Antenna array beam generation method, device, equipment, medium and product

Abstract

The invention relates to an antenna array beam generation method and device, equipment, a medium and a product. The method comprises the steps of obtaining target far-field sequence data determined based on sampling of a target far-field pattern corresponding to an antenna array, and obtaining a target excitation phase and a target excitation amplitude respectively corresponding to each antenna unit in the antenna array according to the target far-field sequence data and a target forming network corresponding to the antenna array, and controlling beam generation of the antenna array based on the target excitation data. By adopting the method, the antenna array beam generation efficiency can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of antenna array design, and in particular to an antenna array beamforming method, device, equipment, medium and product. Background Art

[0002] Antenna array beamforming technology refers to controlling the beam generated by the antenna array by adjusting the excitation amplitude and excitation phase of each antenna unit in the antenna array, thereby achieving directional transmission of signals and improving the performance of the wireless communication system.

[0003] The antenna array beamforming method in the related art is to obtain the excitation amplitude and excitation phase of each antenna unit by solving an optimization problem, such as minimizing a certain objective function (such as the sidelobe level of beamforming) or satisfying specific constraints (such as the main lobe width) to control the beam generated by the antenna array.

[0004] However, the above antenna array beamforming method has low efficiency. Summary of the invention

[0005] Based on this, it is necessary to provide an antenna array beamforming method, device, equipment, medium and product that can improve efficiency in response to the above technical problems.

[0006] In a first aspect, the present application provides an antenna array beamforming method, comprising:

[0007] Obtaining target far-field sequence data corresponding to the antenna array, where the target far-field sequence data is determined based on sampling of the target far-field pattern corresponding to the antenna array;

[0008] According to the target far-field sequence data and the pre-trained target shaping network, the target excitation data is obtained, wherein the target excitation data includes the target excitation phase and the target excitation amplitude corresponding to each antenna unit in the antenna array;

[0009] Beamforming of the antenna array is controlled based on the target excitation data.

[0010] In one embodiment, the target shaping network includes an output layer and a plurality of bidirectional long short-term memory layers, the plurality of bidirectional long short-term memory layers are connected in sequence, and the output layer is connected to the last bidirectional long short-term memory layer of the plurality of bidirectional long short-term memory layers;

[0011] According to the target far-field sequence data and the pre-trained target shaping network, the target excitation data is obtained, including:

[0012] According to the target far-field sequence data and multiple bidirectional long and short-term memory layers, intermediate excitation data are obtained;

[0013] According to the intermediate excitation data and the output layer, the target excitation data is obtained.

[0014] In one embodiment, the method further comprises:

[0015] According to the measured far-field sequence data and the target shaping network, the measured excitation data is obtained, and the measured far-field sequence data is determined based on the measured far-field pattern of the antenna array under the current excitation data;

[0016] The measured excitation data and the current excitation data are compared and processed to determine the faulty antenna unit in the antenna array.

[0017] In one embodiment, the current excitation data includes the current excitation phase and the current excitation amplitude corresponding to each antenna unit;

[0018] The measured excitation data and the current excitation data are compared and processed to determine the faulty unit in the antenna array, including:

[0019] An antenna unit whose measured excitation data meets a fault condition is determined as a faulty unit; wherein the fault condition includes that the phase difference between the measured excitation phase and the current excitation phase is greater than a phase threshold, or the amplitude difference between the measured excitation amplitude and the current excitation amplitude is greater than an amplitude threshold.

[0020] In one embodiment, the method includes:

[0021] Acquire a plurality of sample far-field sequence data corresponding to the antenna array and sample excitation data corresponding to each sample far-field sequence data;

[0022] Iteratively training the initial shaping network according to a plurality of sample far-field sequence data and sample excitation data corresponding to each sample far-field sequence data to obtain a target shaping network;

[0023] Among them, for each iteration process, the intermediate excitation data corresponding to the sample far-field sequence data is obtained according to the sample far-field sequence data and the intermediate shaping network, and the network parameters of the intermediate shaping network are adjusted according to the intermediate excitation data and the sample excitation data corresponding to the sample far-field sequence data.

[0024] In one embodiment, the method further comprises:

[0025] An initial shaping network is determined according to the antenna type or the number of antenna units corresponding to the antenna array.

[0026] In a second aspect, the present application also provides an antenna array beamforming device, comprising:

[0027] An acquisition module is used to acquire target far-field sequence data corresponding to the antenna array, where the target far-field sequence data is determined based on sampling of the target far-field pattern corresponding to the antenna array;

[0028] A determination module is used to obtain target excitation data according to the target far-field sequence data and the pre-trained target shaping network, wherein the target excitation data includes a target excitation phase and a target excitation amplitude corresponding to each antenna unit in the antenna array;

[0029] The generating module is used for controlling the beam forming of the antenna array based on the target excitation data.

