Uplink beam switching method based on joint access point association and beam index selection

Abstract

The invention discloses an uplink beam switching method based on joint access point association and beam index selection, and the method comprises the steps: receiving a plurality of SRS signals of an MITs cluster, and obtaining the channel measurement value of each MIT through matched filtering processing, so as to construct a beam switching decision matrix of each MIT; extracting a sparse principal component matrix from each beam switching decision matrix, and constructing a dominant beam direction solving problem based on the sparse principal component matrix to preliminarily screen dominant beam directions and complete clustering feature matching of the MITs cluster; the dominant beam direction solving problem takes a trace of a maximized sparse principal component matrix as a solving target and takes a support set as a sparse constraint; in the preliminarily screened dominant beam direction, an auxiliary variable and a piecewise linear approximation method are introduced to solve a dominant beam direction solving problem, an optimized beam direction is obtained, and beam switching management of a large-scale MITs cluster in a mobile industrial Internet of Things scene is optimized.

Description

Technical Field

[0001] The present invention belongs to the field of communication technologies, and particularly relates to an uplink beam switching method based on joint access point association and beam index selection. Background Art

[0002] To better support mobility services in industrial environments, it is necessary to ensure seamless data transmission when Mobile IoT Terminals (MITs) move between the coverage areas of different access points. Therefore, effectively managing the beam switching process between the coverage areas of different access points in the factory area is the key to maintaining high-quality services. According to the Fifth Generation New Radio (5G NR) protocol, uplink beam switching management first measures the SRS sent by MITs to evaluate the uplink channel quality. Each SRS (Sounding Reference Signals) is identified by a resource indicator. MITs send SRSs through beam scanning so that the base station can measure the channel states in different uplink directions. After the measurement is completed, the base station notifies the MITs of the uplink beam to be used through downlink control information and the corresponding resource indicator, and then the MITs start uplink data transmission on this beam. This fast reporting mechanism enables the system to respond in a timely manner to channel changes caused by mobility, optimizing communication quality and resource utilization.

[0003] Beam switching management aims to dynamically adjust the beam direction to maintain continuous alignment between MITs and the base station. MITs send SRSs through predefined beam scanning. The base station selects the optimal beam based on the measurement results and triggers the switching process, and selects the optimal direction from the predefined beam set to maximize the signal quality. In millimeter-wave communication, the existing beam alignment strategy based on greedy search reduces the search complexity through hierarchical beam training. In a wideband massive multiple-input multiple-output system, the existing fast beam training method based on frequency scanning reduces the impact of offset on the beamforming gain by pre-compensating the phase difference in the frequency domain. In the scenario of an unmanned aerial vehicle joint base station, the existing adaptive beam reconstruction algorithm dynamically adjusts the beam width and direction by analyzing the user's movement trajectory and channel state information in real time, using the Kalman filter prediction algorithm and combining the Doppler frequency shift compensation mechanism. The existing dynamic adjustment strategies around the beam switching process, combined with channel measurement reports, such as the signal-to-noise ratio and arrival angle estimation of the SRS signal, cope with the movement of MITs by angle offset compensation, such as adding or removing beams in specific directions, but the problems of delay and signaling overhead in multi-node collaborative beam switching have not been effectively solved.

[0004] However, the existing beam switching methods are difficult to adapt to the mobile industrial Internet of Things scenario with large-scale MIT access, with high solution complexity and high computational cost. Summary of the Invention

[0005] To solve the above problems existing in the prior art, the present invention provides an uplink beam switching method based on joint access point association and beam index selection.

[0006] The technical problems to be solved by the present invention are realized through the following technical solutions:

[0007] In a first aspect, the present invention provides an uplink beam switching method based on joint access point association and beam index selection, and the uplink beam switching method includes:

[0008] Receiving multiple SRS signals of the MITs cluster, and obtaining channel measurement values of each MIT through matched filtering processing to construct a beam switching decision matrix for each MIT; the MITs cluster includes multiple MITs;

[0009] Extracting a sparse principal component matrix from the beam switching decision matrices of each MIT, and constructing a dominant beam direction solving problem based on the sparse principal component matrix to preliminarily screen the dominant beam direction and complete the clustering feature matching of the MITs cluster; the dominant beam direction solving problem takes maximizing the trace of the sparse principal component matrix as the solution target and the support set as the sparse constraint; the support set is the position index set of non-zero elements in the sparse principal component matrix;

[0010] Within the preliminarily screened dominant beam directions, introducing auxiliary variables and the piecewise linear approximation method to solve the dominant beam direction solving problem to obtain an optimized beam direction.

[0011] Optionally, extracting a sparse principal component matrix from the beam switching decision matrices of each MIT includes:

[0012] Constructing a sample covariance matrix according to the beam switching decision matrix of each MIT;

[0013] Performing sparse principal component optimization on each sample covariance matrix to obtain a sparse principal component matrix.

[0014] Optionally, the beam switching decision matrix of each MIT includes:

[0015]

[0016] Among them, Ξ(t) represents the beam switching decision matrix; ξ k,n (t) represents the channel measurement value of the nth antenna of the kth MIT in the tth time slot; K represents the total number of MITs; N s represents the number of segments within the time slot; σ 2 represents the variance of the additive noise; s k represents the SRS signal sent by the kth MIT; the superscript H represents the conjugate transpose operation of the matrix; y k,n(t) represents the SRS signal of the k-th MIT received by the base station; ‖·‖ represents the norm operation.

[0017] Optionally, within the initially screened dominant beam direction, introduce auxiliary variables and the piecewise linear approximation method to solve the dominant beam direction solving problem, and the obtained optimized beam direction includes:

[0018] Introduce auxiliary variables and the piecewise linear approximation method to perform convex relaxation on the dominant beam direction solving problem, so as to transform the dominant beam direction solving problem into a convex integer programming problem;

[0019] Within the initially screened dominant beam direction, solve the convex integer programming problem to obtain a relaxed solution, and extract the optimized beam direction from the relaxed solution.

