Analog-digital hybrid beam forming method and device applied to millimeter wave communication system

Abstract

The invention provides an analog-digital hybrid beam forming method and device applied to a millimeter wave communication system, and belongs to the technical field of communication. The millimeter wave communication system comprises a base station and a plurality of single-antenna users in communication connection with the base station, the base station is provided with a plurality of transmitting antennas and a plurality of radio frequency links, and the method comprises the following steps: obtaining interference noise information and at least one channel state matrix from the base station to each single-antenna user, determining a receiving signal-to-noise ratio model of each single-antenna user by combining an unknown analog beam forming matrix used for constraining an analog transmitting behavior between each transmitting antenna and each radio frequency link and a digital beam forming matrix used for constraining a digital transmitting behavior between each radio frequency link and each single-antenna user; and further determining an optimization problem model aiming at maximizing the user and the rate, and solving by adopting a depth expansion technology to obtain the optimal values of the analog beam forming matrix and the digital beam forming matrix. According to the invention, the system communication rate performance can be improved.

Description

Technical Field

[0001] This application relates to the field of communication technologies, and in particular, to an analog-digital hybrid beamforming method and apparatus applied to a millimeter-wave communication system. Background Art

[0002] A millimeter-wave communication system can utilize the high-frequency range of the millimeter-wave spectrum for data transmission, having advantages such as high bandwidth and large capacity, while also facing challenges such as high path loss and rapid signal attenuation. To address this challenge, a hybrid beamforming scheme is proposed to be introduced into the millimeter-wave communication system. However, currently, the gradient projection method with a relatively slow convergence speed is usually adopted for solving, resulting in the need to improve the communication rate performance of the system. Summary of the Invention

[0003] The main objective of this application is to propose an analog-digital hybrid beamforming method and apparatus applied to a millimeter-wave communication system, which can effectively reduce the computational complexity in solving the hybrid beamforming problem and improve the communication rate performance of the millimeter-wave communication system.

[0004] To achieve the above objective, on the one hand, this application proposes an analog-digital hybrid beamforming method applied to a millimeter-wave communication system. The millimeter-wave communication system includes a base station and multiple single-antenna users, and there is a communication connection between the base station and the multiple single-antenna users. The base station is configured with multiple transmit antennas and multiple radio frequency links. The method includes:

[0005] Obtain the interference noise information from the base station to each of the single-antenna users and at least one channel state matrix, and combine the unknown analog beamforming matrix and digital beamforming matrix to determine the received signal-to-noise ratio model for each of the single-antenna users;

[0006] According to the received signal-to-noise ratio model for each of the single-antenna users, determine an optimization problem model with the goal of maximizing the user sum rate;

[0007] Use the deep unfolding technology to solve the optimization problem model to obtain the optimal values of the analog beamforming matrix and the digital beamforming matrix;

[0008] Wherein, the analog beamforming matrix includes the analog transmit control parameters between each of the transmit antennas and each of the radio frequency links, and the digital beamforming matrix includes the digital transmit control parameters between each of the radio frequency links and each of the single-antenna users.

[0009] Further, the received signal-to-noise ratio model for each of the single-antenna users is obtained through the following method:

[0010] Determine the received signal model of each single-antenna user according to the simulated beamforming matrix, the digital beamforming matrix, the interference noise information from the base station to each single-antenna user, and at least one channel state matrix;

[0011] Split the digital beamforming matrix to obtain multiple digital beamforming sub-matrices corresponding to the multiple single-antenna users. The digital beamforming sub-matrix of each single-antenna user includes the digital transmission control parameters between the single-antenna user and each radio frequency link;

[0012] According to the digital beamforming sub-matrices of each single-antenna user, decompose and transform the received signal model of each single-antenna user to obtain the received signal-to-noise ratio model of each single-antenna user.

[0013] Further, the optimization problem model is obtained through the following method:

[0014] According to the received signal-to-noise ratio model of each single-antenna user, use the Shannon formula to determine the user sum rate model in the millimeter-wave communication system;

[0015] According to the user sum rate model and the preset constraint conditions between the simulated beamforming matrix and the digital beamforming matrix, determine the optimization problem model.

[0016] Further, the constraint conditions between the simulated beamforming matrix and the digital beamforming matrix include that the transmit power of the base station determined by the simulated beamforming matrix and the digital beamforming matrix does not exceed the preset transmit power upper limit value.

[0017] Further, the transmit power of the base station is obtained through the following method:

[0018] Determine the F-norm between the simulated beamforming matrix and the digital beamforming matrix, and then use the square value of the F-norm as the transmit power of the base station.

