR beamforming program based on LDLH decomposition being configured to implement the steps of the fast MVDR beamforming method based on LDLH decomposition as described in any one of claims 1 to 7.

10. A storage medium, characterized in that, The storage medium stores a fast MVDR beamforming program based on LDLH decomposition, which, when executed by a processor, implements the steps of the fast MVDR beamforming method based on LDLH decomposition as described in any one of claims 1 to 7.

Images