[0020] Optionally, the convex integer programming problem is defined as:

[0021]

[0022] Wherein, represents a binary variable; j = 1, 2,..., K; K represents the total number of MITs; λ j represents an eigenvalue; β ji represents the auxiliary variable; i = 1, 2,..., ζ; ζ represents the number of principal components.

[0023] In a second aspect, the present invention provides an uplink beam switching device based on joint access point association and beam index selection, and the uplink beam switching device includes:

[0024] A construction module, configured to receive multiple SRS signals of an MITs cluster, obtain channel measurement values of each MIT through matched filtering processing, so as to construct a beam switching decision matrix for each MIT; the MITs cluster includes multiple MITs;

[0025] An extraction module, configured to extract a sparse principal component matrix from the beam switching decision matrix of each MIT, and construct a dominant beam direction solving problem based on the sparse principal component matrix to initially screen the dominant beam direction and complete the clustering feature matching of the MITs cluster; the dominant beam direction solving problem takes maximizing the trace of the sparse principal component matrix as the solving objective, and the support set is used as the sparse constraint; the support set is the position index set of non-zero elements in the sparse principal component matrix;

[0026] A solving module, configured to introduce auxiliary variables and the piecewise linear approximation method to solve the dominant beam direction solving problem within the initially screened dominant beam direction, and obtain an optimized beam direction.

[0027] In a third aspect, the present invention provides an electronic device, including a processor, a communication interface, a memory, and a communication bus. Among them, the processor, the communication interface, and the memory complete communication with each other through the communication bus;

[0028] The memory is used to store a computer program;

[0029] The processor is configured to, when executing the computer program stored on the memory, implement the method steps of any one of the above-mentioned uplink beam switching methods based on joint access point association and beam index selection.

[0030] In a fourth aspect, the present invention provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, it implements the method steps of any one of the above-mentioned uplink beam switching methods based on joint access point association and beam index selection.

[0031] The uplink beam switching method based on joint access point association and beam index selection provided by the present invention utilizes the characteristic that multiple MITs within the same MITs cluster have highly correlated data space information. By using the channel measurement values of each MIT to construct the beam switching decision matrix of each MIT, and then extracting the sparse principal component matrix from the beam switching decision matrix of each MIT, and constructing a dominant beam direction solving problem based on the sparse principal component matrix to preliminarily screen the dominant beam direction and complete the clustering feature matching of the MITs cluster, making the joint access point association and beam index selection method provided by the present invention more suitable for large-scale MITs clusters. Then, within the preliminarily screened dominant beam direction, an auxiliary variable and a piecewise linear approximation method are introduced to solve the dominant beam direction solving problem, reducing the solving complexity of optimizing the beam direction, improving the solving accuracy, and further optimizing the beam switching management of large-scale MITs clusters in the mobile industrial Internet of Things scenario.

[0032] The following will further elaborate on the present invention in conjunction with the accompanying drawings. Description of the Drawings

[0033] Figure 1 is a schematic flowchart of an uplink beam switching method based on joint access point association and beam index selection provided by an embodiment of the present invention;

[0034] Figure 2 is a schematic diagram of a mobile industrial Internet of Things scenario;

[0035] Figure 3 is a schematic diagram of uplink beam switching;

[0036] Figure 4 is a schematic diagram of the throughput comparison of various beam switching strategies under different numbers of MITs;

[0037] Figure 5 It is a schematic diagram of the throughput comparison of various beam switching strategies under different sparsity constraints;

[0038] Figure 6 It is a schematic structural diagram of an uplink beam switching device based on joint access point association and beam index selection provided by an embodiment of the present invention;

[0039] Figure 7 It is a schematic structural diagram of an electronic device provided by an embodiment of the present invention. Specific embodiments

[0040] The present invention will be further described in detail below with reference to specific embodiments, but the embodiments of the present invention are not limited thereto.

[0041] In order to solve the problems that the existing beam switching methods are difficult to adapt to the mobile industrial Internet of Things (IoT) scenarios with large-scale Mobile IoT Terminals (MITs) access, have high solution complexity and high computational cost, an embodiment of the present invention provides a method for joint access point association and beam index selection. Refer to Figure 1 , Figure 1 It is a schematic flowchart of an uplink beam switching method based on joint access point association and beam index selection provided by an embodiment of the present invention, which specifically includes the following steps:

[0042] Step S101: Receive multiple Sounding Reference Signals (SRS) of the MITs cluster, and obtain the channel measurement values of each MIT through matched filtering processing to construct a beam switching decision matrix for each MIT; the MITs cluster includes multiple MITs.

[0043] In an embodiment of the present invention, a MIT (Mobile IoT Terminal) has task orientation and presents the characteristic of cluster-activated operation. Each MIT within the MITs cluster has highly correlated data space information.

[0044] A typical mobile industrial IoT scenario includes multiple access points. Refer to Figure 2 , Figure 2 It is a schematic diagram of a mobile industrial IoT scenario. The base station located in the center of the factory serves as the hub of multiple distributed access points, and each access point covers a specific area to ensure comprehensive communication within the factory. When a MIT moves within the factory area, it needs to switch between different beams of the same access point or beams of different access points to maintain the connection. Refer to Figure 3 , Figure 3It is a schematic diagram of uplink beam switching. When a link failure is detected, the MIT needs to request the network to configure the transmission of SRS (Sounding Reference Signals) to perform uplink beam switching. The MIT sends SRS signals through beam scanning, enabling the access point to measure the multi-directional channel states from different MITs. The base station indicates to the MIT to use a specified beam through the downlink control information in the physical downlink control channel, and then the MIT transmits data using this beam on the physical uplink shared channel.