[0019] Further, the steps of using the deep unfolding technology to solve the optimization problem model to obtain the optimal values of the simulated beamforming matrix and the digital beamforming matrix include:

[0020] According to the optimization problem model, determine the objective function;

[0021] Randomly initialize the simulated beamforming matrix and the digital beamforming matrix, and then input the randomly initialized simulated beamforming matrix and digital beamforming matrix into a preset deep unfolding network, and the deep unfolding network is designed based on the iterative principle of the gradient projection method;

[0022] Iteratively train the deep unfolding network according to the objective function and a preset maximum number of training rounds to obtain the optimal values of the analog beamforming matrix and the digital beamforming matrix.

[0023] Further, each round of training process of the deep unfolding network includes:

[0024] Obtain the current values of the analog beamforming matrix and the digital beamforming matrix;

[0025] Obtain the current values of the first gradient step size and the second gradient step size, where the first gradient step size is used to assist in updating the analog beamforming matrix, and the second gradient step size is used to assist in updating the digital beamforming matrix;

[0026] In the forward propagation stage of the deep unfolding network, use the number of layers of the deep unfolding network as the number of iterations of the gradient projection method, and update the current values of the analog beamforming matrix and the digital beamforming matrix using the gradient projection method according to the objective function and the current values of the first gradient step size and the second gradient step size;

[0027] In the backward propagation stage of the deep unfolding network, update the current values of the first gradient step size and the second gradient step size according to the objective function.

[0028] To achieve the above object, another aspect of the present application proposes an analog-digital hybrid beamforming device applied to a millimeter-wave communication system. The millimeter-wave communication system includes a base station and multiple single-antenna users. The base station is communicatively connected to the multiple single-antenna users, and the base station is configured with multiple transmit antennas and multiple radio frequency links; the device includes:

[0029] A first determination module, configured to obtain interference noise information from the base station to each of the single-antenna users and at least one channel state matrix, and determine a received signal-to-noise ratio model for each of the single-antenna users in combination with an unknown analog beamforming matrix and digital beamforming matrix;

[0030] A second determination module, configured to determine an optimization problem model with the goal of maximizing the user sum rate according to the received signal-to-noise ratio model of each of the single-antenna users;

[0031] A solution module, configured to solve the optimization problem model using a deep unfolding technique to obtain the optimal values of the analog beamforming matrix and the digital beamforming matrix;

[0032] Among them, the analog beamforming matrix includes analog transmission control parameters between each of the transmitting antennas and each of the radio frequency links, and the digital beamforming matrix includes digital transmission control parameters between each of the radio frequency links and each of the single-antenna users.

[0033] To achieve the above object, another aspect of the present application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the above method is implemented.

[0034] To achieve the above object, another aspect of the present application provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, the above method is implemented.

[0035] The present application has at least the following beneficial effects: Considering the received signal-to-noise ratio of each single-antenna user, by using the analog beamforming matrix, the digital beamforming matrix, the interference noise information from the base station to each single-antenna user, and at least one channel state matrix to construct an optimization problem model with the goal of maximizing the user sum rate, and then using the deep unfolding technology for solution, the millimeter-wave communication system can achieve robust and efficient hybrid beamforming by means of the optimal values of the obtained analog beamforming matrix and digital beamforming matrix, which is beneficial to improving the communication rate performance of the millimeter-wave communication system and can effectively reduce the computational complexity in solving the hybrid beamforming problem. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 FIG. is a schematic flowchart of an analog-digital hybrid beamforming method applied to a millimeter-wave communication system provided by an embodiment of the present application;

[0037] Figure 2 FIG. is a schematic structural composition diagram of an analog-digital hybrid beamforming device applied to a millimeter-wave communication system provided by an embodiment of the present application;

[0038] Figure 3 FIG. is a schematic hardware structure diagram of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0039] In order to make the objectives, technical solutions, and advantages of this application more clearly understood, the following further elaborates on this application in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely used to explain this application and are not intended to limit this application. When the following description refers to the accompanying drawings, unless otherwise indicated, the same numerals in different drawings represent the same or similar elements. The implementation manners described in the following exemplary embodiments do not represent all implementation manners consistent with the embodiments of this application. They are merely examples of systems and methods consistent with some aspects of the embodiments of this application as detailed in the appended claims.

[0040] It can be understood that the terms "first", "second", etc. used in this application may be used herein to describe various concepts, but unless otherwise specified, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another. For example, without departing from the scope of the embodiments of this application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the words "if", "when" as used herein may be interpreted as "when...", "while...", or "in response to determining".

[0041] The terms "at least one", "multiple", "each", "any one", etc. used in this application, at least one includes one, two, or more than two, multiple includes two or more than two, each refers to each of the corresponding multiple, and any one refers to any one of the multiple.