[0030] In a third aspect, the present application further provides a communication device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the method in the first aspect are implemented.

[0031] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the method of the first aspect when the computer program is executed by a processor.

[0032] In a fifth aspect, the present application also provides a computer program product, including a computer program, which implements the steps of the method in the first aspect when executed by a processor.

[0033] The above-mentioned antenna array beamforming method, device, equipment, medium and product obtain the target far-field sequence data determined by sampling the target far-field pattern corresponding to the antenna array, and obtain the target excitation phase and target excitation amplitude corresponding to each antenna unit in the antenna array according to the target far-field sequence data and the target shaping network corresponding to the antenna array, and control the beamforming of the antenna array based on the target excitation data; in this way, on the one hand, based on the target far-field sequence data formed by sampling the target far-field pattern that the antenna array needs to achieve, the target far-field sequence data is inferred through the pre-trained target shaping network, and the target excitation phase and target excitation amplitude corresponding to each antenna unit in the antenna array are quickly obtained, and the antenna array is quickly controlled to generate the corresponding beam, avoiding the problem of low efficiency of antenna array beamforming caused by solving the optimization problem to obtain the excitation amplitude and excitation phase of each antenna unit in the related technology. The method provided by the above-mentioned embodiment can quickly provide an approximate optimal solution using the target shaping model that has been trained offline, reduce the calculation time, and improve the efficiency of determining the antenna array excitation data. On the other hand, when the far-field pattern of the antenna array changes, the far-field sampling data corresponding to the changed far-field pattern is input into the target shaping network, and new excitation data can be quickly obtained, avoiding the need to readjust the objective function or constraint conditions, recalculate and solve the problem in the related technology by optimizing the problem solving method, resulting in a slow speed of solving the excitation data when the far-field pattern of the antenna array changes in real time, resulting in a low efficiency of the antenna array beam generation. The method provided by the above embodiment can improve the efficiency of the antenna array beam generation when the far-field pattern of the antenna array changes in real time. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related technologies, the drawings required for use in the embodiments or the related technical descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0035] Figure 1 A diagram of an application environment of an antenna array beamforming method in an embodiment;

[0036] Figure 2 is a schematic flow chart of an antenna array beamforming method in one embodiment;

[0037] Figure 3 A schematic diagram of the structure of a target shaping network in one embodiment;

[0038] Figure 4 A schematic flow chart of the steps of obtaining target incentive data in one embodiment;

[0039] Figure 5 is a schematic flow chart of an antenna array beamforming method in another embodiment;

[0040] Figure 6 is a structural block diagram of an antenna array beamforming device in one embodiment;

[0041] Figure 7 FIG. 4 is a diagram showing the internal structure of a communication device in one embodiment. DETAILED DESCRIPTION

[0042] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0043] The antenna array beam forming method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown in the figure, the terminal 102 communicates with the network device 104 through the network. The terminal 102 may also be referred to as a user equipment (UE for short), and the terminal 102 may be various mobile devices, for example, a mobile phone (or a "cellular" phone), a computer with a mobile terminal, etc., or a portable, pocket-sized, handheld, computer-built-in or vehicle-mounted mobile device, or various Internet of Things devices. The network device 104 may be a device for communicating with the terminal 102, for example, it may be a base station (Base Transceiver Station, BTS) in a Global System for Mobile Communications (GSM) system or a Code Division Multiple Access (CDMA) system, or a base station (NodeB, NB) in a Wideband Code Division Multiple Access (WCDMA) system, or an evolved base station (Evolutional Node B, eNB or eNodeB) in a Long Term Evolution (LTE) system, or a base station (the next Generation Node B, gNB) in a 5th Generation New Radio (5G NR) system, or a base station used in a 6G mobile communication system; the present application does not limit the terminal 102 and the network device 104.

[0044] In an exemplary embodiment, Figure 2 As shown, an antenna array beam forming method is provided, and the method is applied to Figure 1 The network device 104 in FIG. 1 is used as an example for explanation. It can be understood that the method can also be applied to the terminal 102. In this embodiment, the method includes the following steps 202 to 204. Among them:

[0045] Step 202: acquiring target far-field sequence data corresponding to the antenna array, where the target far-field sequence data is determined based on sampling of the target far-field pattern corresponding to the antenna array.