[0045] The SRS configurations of different MITs are different in the time-frequency domain. For the k-th MIT, the starting frequency domain position is denoted as The occupied frequency domain resource is where the frequency index ι ranges from 0 to N RB,k -1, N RB,k is the number of resource blocks; where k = 1, …, K, and K represents the total number of MITs in the MITs cluster. The frequency domain density d fd,k determines the subcarrier spacing, and the subcarriers occupied by SRS are n sub,k = n RB,k ·N sc + d fd,k ·κ, where N sc is the number of subcarriers per resource block, and κ is the subcarrier index. The starting time domain position The corresponding time domain symbol is where the symbol index l ranges from 0 to N sym,k -1, N sym,k is the number of symbols. Let represent the complex value of the SRS signal of the k-th MIT at a specific symbol and subcarrier (q = 1, …, Q k ), then the SRS signal transmitted by the k-th MIT can be expressed as:

[0046]

[0047] where Q k = N sym,k × N RB,k is the total number of sampling points. s k satisfies where is the transmit power; the superscript T represents the transpose operation of the matrix.

[0048] After the base station receives the SRS signal transmitted by the MIT, the base station can obtain the channel measurement values of each MIT through matched filtering processing, which can also be called the channel matrix, and construct a beam switching decision matrix at the time slot level. The specific construction process is as follows:

[0049] For the MIT with low-speed movement in the mobile industrial Internet of Things, a block fading channel model is adopted, assuming that the channel remains constant within the coherence time T c,k ≈1 / (2πf d,k ), where f d,k =v k f c / c is the Doppler shift, f c is the carrier frequency, c is the speed of light, v k is the moving speed of the k-th MIT. Through the n-th channel block, at the t-th time slot, the channel matrix H k,n (t) is composed of the superposition of multipath components and can be expressed as:

[0050]

[0051] where L k is the number of signal paths of the k-th MIT, reflecting the environmental scattering complexity. ρ l,k,n (t) is the time-varying complex gain of the l-th path, including path loss and shadow fading; j represents the imaginary unit. a u (θ l,k ) and a v (φ l,k ) are the uniform linear array response vectors of the base station (with M antennas) and the MIT (with N antennas) respectively, and their expressions are:

[0052]

[0053] where d = λ / 2 is the antenna spacing, λ represents the wavelength, θ l,k is the angle of arrival, and φ l,k is the angle of departure. Through the n-th channel block, at the t-th time slot, the SRS signal y k,n (t) received by the base station from the k-th MIT can be expressed as:

[0054]

[0055] where Ψ(ψ k,n (t)) is the beamforming vector of the MIT, is the beamforming vector of the base station, is the additive noise with variance σ 2 , and I represents the identity matrix.

[0056] Then the base station performs matched filtering on the SRS signal:

[0057]

[0058] where ξ k,n (t) represents the channel measurement value of the n-th antenna of the k-th MIT at the t-th time slot; σ2 represents the variance of the additive noise; s k represents the SRS signal sent by the k-th MIT; the superscript H represents the conjugate transpose operation of the matrix; y k,n (t) represents the SRS signal of the k-th MIT received by the base station; ‖·‖ represents the norm operation.

[0059] And on this basis, a beam switching decision matrix for each MIT at the time slot level is constructed, including:

[0060]

[0061] where, Ξ(t) represents the beam switching decision matrix; K is the total number of MITs in the MITs cluster, and N s is the number of segments within the time slot. For the sake of simplicity, the time slot (t) identifier is omitted in the subsequent steps, and it is defaulted to the current time slot. The beam switching decision matrix is simplified to Ξ.

[0062] In the embodiment of the present invention, the beam switching decision matrix is a strategy guide that instructs the base station how to dynamically adjust its antenna array according to the real-time channel conditions so as to select the best beam direction or configuration for each MIT. The beam switching decision matrix can not only improve the quality and stability of signal transmission, but also effectively reduce interference and improve the spectrum utilization efficiency. Physically speaking, the base station can quickly adjust the direction and width of the beam according to the movement of the MIT and environmental changes, and allow multiple MITs to efficiently share the same frequency band resource simultaneously without interfering with each other, ensuring the optimal connection quality and service experience at any time and anywhere.

[0063] Step S102, extract the sparse principal component matrix from the beam switching decision matrices of each MIT, and construct a dominant beam direction solving problem based on the sparse principal component matrix to preliminarily screen the dominant beam direction and complete the clustering feature matching of the MITs cluster; the dominant beam direction solving problem takes maximizing the trace of the sparse principal component matrix as the solution target and the support set as the sparse constraint; the support set is the set of position indexes of non-zero elements in the sparse principal component matrix.

[0064] In the embodiment of the present invention, extracting the sparse principal component matrix from the beam switching decision matrices of each MIT includes:

[0065] Construct a sample covariance matrix according to the beam switching decision matrices of each MIT;

[0066] Perform sparse principal component optimization on each sample covariance matrix to obtain the sparse principal component matrix.

[0067] First, construct a sample covariance matrix based on the beam switching decision matrix and perform sparse principal component optimization on it, that is, solve the sparse principal component matrix U such that the matrix trace Tr(UT DU) is maximized and satisfies the sparse constraint of the support set, so the problem of solving the dominant beam direction based on spatial correlation can be expressed as:

[0068] (P Beam ):max U Tr(U T DU),stU T U=I ζ ,supp(U)=Ω,|Ω|=η;

[0069] Where ζ represents the number of principal components, corresponding to the size of the beam resource set, and I represents the identity matrix. The support set supp(U) = Ω represents the position index of the nonzero elements in U, and |Ω| = η represents the size of the support set Ω, i.e., the sparsity constraint.

[0070] Spatial correlation refers to the relationship between signal strength and quality across different antennas or beam directions. It reflects the similarity of channel state information (CSI) between geographically close receiving points in a wireless communication environment. Specifically, in beam switching management, spatial correlation is used to describe how the channel characteristics between MITs at different locations within the coverage area of an access point (AP) relate to each other. For example, if two MIT users are close to each other, the signals they receive from the same AP may experience similar fading and interference patterns. This phenomenon is an example of spatial correlation.