[0042] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0043] Before elaborating in detail on the embodiments of this application, some nouns and terms involved in the embodiments of this application are first explained, and the nouns and terms involved in the embodiments of this application are subject to the following explanations.

[0044] The deep unfolding technology combines the traditional model-driven method with the data-driven deep learning technology, not only making full use of the structured information of the model, but also being able to optimize performance by learning the patterns in the data. The deep unfolding technology can retain the basic theory in the communication field and, with the help of the powerful capabilities of deep learning, make it have good interpretability and excellent performance.

[0045] The gradient projection method is a method for solving the optimal solution of constrained nonlinear programming problems. By using the projection technique of the gradient, it searches on the boundary of a convex constraint set and gradually approaches the optimal solution. The basic principle of the gradient projection method is to start from a basic feasible solution, determine the projection direction of the gradient through the constraint conditions, and then perform iterative updates until the accuracy requirement is met.

[0046] Millimeter-wave communication systems can utilize the high-frequency range of the millimeter-wave spectrum for data transmission, with advantages such as high bandwidth and large capacity. At the same time, they also face challenges such as high path loss and rapid signal attenuation. To address this challenge, a digital beamforming scheme is proposed to be introduced into millimeter-wave communication systems. This requires configuring a larger number of radio frequency links for base stations deploying full digital beamforming, which results in high hardware implementation costs. Later, to reduce the hardware cost, a hybrid beamforming scheme is proposed to be introduced into millimeter-wave communication systems. However, the traditional gradient projection method is usually used for solving. Since the gradient projection method uses a fixed update step size and requires manual adjustment, the convergence speed is slow during the application process, resulting in the need to improve the communication rate performance of the system, and the computational complexity is also relatively high when solving the hybrid beamforming problem.

[0047] In view of this, the embodiments of the present application provide an analog-digital hybrid beamforming method and device applied to millimeter-wave communication systems, which can effectively reduce the computational complexity when solving the hybrid beamforming problem and improve the communication rate performance of millimeter-wave communication systems.

[0048] An analog-digital hybrid beamforming method applied to millimeter-wave communication systems provided by the embodiments of the present application relates to the field of communication technologies and can be applied to terminals, servers, or software running on terminals or servers. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, a vehicle-mounted terminal, etc., but is not limited thereto; the server side can be configured as an independent physical server, or can be configured as a server cluster or distributed system composed of multiple physical servers, or can be configured as a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The server can also be a node server in a blockchain network; the software can be an application implementing the above-mentioned analog-digital hybrid beamforming method applied to millimeter-wave communication systems, etc., but is not limited to the above forms.

[0049] This application can be used in numerous general-purpose or special-purpose computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, and so on. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. This application can also be practiced in a distributed computing environment where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0050] Please refer to Figure 1 , Figure 1 FIG. is an alternative flowchart of an analog-digital hybrid beamforming method applied to a millimeter-wave communication system provided by an embodiment of this application. The millimeter-wave communication system includes a base station and multiple single-antenna users. The base station is communicatively connected to the multiple single-antenna users. The base station is configured with multiple transmit antennas and multiple radio frequency links. The method may but is not limited to include the following steps S101 to step S103:

[0051] Step S101: Obtain the interference noise information from the base station to each single-antenna user and at least one channel state matrix, and combine the unknown analog beamforming matrix and digital beamforming matrix to determine the received signal-to-noise ratio model for each single-antenna user.

[0052] Step S102: According to the received signal-to-noise ratio model of each single-antenna user, determine an optimization problem model with the goal of maximizing the sum rate of users.

[0053] Step S103: Use the deep unfolding technique to solve the optimization problem model to obtain the optimal values of the analog beamforming matrix and digital beamforming matrix.

[0054] Steps S101 to S103 shown in the embodiments of the present application, considering the received signal-to-noise ratio of each single-antenna user, construct an optimization problem model aiming to maximize the sum rate of users by using the analog beamforming matrix, the digital beamforming matrix, the interference noise information from the base station to each single-antenna user, and at least one channel state matrix, and then solve it by using the deep unfolding technology, so that the millimeter-wave communication system can achieve robust and efficient hybrid beamforming by means of the optimal values of the obtained analog beamforming matrix and digital beamforming matrix, which is beneficial to improving the communication rate performance of the millimeter-wave communication system and can effectively reduce the computational complexity in solving the hybrid beamforming problem.

[0055] In step S101 of some embodiments, assume that the number of multiple transmit antennas configured by the base station is N, the number of multiple radio frequency links configured by the base station is Na, and the analog beamforming matrix includes the analog transmit control parameters between each transmit antenna and each radio frequency link. The analog beamforming matrix can be denoted as represents the analog transmit control parameter between the nth transmit antenna and the mth radio frequency link included in the analog beamforming matrix, which can be understood as the control parameter of the analog transmitter set between the nth transmit antenna and the mth radio frequency link, including the characteristics (such as frequency, bandwidth, amplitude, phase, etc.) and direction of the transmit signal, where n = 1, 2,..., N and m = 1, 2,..., Na.