[0046] Among them, the directivity of a single antenna is limited. In order to be suitable for applications in various occasions, two or more single antennas working at the same frequency are fed and spatially arranged according to certain requirements to form an antenna array, which can also be called an antenna array; a single antenna constituting the antenna array can be called an array element, or an antenna unit, which is called an antenna unit in this embodiment.

[0047] Far-field sequence data refers to the sampled data corresponding to the far-field pattern. Exemplarily, the far-field pattern is sampled at a sampling interval of 1° to obtain the far-field phase and far-field amplitude corresponding to each sampling point as the far-field sequence data of the antenna array. Exemplarily, the sampling interval can be greater than 1° or less than 1°. Among them, the far-field pattern refers to a graph of the relative field strength (normalized modulus) of the radiation field varying with direction at a certain distance from the antenna. It is a graphical description of the antenna radiation characteristics and can also be called an antenna radiation pattern or an antenna pattern. Exemplarily, the angle range of the far-field pattern is -90° to 90°, wherein the midpoint of the antenna array is taken as the origin, and the normal direction perpendicular to the antenna array is 0° of the far-field pattern. The far-field pattern can be in polar coordinate form or in two-dimensional rectangular coordinate form.

[0048] Among them, the target far-field sequence data refers to the data obtained by sampling the target far-field pattern; the target far-field pattern refers to the target that the radiation field of the current antenna array needs to reach; this embodiment needs to infer the target excitation data applied to the antenna array based on the target far-field pattern.

[0049] In one possible implementation, the target far-field pattern is sampled at a preset sampling interval to obtain the far-field phase and far-field amplitude corresponding to each sampling point, and each far-field phase and far-field amplitude is normalized to obtain target far-field sequence data, so that the data scale corresponding to each sampling point in the target far-field sequence data is consistent, that is, all are between 0 and 1, so that the target shaping network extracts the relative relationship between the data corresponding to each sampling point, thereby improving the accuracy of the target shaping network reasoning.

[0050] Step 204, obtaining target excitation data according to the target far-field sequence data and the pre-trained target shaping network; wherein the target excitation data includes the target excitation phase and the target excitation amplitude corresponding to each antenna unit in the antenna array.

[0051] Among them, the target shaping network is a neural network with memory function and capable of learning the spatial order relationship of input sequence data. In a possible implementation, the target shaping network includes a sequence data feature extraction structure and a mapping structure. Exemplarily, the sequence data feature extraction structure can be implemented based on a recursive neural network (RNN). Exemplarily, the sequence data extraction structure is implemented using a bidirectional long short-term memory network (Bi-directional Long Short-Term Memory, BiLSTM) or a gated recurrent unit (GRU) and other networks; the mapping structure is used to map the positional relationship between the antenna units in the antenna array learned by the sequence data feature extraction structure from the remote sequence data into the excitation data of the antenna array. Exemplarily, the mapping structure can be implemented using a fully connected layer.

[0052] In a possible implementation, the input data dimension of the target shaping network corresponds to the number of sampling points of the far-field pattern, and the output data dimension of the target shaping network corresponds to the number of antenna elements of the antenna array. Exemplarily, the input data dimension of the target shaping network is the same as the number of sampling points of the far-field pattern, and the output data dimension of the target shaping network is the same as the number of antenna elements of the antenna array.

[0053] Among them, a target shaping network is obtained by pre-training with a large amount of far-field sequence data corresponding to known far-field patterns and excitation data corresponding to each far-field pattern, so that the target shaping network can accurately output the excitation data corresponding to the antenna array according to the input far-field sequence data.

[0054] Step 206: Control the beamforming of the antenna array based on the target excitation data.

[0055] The transmission signal of the corresponding antenna unit is adjusted according to the target excitation phase and the target excitation amplitude to control the beam generated by the antenna array.