[0071] In the embodiment of the present invention, the sparse principal component matrix is a key feature representation extracted from the original data by applying sparse principal component analysis, and is suitable for feature selection in high-dimensional data. In beam switching management, it is used to extract the beam direction that has the greatest impact on signal transmission performance from the beam switching decision matrix. Through the optimization process, the embodiment of the present invention seeks a sparse principal component matrix U that maximizes Tr(U T DU), where is the sample covariance matrix, representing the statistical properties of the beam switching decision matrix. The maximization objective here is to find the direction that explains the most data variance—in other words, the most important beam direction. To ensure that only the most relevant beam directions are selected, rather than all possible beam directions, the sparsity constraints supp(U) = Ω, |Ω| = η are introduced to limit the number of nonzero elements. This approach not only simplifies computational complexity but also focuses on the few most influential beam directions, effectively solving the problem of determining the dominant beam direction.

[0072] In the embodiments of the present invention, the support set refers to the set of position indices of non-zero elements in the sparse principal component matrix U, denoted as supp(U) = Ω. Specifically, it identifies which positions in the U matrix are considered important or significant, and these elements correspond to the most critical beam directions in the beam switching decision process. The size of the support set is determined by |Ω| = η, that is, it specifies how many such important elements there are. Physically, the support set helps the present invention screen out those directions that truly make a significant contribution to the communication quality from among numerous potential beam directions, while ignoring other less important directions. For example, in a complex wireless environment, there may be a large number of possible beam directions, but only a few directions can provide the best signal transmission effect. By defining the concept of the support set, one can focus on these key directions, thereby improving the efficiency and accuracy of beam management and resource allocation.

[0073] In the embodiments of the present invention, the principal component is a data feature obtained in a data dimensionality reduction technique, mainly used to capture the direction with the largest change in the dataset, that is, the direction of the largest variance. In the application scenario of beam switching decision, the principal component can be understood as an abstract expression of the beam direction, which reflects the most significant change trend or pattern in the channel state information. Each principal component is an orthogonal basis vector calculated based on the original data, and its arrangement order decreases sequentially according to the variance explained. In the present invention, the number of principal components ζ is directly related to the size of the beam resource set, that is, the number of main beam directions that the system needs to consider. By using the principal component analysis method, one can identify those directions that contribute the most to the variance from a large number of beam directions, thereby guiding the base station to perform efficient beam scheduling and resource management to improve the overall network performance and service quality.

[0074] To approximate the optimal sparse structure of the dominant beam direction solution problem (P Beam ), in the embodiments of the present invention, the support set Ω is dynamically adjusted through three-step cluster matching. The first step is to calculate the removal candidate set. In each iteration τ, for each index j in the previous support set Ω τ-1 the removal loss is calculated

[0075]

[0076] where represents the square root energy of the j-th column in the sample covariance matrix D; represents the submatrix corresponding to the support set Ω τ-1 extracted from the sample covariance matrix D and subjected to a square root form transformation (normalization); represents the submatrix in the sparse principal component matrix U corresponding to the support set Ω τ-1 ; ‖·‖ Fdenotes the Frobenius norm; ‖·‖2 denotes the L2 norm; Ω τ-1 denotes the support set in the previous iteration.

[0077] Select the index with the minimum loss

[0078] In the second step, calculate the insertion candidate set, which is the complement of the removal candidate set For each index j in, calculate the insertion gain

[0079] [[ID=,15]]

[0080] where Select the index with the maximum gain

[0081] In the embodiments of the present invention, the candidate set refers to the set of potential beam direction indices in the beam switching decision matrix, and these indices represent the candidate beams that may be selected as the dominant beam direction. Specifically, the candidate set is an extension or alternative pool of the support set Ω, containing all beam direction indices that may contribute to the optimization objective (i.e., maximizing Tr(U T DU)). During each iteration, the present invention approximates the optimal solution by dynamically adjusting the content of the support set Ω, and the candidate set provides the beam direction indices for this adjustment process. The existence of the candidate set enables the present invention to evaluate which beam directions should be retained, removed, or added in each iteration, thereby gradually optimizing the structure of the sparse principal component matrix U. Physically speaking, the indices in the candidate set correspond to different antenna array configurations or beam directions, which are the directions that have the greatest impact on the signal transmission performance of the MIT cluster in the current time slot.

[0082] In the embodiments of the present invention, the index j is the mapping relationship from the set of beam direction indices in the support set Ω τ-1 to the beam direction represented by the beam switching decision matrix Ξ. Specifically, the support set Ω τ-1 is a set containing the positions of non-zero elements, representing the beam indices selected as the dominant beam direction in the current iteration. These indices directly correspond to the column values in the beam switching decision matrix Ξ, and each column represents a specific beam direction. Therefore, the index j is actually the identifier of a specific beam direction in the beam switching decision matrix from the support set Ω τ-1 to.

[0083] In each iteration, by calculating the removal loss corresponding to each index j to evaluate how much impact it will have on the objective function (i.e., the matrix trace Tr(U T DU)) if this beam direction is removed from the current support set. The range of the index j is limited to the current support set Ωτ-1 Within this range, it means that the evaluation and adjustment will only be carried out in the already selected dominant beam directions, rather than reselecting from the entire candidate set. This mechanism ensures the efficiency of the present invention while gradually optimizing the sparse structure of the support set to better reflect the distribution characteristics of the dominant beam directions.

[0084] The third step is to update the support set. If Update Ω τ = Ω τ-1 -{j rem}+{j in}, and recalculate the eigenvalue decomposition, otherwise terminate the iteration.

[0085] So far, the dominant beam directions can be preliminarily screened through the support set Ω, that is, the position indices of the non-zero elements in the sparse principal component matrix, namely, the column vectors of the sparse principal component matrix U are summed up to complete the MIT clustering feature matching.