[0056] In step S101 of some embodiments, assume that the number of multiple single-antenna users included in the millimeter-wave communication system is K, and the digital beamforming matrix includes the digital transmit control parameters between each radio frequency link and each single-antenna user. The digital beamforming matrix can be denoted as represents the digital transmit control parameter between the mth radio frequency link and the kth single-antenna user included in the digital beamforming matrix, which can be understood as the control parameter of the digital transmitter set between the mth radio frequency link and the kth single-antenna user, including the weight parameters, filter coefficients, and modulation parameters required for processing digital signals, and is used to control the behavior of digital signals during the beamforming process, where k = 1, 2,..., K.

[0057] In step S101 of some embodiments, the determination process of the received signal-to-noise ratio model for each single-antenna user may, but is not limited to, include the following steps S201 to S203:

[0058] Step S201: According to the analog beamforming matrix, the digital beamforming matrix, the interference noise information from the base station to each single-antenna user, and at least one channel state matrix, determine the received signal model for each single-antenna user as follows:

[0059]

[0060] where y k is the received signal of the k-th single-antenna user, h k is the single channel state matrix from the base station to the k-th single-antenna user and H represents the conjugate transpose symbol, s is the transmit coefficient with zero mean and I K is the K-order identity matrix, represents the expectation operator, n k is the interference and noise information from the base station to the k-th single-antenna user, and the corresponding distribution information is described as σ k is the noise variance.

[0061] Step S202: Split the digital beamforming matrix to obtain multiple digital beamforming sub-matrices corresponding to multiple single-antenna users, where the digital beamforming sub-matrix of each single-antenna user includes the digital transmit control parameters between the single-antenna user and each RF link.

[0062] Step S203: According to the digital beamforming sub-matrices of each single-antenna user, decompose and transform the received signal model of each single-antenna user to obtain the received signal-to-noise ratio model of each single-antenna user as follows:

[0063]

[0064] where SINR k is the signal-to-noise ratio of the received signal of the k-th single-antenna user, g Dk is the digital beamforming sub-matrix of the k-th single-antenna user and i.e., the k-th column parameter in the digital beamforming matrix.

[0065] Steps S201 to S203 provided by the embodiments of the present application, through preliminary splitting of the digital beamforming matrix and then combining the analog beamforming matrix, the interference and noise information from the base station to each single-antenna user, and at least one channel state matrix for comprehensive modeling and analysis, can obtain a more reasonable received signal-to-noise ratio model for each single-antenna user, laying a foundation for subsequent construction of a reliable optimization problem model to solve the hybrid beamforming problem.

[0066] In step S102 of some embodiments, the determination process of the optimization problem model with the goal of maximizing the user sum rate may but is not limited to including the following steps S301 to S303:

[0067] Step S301. According to the received signal-to-noise ratio models of individual single-antenna users, the user sum-rate model in this millimeter-wave communication system is determined using the Shannon formula as follows:

[0068]

[0069] In the formula, R(G A ,G D ) is the sum rate of multiple single-antenna users in this millimeter-wave communication system, which can be understood as the sum of the data transmission rates of multiple single-antenna users.

[0070] Step S302. According to this user sum-rate model and the preset constraint conditions regarding the analog beamforming matrix and the digital beamforming matrix, the optimization problem model is determined.

[0071] In this step, the constraint conditions regarding the analog beamforming matrix and the digital beamforming matrix include that the transmit power of the base station jointly determined by the analog beamforming matrix and the digital beamforming matrix is less than or equal to the preset transmit power upper limit value; wherein, the transmit power of the base station is obtained by the following method: determining the F-norm (i.e., Frobenius norm) between the analog beamforming matrix and the digital beamforming matrix, and then taking the square value of the F-norm as the transmit power of the base station, denoted as In addition, considering that the analog beamforming matrix needs to be implemented by phase shifters, in this application, it is defined that the analog beamforming matrix satisfies the unit modulus constraint. On this basis, the optimization problem model is determined as follows:

[0072]

[0073] In the formula, P t is the transmit power upper limit value of the base station, and G A (n,m) are the parameters falling in the nth row and mth column of the analog beamforming matrix.

[0074] Steps S301 to S303 provided in the embodiments of this application, considering power constraints and constant modulus constraints, can obtain a more reasonable optimization problem model through combining the Shannon formula and the received signal-to-noise ratio models of individual single-antenna users for modeling analysis, laying a foundation for subsequent assisting the millimeter-wave communication system to achieve robust and efficient hybrid beamforming.