[0056] In the above-mentioned antenna array beamforming method, target far-field sequence data determined by sampling the target far-field pattern corresponding to the antenna array is obtained, and the target excitation phase and target excitation amplitude corresponding to each antenna unit in the antenna array are obtained according to the target far-field sequence data and the target shaping network corresponding to the antenna array, and the beamforming of the antenna array is controlled based on the target excitation data; in this way, on the one hand, based on the target far-field sequence data formed by sampling the target far-field pattern that the antenna array needs to achieve, the target far-field sequence data is inferred through a pre-trained target shaping network, and the target excitation phase and target excitation amplitude corresponding to each antenna unit in the antenna array are quickly obtained, and the antenna array is quickly controlled to generate the corresponding beam, thereby avoiding the related technology of obtaining the excitation amplitude and excitation phase of each antenna unit by solving the optimization problem, and the calculation process is highly complex and the collection is complicated. The problem of low efficiency of antenna array beamforming caused by slow convergence speed is solved by adopting the method provided in the above embodiment and using the target shaping network that has been trained offline to quickly provide an approximate optimal solution, reduce the calculation time, and improve the efficiency of determining the antenna array excitation data; on the other hand, when the far-field pattern of the antenna array changes, the far-field sampling data corresponding to the changed far-field pattern is input into the target shaping network, so that new excitation data can be quickly obtained, avoiding the need to readjust the objective function or constraint conditions, recalculate and solve the problem in the related technology through optimization problem solving, resulting in a slow speed for solving the excitation data when the far-field pattern of the antenna array changes in real time, resulting in low efficiency of antenna array beamforming. The method provided in the above embodiment can improve the efficiency of antenna array beamforming when the far-field pattern of the antenna array changes in real time.

[0057] In an exemplary embodiment, please refer to Figure 3 The target shaping network includes an output layer and multiple bidirectional long short-term memory layers, wherein the multiple bidirectional long short-term memory layers are connected in sequence, and the output layer is connected to the last bidirectional long short-term memory layer of the multiple bidirectional long short-term memory layers.

[0058] Among them, a bidirectional long short-term memory layer includes two long short-term memory sublayers, one long short-term memory sublayer processes the input sequence data from the forward direction, and the other long short-term memory sublayer processes the input sequence data from the reverse direction. After the processing is completed, the processing results of the two long short-term memory sublayers are spliced ​​or combined to obtain the output data of the bidirectional long short-term memory layer. For example, Figure 3 As shown, BiLSTM represents a bidirectional long short-term memory layer, from network parameters C0, H0 to network parameters C M ,H M The processing along the solid arrow direction is the positive long short-term memory sublayer, from network parameters C0, H0 to network parameters C M ,H MThe processing process along the dotted arrow direction is the reverse long short-term memory sublayer. LSTM cell represents the long short-term memory unit, M represents the number of sampling points corresponding to the target far-field sequence data, and FP m represents the far-field phase of the mth sampling point, FA m represents the far-field amplitude of the mth sampling point; N represents the N antenna elements, SP n represents the excitation phase of the nth antenna element, SA n Represents the excitation amplitude of the nth antenna element.

[0059] In this embodiment, multiple bidirectional long short-term memory layers are used as the sequence data feature extraction structure, which can provide more neurons and learnable parameters, improve the effect of sequence data feature extraction, and achieve better incentive data prediction effect. In a possible implementation, Figure 3 As shown, the input data dimension and output data dimension of each bidirectional long short-term memory layer are the same as the number of sampling points corresponding to the target far-field sequence data. The output layer serves as a feature mapping structure, and the input data dimension of the output layer is the same as the output data dimension of the bidirectional long short-term memory layer. The output data dimension of the output layer is the same as the number of antenna units of the antenna array.

[0060] Please refer to Figure 4 In this embodiment, the process of obtaining target excitation data according to the target far-field sequence data and the pre-trained target shaping network includes steps 402 to 404, wherein:

[0061] Step 402, obtaining intermediate excitation data according to the target far-field sequence data and a plurality of bidirectional long short-term memory layers.

[0062] Among them, the target far-field sequence data is input into the first bidirectional long short-term memory layer to obtain the feature extraction data output by the first bidirectional long short-term memory layer, and then the feature extraction data is input into the second bidirectional long short-term memory layer, until the last bidirectional long short-term memory layer, to obtain the intermediate excitation data.

[0063] Exemplarily, the target far-field sequence data is a 2×M matrix, each column in the matrix corresponds to a sampling point, and each column includes two data, one is the far-field phase, and the other is the far-field amplitude.

[0064] Step 404, obtaining target excitation data according to the intermediate excitation data and the output layer.

[0065] Among them, the intermediate excitation data is input into the output layer to obtain the target excitation data output by the output layer.

[0066] Exemplarily, the target excitation data is a 2×N matrix, each column in the matrix corresponds to an antenna unit, and each column includes two data, one is the excitation phase, and the other is the excitation amplitude.

[0067] In the antenna array beamforming method provided in the above embodiment, the target shaping network utilizes the bidirectional long short-term memory layer with bidirectional information processing and sequence modeling capabilities to capture the position information between each antenna unit. By making full use of the spatial connection between each antenna unit, it can effectively understand and predict the behavior and interaction of the antenna units, perform fast and accurate feature extraction, and improve the accuracy of antenna array excitation data prediction.