[0086] In step S103, within the preliminarily screened dominant beam directions, an auxiliary variable and the piecewise linear approximation method are introduced to solve the dominant beam direction solving problem to obtain the optimized beam direction.

[0087] In the embodiment of the present invention, according to the preliminarily screened dominant beam directions, it is only necessary to optimize within the sparse subspace defined by the support set Ω instead of global search, which can significantly reduce the computational complexity.

[0088] In the embodiment of the present invention, within the preliminarily screened dominant beam directions, introducing an auxiliary variable and the piecewise linear approximation method to solve the dominant beam direction solving problem to obtain the optimized beam direction includes:

[0089] Introduce an auxiliary variable and the piecewise linear approximation method to perform convex relaxation on the dominant beam direction solving problem to transform the dominant beam direction solving problem into a convex integer programming problem;

[0090] Within the preliminarily screened dominant beam directions, solve the convex integer programming problem to obtain a relaxed solution, and extract the optimized beam direction from the relaxed solution.

[0091] The specific process is as follows:

[0092] First, perform piecewise linear approximation relaxation on the dominant beam direction solving by introducing an auxiliary variable. The eigenvalue decomposition of the sample covariance matrix D is where λ1 ≥ … ≥ λ K ≥ 0 are eigenvalues, and c j correspond to the eigenvectors. U = [u1, …, u ζ is the sparse principal component matrix, and the column vectors u i satisfy the orthogonality constraint. Define Denote the projection of the j-th eigenvector on the i-th principal component. On this basis, introduce the auxiliary variable β ji Approximate And constrain their relationship through a piecewise linear function.

[0093] Next, perform piecewise interval partitioning to convexly relax the problem of solving the dominant beam direction (P Beam ). Divide the range of values of α ji [-δ j , δ j into 2Z subintervals evenly, where is the sparse subset norm of the eigenvector c j on the support set Ω. The splitting points are set as z = -Z, …, Z, where the number of segments z is used to control the trade-off between approximation accuracy and computational complexity.

[0094] Finally, combine the auxiliary variable and the piecewise interval partitioning to define a binary variable for each (j, i) And satisfy Approximate α ji and β ji :

[0095]

[0096] The objective function of the original problem of solving the dominant beam direction (P Beam ) can be relaxed to:

[0097]

[0098] Since (in adjacent intervals), the maximum single-point error is The total error decreases as Z increases.

[0099] So far, the original problem (P Beam ) can be transformed into a convex integer programming problem (subject to the above piecewise linear constraints):

[0100]

[0101] Among them, represents the binary variable; j = 1, 2, …, K; K represents the total number of MITs; λ j represents the eigenvalue; β ji represents the auxiliary variable; i = 1, 2, …, ζ; ζ represents the number of principal components.

[0102] The branch and bound method can be used to handle the binary variable Combined with the aforementioned support set Ω, extract from the relaxed solution to make α jiThe largest beam direction index is used as the final optimized beam direction.

[0103] This process only needs to search within the sparse subspace corresponding to Ω, greatly reducing the scale of variables, significantly reducing the computational complexity, and at the same time ensuring approximate optimality. Finally, according to the optimization result of the beam direction index, the base station issues a resource indication through the physical downlink control channel, notifying the MIT to switch to the specified beam for physical uplink shared channel data transmission.

[0104] In the embodiment of the present invention, the characteristics that multiple MITs within the same MITs cluster have highly correlated data space information are utilized. The beam switching decision matrix of each MIT is constructed through the channel measurement values of each MIT, and then the sparse principal component matrix is extracted from the beam switching decision matrix of each MIT. Based on the sparse principal component matrix, a problem of solving the dominant beam direction is constructed to preliminarily screen the dominant beam direction and complete the clustering feature matching of the MITs cluster, making the joint access point association and beam index selection method provided by the present invention more suitable for large-scale MITs clusters. Then, within the preliminarily screened dominant beam directions, auxiliary variables and the piecewise linear approximation method are introduced to solve the problem of solving the dominant beam direction, reducing the solution complexity of the optimized beam direction, improving the solution accuracy, and further optimizing the beam switching management of large-scale MITs clusters in the mobile industrial Internet of Things scenario.

[0105] Based on the above content, the core idea of the present invention includes two aspects: First, the mobile Internet of Things terminals are task-oriented and exhibit the characteristics of cluster activation operation. The MITs within the same cluster have highly correlated data space information; Second, in an unmanned chemical plant, due to the low complexity of the scattering environment and the strong periodicity and regularity of the running trajectories of medium- and low-speed moving MITs, the sounding reference signal measurement report can effectively indicate the optimal beam switching index based on predictable channel information. Therefore, the present invention first extracts the characteristics of the MITs cluster to match the access points in the corresponding coverage area, and then selects the optimal beam index by identifying the main space direction of the channel information. Compared with the existing beam switching management methods, the present invention improves the solution accuracy and feasibility of access point association and beam index selection while streamlining the processing flow through index exchange and ordered subset partitioning techniques, can efficiently handle large-scale SRS measurement problems in the mobile Internet of Things scenario, and has significant engineering application value.