[0075] In step S103 of some embodiments, regarding the solution process of the optimization problem model using the deep unfolding technique, it may but is not limited to include the following steps S401 to S403:

[0076] Step S401. According to this optimization problem model, the objective function is determined as follows:

[0077]

[0078] In the formula, J is the objective function, D b is the set of channel state matrices, which contains multiple different subsets of channel state matrices. Each subset of channel state matrices contains a single channel state matrix from this base station to each single-antenna user. The summation symbol represents the summation process for the multiple user sum rates calculated corresponding to multiple subsets of channel state matrices.

[0079] Step S402: Randomly initialize the analog beamforming matrix and the digital beamforming matrix, and then input the randomly initialized analog beamforming matrix and digital beamforming matrix into a preset deep unfolding network, which is designed based on the iterative principle of the gradient projection method.

[0080] Step S403: Iteratively train the deep unfolding network according to the objective function and a preset maximum number of training rounds to obtain the optimal values of the analog beamforming matrix and the digital beamforming matrix.

[0081] In this step, the deep unfolding network is iteratively trained according to the objective function until the maximum number of training rounds is reached. Then, the latest values of the analog beamforming matrix and the digital beamforming matrix obtained after performing the last round of training on the deep unfolding network are directly used as the optimal values of the analog beamforming matrix and the digital beamforming matrix for output; in this application, it is preferably set that the maximum number of training rounds is 50.

[0082] Steps S401 to S403 provided by the embodiments of this application can significantly reduce the number of iterative operations of the gradient projection method by using a deep unfolding network integrating the gradient projection method to solve the objective optimization problem, thereby reducing the computational complexity and improving the operation speed.

[0083] In step S403 of some embodiments, each round of training process of the deep unfolding network may, but is not limited to, include the following steps S501 to S503:

[0084] Step S501: Obtain the current values of the analog beamforming matrix and the digital beamforming matrix, and obtain the current values of the first gradient step size and the second gradient step size; wherein, the first gradient step size is used to assist in updating the analog beamforming matrix, and the second gradient step size is used to assist in updating the digital beamforming matrix.

[0085] In this step, if the current execution is the first round of training for the deep unfolding network, the current values of the analog beamforming matrix and the digital beamforming matrix are the randomly initialized values of the analog beamforming matrix and the digital beamforming matrix obtained through the above step S402, and the current values of the first gradient step size and the second gradient step size are both preset values; if the current execution is other rounds of training for the deep unfolding network, the current values of the analog beamforming matrix and the digital beamforming matrix are the latest values of the analog beamforming matrix and the digital beamforming matrix obtained after the previous round of training for the deep unfolding network, and the current values of the first gradient step size and the second gradient step size are the latest values of the first gradient step size and the second gradient step size obtained after the previous round of training for the deep unfolding network.

[0086] Step S502: In the forward propagation stage of the deep unfolding network, use the number of layers of the deep unfolding network as the number of iterations of the gradient projection method, and update the current values of the analog beamforming matrix and the digital beamforming matrix by using the gradient projection method according to the objective function and the current values of the first gradient step size and the second gradient step size.

[0087] In this step, set the deep unfolding network to include H network layers connected in sequence. In the forward propagation process of the h-th network layer, first obtain the current latest values of the analog beamforming matrix and the digital beamforming matrix, the current value of the first gradient sub-step size included in the first gradient step size required for updating the analog beamforming matrix in the h-th network layer, and the current value of the second gradient sub-step size included in the second gradient step size required for updating the digital beamforming matrix in the h-th network layer; then update the current latest value of the analog beamforming matrix according to the objective function, the current value of the first gradient sub-step size, and the current latest value of the digital beamforming matrix; and update the current latest value of the digital beamforming matrix according to the objective function, the current value of the second gradient sub-step size, and the current latest value of the analog beamforming matrix.

[0088] It should be noted that the current latest values of the analog beamforming matrix and the digital beamforming matrix obtained in the forward propagation process of the first network layer are the current values of the analog beamforming matrix and the digital beamforming matrix obtained in the above step S501; the current latest values of the analog beamforming matrix and the digital beamforming matrix obtained in the forward propagation process of other network layers are the current latest values of the analog beamforming matrix and the digital beamforming matrix obtained after the forward propagation of the previous connected network layer.