[0068] In an exemplary embodiment, the target shaping network includes a bidirectional long short-term memory layer and an output layer.

[0069] In an exemplary embodiment, please refer to Figure 5 The antenna array beamforming method provided further includes steps 502 to 504, wherein:

[0070] Step 502, obtaining measured excitation data according to the measured far-field sequence data and the target shaping network, wherein the measured far-field sequence data is determined based on the measured far-field pattern of the antenna array under the current excitation data.

[0071] Step 504: performing comparison processing based on the measured excitation data and the current excitation data to determine the faulty antenna unit in the antenna array.

[0072] In this embodiment, the antenna array can be fault-checked by using a trained target shaping network. Based on the far-field pattern currently measured, the trained target shaping network is used to infer the excitation data corresponding to the far-field pattern, and the excitation data is compared with the excitation data currently set for the antenna array to quickly identify the faulty unit.

[0073] In a possible implementation manner, the current excitation data is pre-set excitation data and is used to perform fault detection on the antenna array.

[0074] In another possible implementation, the current excitation data is also determined by the target shaping network. Exemplarily, the current excitation data is the target excitation data; exemplarily, based on the acquired far-field pattern A1, the far-field sequence data B1 is determined, and based on the far-field sequence data B1 and the target shaping network, the excitation data C1 is obtained, and the antenna array is configured according to the excitation data C1; however, during the application process, it is found that the performance of the antenna array does not meet expectations or the measured far-field pattern A2 is different from the far-field pattern A1, it is believed that an antenna unit in the antenna array may have a fault, and based on the measured far-field pattern A2, the far-field sequence data B2 is obtained, and based on the far-field sequence data B2 and the target shaping network, the excitation data C2 is obtained, and the excitation data C1 and the excitation data C2 are compared to determine the faulty unit in the antenna array.

[0075] Exemplarily, the method of obtaining the measured far-field pattern includes: rotating the sample around the antenna array during the sampling process, or at least rotating the antenna array itself, to obtain the measured far-field pattern.

[0076] In a possible implementation, the current excitation data includes the current excitation phase and the current excitation amplitude corresponding to each antenna unit, wherein the process of determining the faulty unit in the antenna array by comparing the target excitation data with the current excitation data includes: determining the antenna unit whose target excitation data meets the fault condition as the faulty unit. The fault condition includes that the phase difference between the target excitation phase and the current excitation phase is greater than the phase threshold, or the amplitude difference between the target excitation amplitude and the current excitation amplitude is greater than the amplitude threshold.

[0077] Exemplarily, an antenna unit whose phase difference between the target excitation phase and the current excitation phase is greater than a phase threshold is determined as a faulty unit.

[0078] As another example, an antenna unit whose amplitude difference between the target excitation amplitude and the current excitation amplitude is greater than an amplitude threshold is determined as a faulty unit.

[0079] In the above implementation, the faulty unit is determined according to a parameter in the excitation amplitude or the excitation phase in the excitation data, thereby improving the speed of fault detection.

[0080] In a possible implementation, an antenna unit whose phase difference between a target excitation phase and a current excitation phase is greater than a phase threshold and whose amplitude difference between a target excitation amplitude and a current excitation amplitude is greater than an amplitude threshold is determined as a faulty unit.

[0081] In an exemplary embodiment, the provided antenna array beamforming method involves a training process of a target shaping model. Figure 5 , the process includes steps 502 to 504, wherein:

[0082] Step 502: Acquire a plurality of sample far-field sequence data corresponding to the antenna array and sample excitation data corresponding to each sample far-field sequence data.

[0083] In a possible implementation, multiple groups of sample far-field sequence data and corresponding sample excitation data are determined according to a mathematical shaping model of the antenna array.

[0084] For example, the antenna array is a linearly arranged phased array antenna including N antenna units. The mathematical shaping model of the antenna array can be expressed as Formula 1. The far-field phase FP can be calculated based on the excitation amplitude and excitation phase of each antenna unit. m and the far-field amplitude FA m , where m refers to the mth sampling point in the far-field sequence data.

[0085] ; Formula 1

[0086] in, Refers to the far-field pattern, which is in complex form. The far-field phase FP can be obtained by processing m and the far-field amplitude FA m , N represents N antenna units, SP n represents the excitation phase of the nth antenna element, SA n represents the excitation amplitude of the nth antenna unit; the value range of θ is -90° to 90°, where the midpoint of the antenna array is taken as the origin and the normal direction perpendicular to the array is 0°.