[0106] The simulation experiment of the uplink beam switching method based on joint access point association and beam index selection provided by the embodiment of the present invention is as follows:

[0107] (1) Instance scenario setting

[0108] The experimental system is deployed in a 500m×500m area. The base station is located at the center point (250, 250), and four groups of 64-antenna access points are respectively configured at the coordinate nodes (75, 75), (75, 425), (425, 75), and (425, 425). Each MIT is equipped with 4 antennas, and a minimum spacing constraint of 20m is imposed during deployment. The SRS uses orthogonal sequence allocation (n sta,k = 5 + 3(k - 1)), and each MIT fixedly occupies N RB,k = 24 resource blocks, where the subcarrier spacing d fd,k is randomly set between 0.5 and 3 to simulate heterogeneous service requirements. The time-domain resource configuration adopts a misaligned allocation strategy for symbols 0 to 3 to ensure asynchronous transmission of the SRS. The transmit power is dynamically adjusted to the range of 20dBm to 26dBm according to the link budget, reflecting the spatial location difference between the MIT and the base station. The number of beam resource sets configured for each access point is set to 4. To achieve ordered subset partitioning, the interval [-1, 1] is divided into 10 equal-length sub-segments, each with a length of 0.2. In the channel model parameter settings, the carrier frequency f c = 3500MHz, the number of channel blocks N s = 64. The number of multipaths L k of each MIT varies between 4 and 8, reflecting propagation environments with different scattering intensities. Since the moving speed v k of the MIT varies between 1m / s and 5m / s, the resulting Doppler frequency shift f d,k is different, making the channel coherence time T c,k distribute in the range of approximately 3.5ms to 17.9ms. Therefore, the transmission time slot T s is set to be less than or equal to the corresponding coherence time, and usually 1ms is taken to ensure channel stability within each transmission block.

[0109] (2) Experimental details:

[0110] a) Beam switching throughput analysis: When the number of MITs increases from 12 to 32, the proposed JAB method maintains high throughput through ordered subset partitioning (such as dividing the interval [-1, 1] into 10 segments, each with a length of 0.2). By comparing greedy selection switching, periodic beam switching, and no beam switching, it is shown that the present invention can significantly improve the uplink beam switching management performance in the mobile industrial Internet of Things scenario.

[0111] b) Beam switching robustness verification: When the sparsity η overshoots (e.g., η > 8), the proposed uplink beam switching method based on JAB (Joint Access Point Association and Beam Index Selection) in the present invention suppresses beam conflicts through convex relaxation, avoiding a sharp drop in throughput. For example, when η = 10, the throughput only drops by 3.6%. By comparing greedy selection switching, periodic beam switching, and no beam switching, it is shown that the present invention has beam switching robustness in mobile industrial Internet of Things scenarios with different sparse MIT cluster characteristics.

[0112] To verify the performance of the proposed method, throughput is used as the evaluation metric in Figure 4 and Figure 5 to conduct a comparative analysis of the following three existing beam switching strategies, including greedy selection switching, periodic beam switching, and no beam switching. As Figure 4 shown, Figure 4 is a schematic diagram of the throughput comparison of various beam switching strategies under different numbers of MITs. Under different numbers of MITs, the proposed method can closely approach the optimal theoretical upper bound of greedy selection switching. In particular, as the number of MITs increases, the proposed method benefits from the ordered subset partitioning mechanism, and while maintaining the solution quality, the gap in throughput improvement with periodic beam switching significantly expands. Periodic beam switching is only applicable to scenarios with a small number of MITs. When the number of MITs increases, the mechanical periodic switching cannot meet the service requirements of a large number of beam switches in mobile industrial Internet of Things scenarios, significantly affecting the overall network performance. Figure 4 shows the performance comparison of various beam switching strategies under different sparsity scenarios. The experimental results show that when the sparsity level changes, the proposed method can still closely approach the theoretical upper bound and its performance is better than the periodic switching strategy. Figure 5 is a schematic diagram of the throughput comparison of various beam switching strategies under different sparsity constraints. For the no-switching strategy, an increase in the sparsity level will cause the beam coverage area to expand, thus improving the overall service quality. In addition, as the sparsity constraint η increases (except for the no-switching strategy), the system throughput shows a trend of first rising and then falling. This is because when the sparsity η increases, the switching selection range expands, and effective beam switching can improve the uplink transmission rate. However, when the sparsity η exceeds the critical threshold, the increase in the required accuracy of switching and the increase in selection errors will lead to an exacerbation of switching conflicts between different multi-beam index tasks, thereby affecting the overall throughput.

[0113] In the embodiments of the present invention, an uplink beam switching management method based on a cluster-level protocol interaction strategy is proposed to protect the method of realizing SRS resource pooling by utilizing the spatial correlation of MITs clusters, including the specific implementation processes of sharing time-frequency parameters and differential subcarrier allocation. A sparse feature extraction and index optimization method is proposed to protect the technical solution of transforming the beam selection problem into a convex optimization by piecewise linear approximation and auxiliary variable definition, including the specific steps of dynamic support set update.

[0114] Compared with the prior art, the present invention fully processes the protocol interaction details: the proposed JAB method significantly improves the accuracy and real-time performance of beam switching by deeply optimizing the protocol interaction process. By staggering the SRS start symbols and frequency domain densities of different MITs, signal orthogonality is ensured and measurement conflicts are avoided. At the same time, a cluster-level resource pooling strategy is proposed for clustered MITs, sharing the time-frequency start parameters, differential subcarrier allocation, and reducing the independent signaling overhead. Adapt to large-scale scenarios and reduce complexity: the proposed JAB method realizes low-complexity and high-scalability beam switching management through algorithm innovation and architecture design. For large-scale MIT access, a clustering cooperation mechanism based on spatial correlation is designed, and each access point only processes the local information of the MITs within its coverage area, reducing the global data synchronization requirement through an asynchronous update strategy.

[0115] Based on the same inventive concept, the embodiments of the present invention also provide an uplink beam switching device based on joint access point association and beam index selection. See Figure 6 , Figure 6 FIG. is a schematic structural diagram of an uplink beam switching device based on joint access point association and beam index selection provided by the embodiments of the present invention. The uplink beam switching device includes:

[0116] A construction module 601, configured to receive multiple SRS signals of an MITs cluster, obtain channel measurement values of each MIT through matched filtering processing, and construct a beam switching decision matrix for each MIT; the MITs cluster includes multiple MITs;

[0117] An extraction module 602, configured to extract a sparse principal component matrix from the beam switching decision matrix of each MIT, and construct a dominant beam direction solving problem based on the sparse principal component matrix to preliminarily screen the dominant beam direction and complete the clustering feature matching of the MITs cluster; the dominant beam direction solving problem takes maximizing the trace of the sparse principal component matrix as the solution target and the support set as the sparse constraint; the support set is the position index set of non-zero elements in the sparse principal component matrix;

[0118] A solving module 603, configured to introduce an auxiliary variable and a piecewise linear approximation method to solve the dominant beam direction solving problem within the preliminarily screened dominant beam directions, and obtain an optimized beam direction.