[0089] Specifically, according to the objective function, the current value of the first gradient sub-step size, and the current latest value of the digital beamforming matrix, the current latest value of the analog beamforming matrix is updated, which can be implemented using the following formula:

[0090]

[0091] In the formula, G A0 is the current latest value of the analog beamforming matrix to be updated, G A1 is the first latest value of the analog beamforming matrix, which is the gradient update result with respect to G A0 , GA1(n, m) is the first latest value of the parameter falling in the n-th row and m-th column of the analog beamforming matrix, is the conjugate gradient of the objective function J with respect to G A0 , μ1(h) is the current value of the first gradient sub-step size required to be applied when updating the analog beamforming matrix during the forward propagation process of the h-th network layer, g Dk0 is the current latest value of the digital beamforming sub-matrix of the k-th single-antenna user to be updated and g Dk0 ∈G D0 , GD0 is the current latest value of the digital beamforming matrix to be updated, G A2 is the second latest value of the analog beamforming matrix, which is the projection result with respect to G A1 , and can also be understood as the final update result with respect to G A0 , GA2(n, m) is the second latest value of the parameter falling in the n-th row and m-th column of the analog beamforming matrix, and both A and B are auxiliary parameters set for convenience of description. In this application, by performing a projection operation on the first latest value G A1 of the analog beamforming matrix, the second latest value G A2 of the analog beamforming matrix can satisfy the constant modulus constraint.

[0092] Specifically, according to the objective function, the current value of the second gradient sub-step size, and the current latest value of the analog beamforming matrix, the current latest value of the digital beamforming matrix is updated, which can be implemented using the following formula:

[0093]

[0094] In the formula, G D1 is the first latest value of the digital beamforming matrix, which is the gradient update result with respect to GD0, is the conjugate gradient of the objective function J with respect to GD0, and μ2(h) is the current value of the second gradient sub-step required to update the digital beamforming matrix during the forward propagation of the h-th network layer. is the conjugate gradient of the objective function J with respect to , where both C and D are auxiliary parameters set for convenience of description, and G D2 is the second latest value of the digital beamforming matrix, which is the projection result with respect to G D1 , and can also be understood as the final update result with respect to GD0. In this application, by performing a projection operation on the first latest value G D1 of the digital beamforming matrix, the second latest value G D2 of the digital beamforming matrix can satisfy the power constraint.

[0095] Step S503: In the backpropagation stage of the deep unfolding network, according to the objective function, update the current values of the first gradient step and the second gradient step, which can be implemented using the following formula:

[0096]

[0097] In the formula, μ1 is the current value of the first gradient step to be updated, which includes the current values of the H first gradient sub-steps required to update the analog beamforming matrix during the forward propagation of the H network layers. is the update result with respect to μ1. is the conjugate gradient of the objective function with respect to μ1, α is a preset gradient update coefficient, μ2 is the current value of the second gradient step to be updated. is the update result with respect to μ2. is the conjugate gradient of the objective function with respect to μ2. In this application, the first gradient step and the second gradient step are hyperparameters of the deep unfolding network, and the Adam optimizer in the deep learning framework PyTorch can be used to update the gradient step during backpropagation.

[0098] In this application, by using the deep unfolding network to synchronously and flexibly update all the gradient steps required during the iterative process of the gradient projection method, the convergence speed of the algorithm can be improved, and the deep unfolding network integrating the gradient projection method still uses the model-driven optimization algorithm in practical applications, which can maintain strong interpretability.

[0099] Please refer to Figure 2 , Figure 2FIG. 0 is an optional schematic structural diagram of an analog-digital hybrid beamforming device applied to a millimeter-wave communication system, which can implement the above-mentioned analog-digital hybrid beamforming method applied to a millimeter-wave communication system. The millimeter-wave communication system includes a base station and multiple single-antenna users. The base station is communicatively connected to the multiple single-antenna users. The base station is configured with multiple transmit antennas and multiple radio frequency links. The device may but is not limited to include the following:

[0100] A first determination module 601, configured to obtain interference noise information from the base station to each single-antenna user and at least one channel state matrix, and determine a received signal-to-noise ratio model for each single-antenna user by combining an unknown analog beamforming matrix and a digital beamforming matrix. Wherein, the analog beamforming matrix includes analog transmission control parameters between each transmit antenna and each radio frequency link, and the digital beamforming matrix includes digital transmission control parameters between each radio frequency link and each single-antenna user.

[0101] A second determination module 602, configured to determine an optimization problem model aiming at maximizing the user sum rate according to the received signal-to-noise ratio model of each single-antenna user.

[0102] A solving module 603, configured to solve the optimization problem model by using a deep unfolding technique to obtain the optimal values of the analog beamforming matrix and the digital beamforming matrix.

[0103] It can be understood that the content in the above method embodiments is applicable to the device embodiments. The functions specifically implemented by the device embodiments are the same as those specifically implemented by the above method embodiments, and the beneficial effects achieved by the device embodiments are also the same as those achieved by the above method embodiments.