[0087] Among them, different mathematical shaping models are used for different antenna arrays to determine the sample far-field sequence data and the corresponding sample excitation data corresponding to the current antenna array; illustratively, the mathematical shaping model corresponding to the antenna array with coupling between antenna units, or the antenna array with unequal spacing between antenna units is different from the mathematical shaping model shown in the above formula 1. In this way, different antenna arrays select the corresponding training sample data to obtain the target shaping network corresponding to the antenna array, thereby improving the accuracy of the target shaping network.

[0088] In a possible implementation, the excitation amplitude and the excitation phase are normalized to obtain sample excitation data, and the far-field phase and the far-field amplitude are normalized to obtain sample far-field phase and sample far-field amplitude to obtain sample far-field sequence data.

[0089] Exemplarily, the far-field phase and far-field amplitude obtained according to the mathematical shaping model of the antenna array are normalized to obtain the sample far-field phase and the sample far-field amplitude The process is shown in Formula 2 and Formula 3 respectively:

[0090] ; Formula 2

[0091] . Formula 3

[0092] The excitation amplitude of the antenna unit is defined in the range of [0, 1], and no normalization is required. Exemplarily, the process of normalizing the excitation phase corresponding to each antenna unit to obtain the sample excitation phase includes: first, taking the excitation phase of the first antenna unit as a reference, calculating the relative excitation phase , as shown in Formula 5, and then normalized to obtain the sample excitation phase , as shown in Formula 6:

[0093] , Formula 5

[0094] . Formula 6

[0095] Step 504, iteratively training the initial shaping network according to the plurality of sample far-field sequence data and the sample excitation data corresponding to each sample far-field sequence data, to obtain a target shaping network.

[0096] Among them, for each iteration process, the intermediate excitation data corresponding to the sample far-field sequence data is obtained according to the sample far-field training sequence data and the intermediate shaping network, and the network parameters of the intermediate shaping network are adjusted according to the intermediate excitation data and the sample excitation data corresponding to the sample far-field sequence data.

[0097] In a possible implementation, initial values ​​of parameters such as weight parameters, bias parameters, and hyperparameters (for example, learning rate, regularization parameters, and related parameters in the Adam optimization algorithm) in the initial shaping network are obtained in a random manner.

[0098] In a possible implementation manner, the initial shaping network is determined according to the antenna type or the number of antenna units corresponding to the antenna array.

[0099] Among them, the depth of the sequence data feature extraction structure in the initial shaping network corresponding to different antenna types is different; exemplarily, the number of bidirectional long short-term memory layers in the initial shaping network corresponding to different embodiment types is different; exemplarily, the number of bidirectional long short-term memory layers corresponding to the sparse array of lines with unequal spacing is greater than the number of bidirectional long short-term memory layers corresponding to the linear array with equal spacing; exemplarily, the number of bidirectional long short-term memory layers corresponding to the antenna array with mutual coupling between antenna units is greater than the number of bidirectional long short-term memory layers corresponding to the antenna array without mutual coupling; Among them, the sparse array of lines with unequal spacing or the antenna array with coupling means that the far-field radiation of each antenna unit will have a greater impact on each other, that is, the synthesized far-field amplitude and far-field phase are more complex. Using a larger number of bidirectional long short-term memory layers can increase the learning ability of the network and improve the accuracy of antenna array beam generation.

[0100] Exemplarily, the number of bidirectional long and short-term memory layers corresponding to an antenna array with a larger number of antenna units is greater than the number of bidirectional long and short-term memory layers corresponding to an antenna array with a smaller number of antenna units. The more antenna units there are, the greater the influence between the far-field radiation of each antenna unit. By adopting a larger number of bidirectional long and short-term memory layers, the learning ability of the network can be increased and the accuracy of antenna array beam generation can be improved.

[0101] In a possible implementation, during the iterative training process of the initial shaping network, the loss function F is as shown in Formula 7:

[0102] . Formula 7

[0103] In a possible implementation, online learning and incremental training can be used in the iterative training of the initial shaping network, and data can be input while the data is accurate, so as to prepare data and train the network at the same time, thereby improving training efficiency.

[0104] In an exemplary embodiment, an antenna array beamforming method is provided, wherein the method is applied to Figure 1 The network device 104 in FIG. 1 is used as an example for explanation. It can be understood that the method can also be applied to the terminal 102. In this embodiment, the method includes the following steps S2 to S12. Among them:

[0105] Step S2, obtaining target far-field sequence data corresponding to the antenna array, where the target far-field sequence data is determined based on sampling of the target far-field pattern corresponding to the antenna array.