[0119] In the embodiment of the present invention, the characteristics that multiple MITs within the same MITs cluster have highly correlated data space information are utilized. By using the channel measurement values of each MIT to construct the beam switching decision matrix of each MIT, a sparse principal component matrix is then extracted from the beam switching decision matrix of each MIT, and a dominant beam direction solving problem is constructed based on the sparse principal component matrix to preliminarily screen the dominant beam direction and complete the clustering feature matching of the MITs cluster, making the joint access point association and beam index selection method provided by the present invention more suitable for large-scale MITs clusters. Then, within the preliminarily screened dominant beam directions, auxiliary variables and the piecewise linear approximation method are introduced to solve the dominant beam direction solving problem, reducing the solving complexity of optimizing the beam direction, improving the solving accuracy, and further optimizing the beam switching management of large-scale MITs clusters in the mobile industrial Internet of Things scenario.

[0120] Optionally, the extraction module extracts the sparse principal component matrix from the beam switching decision matrix of each MIT, including:

[0121] Constructing a sample covariance matrix according to the beam switching decision matrix of each MIT;

[0122] Performing sparse principal component optimization on each sample covariance matrix to obtain a sparse principal component matrix.

[0123] Optionally, the beam switching decision matrix of each MIT includes:

[0124]

[0125] where, Ξ(t) represents the beam switching decision matrix; ξ k,n (t) represents the channel measurement value of the nth antenna of the kth MIT in the tth time slot; K represents the total number of MITs; N s represents the number of segments within the time slot; σ 2 represents the variance of the additive noise; s k represents the SRS signal transmitted by the kth MIT; the superscript H represents the conjugate transpose operation of the matrix; y k,n (t) represents the SRS signal of the kth MIT received by the base station; ‖·‖ represents the norm operation.

[0126] Optionally, the solving module is specifically configured to:

[0127] Introduce auxiliary variables and the piecewise linear approximation method to perform convex relaxation on the dominant beam direction solving problem, so as to transform the dominant beam direction solving problem into a convex integer programming problem; within the preliminarily screened dominant beam directions, solve the convex integer programming problem to obtain a relaxed solution, and extract the optimized beam direction from the relaxed solution.

[0128] Optionally, the convex integer programming problem is defined as:

[0129]

[0130] where represents a binary variable; j = 1, 2, ..., K; K represents the total number of MITs; λ j represents an eigenvalue; β ji represents the auxiliary variable; i = 1, 2, ..., ζ; ζ represents the number of principal components.

[0131] An embodiment of the present invention further provides an electronic device, as Figure 7 shown, including a processor 701, a communication interface 702, a memory 703, and a communication bus 704. Among them, the processor 701, the communication interface 702, and the memory 703 communicate with each other through the communication bus 704.

[0132] The memory 703 is used to store a computer program;

[0133] When the processor 701 executes the program stored on the memory 703, it implements the method steps of any one of the above-mentioned uplink beam switching methods based on joint access point association and beam index selection.

[0134] The communication bus mentioned in the above electronic device may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of representation, only a thick line is used in the figure to represent it, but it does not mean that there is only one bus or one type of bus.

[0135] The communication interface is used for communication between the above electronic device and other devices.

[0136] The memory may include a Random Access Memory (RAM), or may also include a Non-Volatile Memory (NVM), such as at least one disk memory. Optionally, the memory may also be at least one storage device located far from the aforementioned processor.

[0137] The above-mentioned processor may be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0138] The present invention also provides a computer-readable storage medium. A computer program is stored in the computer-readable storage medium. When the computer program is executed by a processor, the method steps of any one of the above-mentioned uplink beam switching methods based on joint access point association and beam index selection are implemented.

[0139] Optionally, the computer-readable storage medium may be a non-volatile memory (NVM), such as at least one disk memory.

[0140] Optionally, the above-mentioned computer-readable storage medium may also be at least one storage device located far from the aforementioned processor.

[0141] In another embodiment of the present invention, a computer program product containing instructions is also provided. When it runs on a computer, it causes the computer to execute the method steps of any one of the above-mentioned uplink beam switching methods based on joint access point association and beam index selection.

[0142] It should be noted that terms such as "first" and "second" are used to distinguish similar objects and do not necessarily describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances so that the embodiments of the present invention described here can be implemented in an order other than those illustrated or described here. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present invention.

[0143] In the description of this specification, the description referring to terms such as "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features or characteristics described can be combined in a suitable manner in any one or more embodiments or examples. In addition, those skilled in the art can combine and combine the different embodiments or examples described in this specification.

[0144] Although the present invention has been described herein in connection with various embodiments, however, in the process of implementing the claimed invention, those skilled in the art can understand and achieve other variations of the disclosed embodiments by viewing the accompanying drawings and the disclosure. In the description of the present invention, the term "including" does not exclude other components or steps, the term "a" or "one" does not exclude a plurality of cases, and the meaning of "a plurality" is two or more, unless otherwise specifically defined. In addition, certain measures are described in different embodiments, but this does not mean that these measures cannot be combined to produce good results.

[0145] The method provided by the embodiments of the present invention can be applied to an electronic device. Specifically, the electronic device can be: a desktop computer, a portable computer, a smart mobile terminal, a server, etc. There is no limitation here, and any electronic device that can implement the present invention belongs to the protection scope of the present invention.