[0104] An embodiment of the present application further provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the above-mentioned analog-digital hybrid beamforming method applied to a millimeter-wave communication system. The electronic device may include any intelligent terminal such as a tablet computer or an in-vehicle computer.

[0105] It can be understood that the content in the above method embodiments is applicable to the device embodiments. The functions specifically implemented by the device embodiments are the same as those specifically implemented by the above method embodiments, and the beneficial effects achieved by the device embodiments are also the same as those achieved by the above method embodiments.

[0106] Please refer to Figure 3 , Figure 3 FIG. shows the hardware structure of an electronic device according to another embodiment. The electronic device includes:

[0107] The processor 701 can be implemented in the form of a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, etc., and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application;

[0108] The memory 702 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM), etc. The memory 702 can store an operating system and other application programs. When implementing the technical solutions provided in the embodiments of the present application through software or firmware, the relevant program codes are stored in the memory 702 and are called by the processor 701 to execute the technical solutions provided in the embodiments of the present application;

[0109] The input / output interface 703 is used to implement information input and output;

[0110] The communication interface 704 is used to implement communication interaction between this device and other devices, and can implement communication through a wired manner (such as USB, network cable, etc.) or through a wireless manner (such as mobile network, WIFI, Bluetooth, etc.);

[0111] The bus 705 transmits information between various components of the device (such as the processor 701, the memory 702, the input / output interface 703, and the communication interface 704);

[0112] Among them, the processor 701, the memory 702, the input / output interface 703, and the communication interface 704 are communicatively connected to each other inside the device through the bus 705.

[0113] The embodiments of the present application further provide a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, it implements the above-mentioned analog-digital hybrid beamforming method applied to a millimeter-wave communication system.

[0114] It can be understood that the content in the above method embodiments is applicable to the present storage medium embodiments. The functions specifically implemented by the present storage medium embodiments are the same as those specifically implemented by the above method embodiments, and the beneficial effects achieved by the present storage medium embodiments are also the same as those achieved by the above method embodiments.

[0115] The memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. In addition, the memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state storage devices. In some embodiments, the memory may optionally include a memory remotely located relative to the processor, and these remote memories can be connected to the processor through a network. Examples of the above network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0116] The embodiments described in the embodiments of the present application are for more clearly illustrating the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided by the embodiments of the present application. Those skilled in the art will know that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of the present application are equally applicable to similar technical problems.

[0117] Those skilled in the art can understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than those shown in the figures, or combine certain steps, or different steps.

[0118] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0119] Those of ordinary skill in the art can understand that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, and appropriate combinations thereof.

[0120] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and do not have to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present application described here can be implemented in an order other than those illustrated or described here. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products, or devices.

[0121] It should be understood that in this application, "at least one (item)" means one or more, and "a plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships can exist. For example, "A and / or B" can mean: only A exists, only B exists, and both A and B exist at the same time. Among them, A and B can be singular or plural. The character " / " generally indicates that the associated objects before and after are in an "or" relationship. "At least one (item) of the following" or its similar expression refers to any combination of these items, including any combination of single item (item) or plural items (items). For example, at least one (item) of a, b, or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0122] In several embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the above division of units is only a logical function division. In actual implementation, there can be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection to each other can be through some interfaces. The indirect coupling or communication connection of systems or units can be in electrical, mechanical or other forms.

[0123] The units described above as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place, or they can be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0124] In addition, each functional unit in various embodiments of this application can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0125] When an integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods of various embodiments of this application. The aforementioned storage medium includes: various media that can store programs, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks, or optical discs.

[0126] The preferred embodiments of the embodiments of this application have been described above with reference to the accompanying drawings, but this does not limit the scope of the rights of the embodiments of this application. Any modifications, equivalent replacements, and improvements made by those skilled in the art without departing from the scope and essence of the embodiments of this application shall be within the scope of the rights of the embodiments of this application.

Claims

1. A hybrid analog-digital beamforming method for a millimeter wave communication system, characterized in that: The millimeter wave communication system includes a base station and multiple single-antenna users, the base station and the multiple single-antenna users are connected in communication, and the base station is configured with multiple transmitting antennas and multiple radio frequency links; the method includes: Obtain interference noise information and at least one channel state matrix from the base station to each of the single-antenna users, and determine a received signal-to-noise ratio model of each of the single-antenna users in combination with an unknown analog beamforming matrix and a digital beamforming matrix; Determining an optimization problem model with the goal of maximizing the user sum rate according to a received signal-to-noise ratio model of each single-antenna user; The optimization problem model is solved by using a deep expansion technique to obtain optimal values ​​of the analog beamforming matrix and the digital beamforming matrix; The analog beamforming matrix includes analog transmission control parameters between each of the transmitting antennas and each of the radio frequency links, and the digital beamforming matrix includes digital transmission control parameters between each of the radio frequency links and each of the single-antenna users.