[0106] Step S4, obtaining intermediate excitation data according to the target far-field sequence data and multiple bidirectional long and short-term memory layers of the target shaping network.

[0107] Step S6, obtaining target excitation data according to the intermediate excitation data and the output layer of the target shaping network; wherein the target excitation data includes the target excitation phase and the target excitation amplitude corresponding to each antenna unit in the antenna array.

[0108] Step S8: controlling the beamforming of the antenna array based on the target excitation data.

[0109] Step S10, obtaining measurement excitation data according to the measurement far-field sequence data and the target shaping network. The measurement excitation far-field sequence data is determined based on the measurement far-field pattern of the antenna array under the current excitation data.

[0110] Step S12, determining the antenna unit whose measured excitation data meets the fault condition as a faulty unit; wherein the fault condition includes that the phase difference between the measured excitation phase and the current excitation phase is greater than the phase threshold, or the amplitude difference between the measured excitation amplitude and the current excitation amplitude is greater than the amplitude threshold.

[0111] Optionally, the target shaping network is trained by the following steps: obtaining multiple sample far-field sequence data corresponding to the antenna array and sample excitation data corresponding to each sample far-field sequence data; iteratively training the initial shaping network according to the multiple sample far-field sequence data and the sample excitation data corresponding to each sample far-field sequence data to obtain the target shaping network; wherein, for each iterative process, according to the sample far-field sequence data and the intermediate shaping network, intermediate excitation data corresponding to the sample far-field sequence data are obtained, and according to the intermediate excitation data and the sample excitation data corresponding to the sample far-field sequence data, the network parameters of the intermediate shaping network are adjusted.

[0112] Optionally, the initial shaping network is determined according to the antenna type or the number of antenna units corresponding to the antenna array.

[0113] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0114] It is understood that the term "based on" used in this application is used to describe one or more factors that affect the determination, and does not exclude other factors that may affect the determination. For example, the phrase "determine A based on B" means that the determination of A can be based entirely or at least partially on factor B, that is, B is a factor that affects the determination of A, but does not exclude that the determination of A is also based on C.

[0115] Based on the same inventive concept, the embodiment of the present application also provides an antenna array beam forming device for implementing the antenna array beam forming method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations of one or more antenna array beam forming device embodiments provided below can refer to the limitations of the antenna array beam forming method above, and will not be repeated here.

[0116] In an exemplary embodiment, Figure 6 As shown, an antenna array beam forming device is provided, including: an acquisition module 602 and a determination module 604, wherein:

[0117] An acquisition module 602 is used to acquire target far-field sequence data corresponding to the antenna array, where the target far-field sequence data is determined based on a sampling of a target far-field pattern corresponding to the antenna array;

[0118] The determination module 604 is used to obtain target excitation data according to the target far-field sequence data and the pre-trained target shaping network; wherein the target excitation data includes the target excitation phase and the target excitation amplitude corresponding to each antenna unit in the antenna array;

[0119] The generating module 606 is configured to control the beam forming of the antenna array based on the target excitation data.

[0120] In an exemplary embodiment, the target shaping network includes an output layer and multiple bidirectional long short-term memory layers, the multiple bidirectional long short-term memory layers are connected in sequence, and the output layer is connected to the last bidirectional long short-term memory layer of the multiple bidirectional long short-term memory layers; the determination module 604 is used to obtain intermediate excitation data based on the target far-field sequence data and the multiple bidirectional long short-term memory layers; and the target excitation data is obtained based on the intermediate excitation data and the output layer.

[0121] In an exemplary embodiment, the determination module 604 is also used to obtain measured excitation data based on the measured far-field sequence data and the target shaping network, and the measured far-field sequence data is determined based on the measured far-field radiation pattern of the antenna array under the current excitation data; the device also includes a detection module, which is used to perform comparative processing based on the measured excitation data and the current excitation data to determine the faulty antenna unit in the antenna array.

[0122] In an exemplary embodiment, the current excitation data includes the current excitation phase and the current excitation amplitude corresponding to each antenna unit respectively; the detection module is used to determine the antenna unit whose measured excitation data meets the fault condition as a faulty unit; wherein the fault condition includes that the phase difference between the measured excitation phase and the current excitation phase is greater than the phase threshold, or the amplitude difference between the measured excitation amplitude and the current excitation amplitude is greater than the amplitude threshold.