[0146] For the embodiments of the apparatus / electronic device / storage medium, since they are basically similar to the method embodiments, the description is relatively simple, and for the relevant parts, reference can be made to the partial description of the method embodiments.

[0147] It should be noted that the apparatus, electronic device, and storage medium of the embodiments of the present invention are respectively the apparatus, electronic device, and storage medium applying the above-mentioned uplink beam switching method based on joint access point association and beam index selection. Then all the embodiments of the above-mentioned uplink beam switching method based on joint access point association and beam index selection are applicable to the apparatus, electronic device, and storage medium, and can achieve the same or similar beneficial effects.

[0148] The above content is a further detailed description of the present invention in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present invention is only limited to these descriptions. For those of ordinary skill in the technical field to which the present invention pertains, without departing from the concept of the present invention, several simple deductions or substitutions can still be made, and all should be regarded as belonging to the protection scope of the present invention.

Claims

1. An uplink beam switching method based on joint access point association and beam index selection, characterized in that The uplink beam switching method includes: Receiving multiple SRS signals of an MITs cluster, obtaining channel measurement values of each MIT through matched filtering processing to construct a beam switching decision matrix for each MIT; the MITs cluster includes multiple MITs; Extracting a sparse principal component matrix from the beam switching decision matrix of each MIT, and constructing a dominant beam direction solving problem based on the sparse principal component matrix to preliminarily screen the dominant beam direction and complete the clustering feature matching of the MITs cluster; the dominant beam direction solving problem aims to maximize the trace of the sparse principal component matrix, and the support set is used as the sparse constraint; the support set is the position index set of non-zero elements in the sparse principal component matrix; Within the preliminarily screened dominant beam directions, introducing auxiliary variables and the piecewise linear approximation method to solve the dominant beam direction solving problem to obtain an optimized beam direction.

2. The uplink beam switching method according to claim 1, characterized in that Extracting a sparse principal component matrix from the beam switching decision matrix of each MIT includes: Constructing a sample covariance matrix according to the beam switching decision matrix of each MIT; Performing sparse principal component optimization on each sample covariance matrix to obtain a sparse principal component matrix.

3. The uplink beam switching method according to claim 1, wherein The beam switching decision matrix of each MIT includes: Among them, Ξ(t) represents the beam switching decision matrix; ξ k,n (t) represents the channel measurement value of the nth antenna of the kth MIT in the tth time slot; K represents the total number of MITs; N s represents the number of segments within a time slot; σ 2 represents the variance of additive noise; s k represents the SRS signal transmitted by the kth MIT; the superscript H represents the conjugate transpose operation of the matrix; y k,n (t) represents the SRS signal of the kth MIT received by the base station; ‖·‖ represents the norm operation.

4. The uplink beam switching method according to claim 1, wherein Within the preliminarily screened dominant beam directions, introducing auxiliary variables and the piecewise linear approximation method to solve the dominant beam direction solving problem to obtain an optimized beam direction includes: Introducing auxiliary variables and the piecewise linear approximation method to perform convex relaxation on the dominant beam direction solving problem to transform the dominant beam direction solving problem into a convex integer programming problem; Within the preliminarily screened dominant beam directions, solving the convex integer programming problem to obtain a relaxed solution, and extracting the optimized beam direction from the relaxed solution.

5. The uplink beam switching method according to claim 4, wherein The convex integer programming problem is defined as: Among them, represents a binary variable; j = 1, 2, ..., K; K represents the total number of MITs; λ j represents an eigenvalue; β ji represents the auxiliary variable; i = 1, 2, ..., ζ; ζ represents the number of principal components.

6. An uplink beam switching device based on joint access point association and beam index selection, characterized in that The uplink beam switching device includes: A construction module, configured to receive multiple SRS signals of an MITs cluster, obtain channel measurement values of each MIT through matched filtering processing to construct a beam switching decision matrix for each MIT; the MITs cluster includes multiple MITs; An extraction module, configured to extract a sparse principal component matrix from the beam switching decision matrix of each MIT, and construct a dominant beam direction solving problem based on the sparse principal component matrix to preliminarily screen the dominant beam direction and complete the clustering feature matching of the MITs cluster; the dominant beam direction solving problem aims to maximize the trace of the sparse principal component matrix, and the support set is used as the sparse constraint; the support set is the position index set of non-zero elements in the sparse principal component matrix; A solving module, configured to introduce auxiliary variables and the piecewise linear approximation method to solve the dominant beam direction solving problem within the preliminarily screened dominant beam directions to obtain an optimized beam direction.

7. The uplink beam switching device according to claim 6, wherein The extraction module, extracting a sparse principal component matrix from the beam switching decision matrix of each MIT, includes: Constructing a sample covariance matrix according to the beam switching decision matrix of each MIT; Performing sparse principal component optimization on each sample covariance matrix to obtain a sparse principal component matrix.

8. The uplink beam switching device according to claim 7, wherein The beam switching decision matrix of each MIT includes: Among them, Ξ(t) represents the beam switching decision matrix; ξ k,n (t) represents the channel measurement value of the nth antenna of the kth MIT in the tth time slot; K represents the total number of MITs; N s represents the number of segments within the time slot; σ 2 represents the variance of the additive noise; s k represents the SRS signal transmitted by the kth MIT; the superscript H represents the conjugate transpose operation of the matrix; y k,n (t) represents the SRS signal of the kth MIT received by the base station; ‖·‖ represents the norm operation.

9. An electronic device, characterized in that, It includes a processor, a communication interface, a memory, and a communication bus. Among them, the processor, the communication interface, and the memory complete mutual communication through the communication bus; The memory is used to store computer programs; The processor, when executing the computer programs stored on the memory, implements the uplink beam switching method based on joint access point association and beam index selection according to any one of claims 1-5.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer programs, and when the computer programs are executed by the processor, the uplink beam switching method based on joint access point association and beam index selection according to any one of claims 1-5 is implemented.

Images