2. The analog-digital hybrid beamforming method applied to a millimeter wave communication system according to claim 1, characterized in that: The received signal-to-noise ratio model of each single-antenna user is obtained in the following manner: Determine a received signal model of each of the single-antenna users according to the analog beamforming matrix, the digital beamforming matrix, interference noise information from the base station to each of the single-antenna users, and at least one channel state matrix; Splitting the digital beamforming matrix to obtain multiple digital beamforming sub-matrices corresponding to the multiple single-antenna users, wherein the digital beamforming sub-matrix of each single-antenna user includes a digital transmission control parameter between the single-antenna user and each of the radio frequency links; According to the digital beamforming sub-matrix of each single-antenna user, the received signal model of each single-antenna user is decomposed and transformed to obtain the received signal-to-noise ratio model of each single-antenna user.

3. The analog-digital hybrid beamforming method applied to a millimeter wave communication system according to claim 1, characterized in that: The optimization problem model is obtained in the following way: According to the received signal-to-noise ratio model of each single-antenna user, the Shannon formula is used to determine the user and rate model in the millimeter wave communication system; The optimization problem model is determined according to the user and rate model and preset constraints between the analog beamforming matrix and the digital beamforming matrix.

4. The analog-digital hybrid beamforming method applied to a millimeter wave communication system according to claim 3, characterized in that: The constraint condition between the analog beamforming matrix and the digital beamforming matrix includes that the transmission power of the base station determined by the analog beamforming matrix and the digital beamforming matrix does not exceed a preset transmission power upper limit value.

5. The analog-digital hybrid beamforming method applied to a millimeter wave communication system according to claim 4, characterized in that: The transmission power of the base station is obtained in the following manner: An F-norm between the analog beamforming matrix and the digital beamforming matrix is ​​determined, and a square value of the F-norm is used as the transmit power of the base station.

6. The analog-digital hybrid beamforming method for millimeter wave communication system according to claim 1, characterized in that: The adopting of the deep expansion technology to solve the optimization problem model to obtain the optimal values ​​of the analog beamforming matrix and the digital beamforming matrix includes: Determine the objective function according to the optimization problem model; Randomly initializing the analog beamforming matrix and the digital beamforming matrix, and then inputting the randomly initialized analog beamforming matrix and the digital beamforming matrix into a preset deep unfolding network, wherein the deep unfolding network is designed based on the iterative principle of the gradient projection method; According to the objective function and a preset maximum number of training rounds, the deep unfolding network is iteratively trained to obtain optimal values ​​of the analog beamforming matrix and the digital beamforming matrix.

7. The analog-digital hybrid beamforming method applied to a millimeter wave communication system according to claim 6, characterized in that: Each round of training process for the deep unfolded network includes: Obtaining current values ​​of the analog beamforming matrix and the digital beamforming matrix; Acquire current values ​​of a first gradient step size and a second gradient step size, wherein the first gradient step size is used to assist in updating the analog beamforming matrix, and the second gradient step size is used to assist in updating the digital beamforming matrix; In the forward propagation stage of the deep unfolding network, the number of layers of the deep unfolding network is used as the number of iterations of the gradient projection method, and the current values ​​of the analog beamforming matrix and the digital beamforming matrix are updated by the gradient projection method according to the objective function and the current values ​​of the first gradient step size and the second gradient step size; In the back propagation phase of the deep expansion network, current values ​​of the first gradient step size and the second gradient step size are updated according to the objective function.

8. An analog-digital hybrid beamforming device applied to a millimeter wave communication system, characterized in that: The millimeter wave communication system includes a base station and multiple single-antenna users, the base station and the multiple single-antenna users are connected in communication, and the base station is configured with multiple transmitting antennas and multiple radio frequency links; the device includes: A first determination module is used to obtain interference noise information and at least one channel state matrix from the base station to each of the single-antenna users, and determine a received signal-to-noise ratio model of each of the single-antenna users in combination with an unknown analog beamforming matrix and a digital beamforming matrix; A second determination module is used to determine an optimization problem model with the goal of maximizing the user sum rate according to the received signal-to-noise ratio model of each single-antenna user; A solution module, used for solving the optimization problem model by using a deep expansion technology to obtain optimal values ​​of the analog beamforming matrix and the digital beamforming matrix; The analog beamforming matrix includes analog transmission control parameters between each of the transmitting antennas and each of the radio frequency links, and the digital beamforming matrix includes digital transmission control parameters between each of the radio frequency links and each of the single-antenna users.

9. An electronic device, characterized in that: The electronic device comprises a memory and a processor, the memory stores a computer program, and the processor implements the method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images