[0123] In an exemplary embodiment, the device also includes a training module for obtaining multiple sample far-field sequence data corresponding to the antenna array and sample excitation data corresponding to each sample far-field sequence data; iteratively training the initial shaping network according to the multiple sample far-field sequence data and the sample excitation data corresponding to each sample far-field sequence data to obtain a target shaping network; wherein, for each iterative process, according to the sample far-field sequence data and the intermediate shaping network, intermediate excitation data corresponding to the sample far-field sequence data are obtained, and according to the intermediate excitation data and the sample excitation data corresponding to the sample far-field sequence data, the network parameters of the intermediate shaping network are adjusted.

[0124] In an exemplary embodiment, the training module is used to determine an initial shaping network according to the antenna type or the number of antenna units corresponding to the antenna array.

[0125] Each module in the above antenna array beamforming device can be implemented in whole or in part by software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each module.

[0126] In an exemplary embodiment, a communication device is provided. The computer device can be a base station, a terminal, or a server, and its internal structure diagram can be shown in Figure Y. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program library. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, an antenna array beam forming method is implemented.

[0127] Those skilled in the art will understand that Figure 7 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0128] In one embodiment, a communication device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

[0129] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0130] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0131] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0132] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0133] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0134] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be construed as limiting the scope of the present application. It should be noted that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. An antenna array beamforming method, characterized in that: The method comprises: Acquire target far-field sequence data corresponding to the antenna array, wherein the target far-field sequence data is determined based on sampling of the target far-field pattern corresponding to the antenna array; According to the target far-field sequence data and the pre-trained target shaping network, target excitation data is obtained, wherein the target excitation data includes a target excitation phase and a target excitation amplitude corresponding to each antenna unit in the antenna array; Beamforming of the antenna array is controlled based on the target excitation data.

2. The method according to claim 1, characterized in that The target shaping network includes an output layer and a plurality of bidirectional long short-term memory layers, the plurality of bidirectional long short-term memory layers are connected in sequence, and the output layer is connected to the last bidirectional long short-term memory layer of the plurality of bidirectional long short-term memory layers; The step of obtaining target excitation data according to the target far-field sequence data and a pre-trained target shaping network includes: Obtaining intermediate excitation data according to the target far-field sequence data and the plurality of bidirectional long and short-term memory layers; The target excitation data is obtained according to the intermediate excitation data and the output layer.

3. The method according to claim 1, characterized in that The method further comprises: According to the measured far-field sequence data and the target shaping network, the measured excitation data are obtained, wherein the measured far-field sequence data is determined based on the measured far-field pattern of the antenna array under the current excitation data; A comparison process is performed based on the measured excitation data and the current excitation data to determine a faulty antenna unit in the antenna array.

4. The method according to claim 3, characterized in that The current excitation data includes the current excitation phase and the current excitation amplitude corresponding to each antenna unit, and the measured excitation data includes the measured excitation phase and the measured excitation amplitude corresponding to each antenna unit; The comparing and processing the measured excitation data and the current excitation data to determine the faulty unit in the antenna array includes: The antenna unit whose measured excitation data meets the fault condition is determined as the faulty unit; wherein the fault condition includes that the phase difference between the measured excitation phase and the current excitation phase is greater than a phase threshold, or the amplitude difference between the measured excitation amplitude and the current excitation amplitude is greater than an amplitude threshold.

5. The method according to claim 1, characterized in that The method comprises: Acquire a plurality of sample far-field sequence data corresponding to the antenna array and sample excitation data corresponding to each of the sample far-field sequence data; Performing iterative training processing on the initial shaping network according to the plurality of sample far-field sequence data and the sample excitation data respectively corresponding to each of the sample far-field sequence data, to obtain the target shaping network; Among them, for each iterative process, the intermediate excitation data corresponding to the sample far-field sequence data is obtained according to the sample far-field sequence data and the intermediate shaping network, and the network parameters of the intermediate shaping network are adjusted according to the intermediate excitation data and the sample excitation data corresponding to the sample far-field sequence data.

6. The method according to claim 5, characterized in that The method further comprises: The initial shaping network is determined according to the antenna type or the number of antenna units corresponding to the antenna array.

7. An antenna array beam forming device, characterized in that: The device comprises: An acquisition module, used to acquire target far-field sequence data corresponding to the antenna array, wherein the target far-field sequence data is determined based on a sampling of a target far-field pattern corresponding to the antenna array; A determination module is used to obtain target excitation data according to the target far-field sequence data and a pre-trained target shaping network, wherein the target excitation data includes a target excitation phase and a target excitation amplitude corresponding to each antenna unit in the antenna array; A generating module is used to control the beam forming of the antenna array based on the target excitation data.

8. A communication device, comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images