Method and device for determining optimal beam and electronic equipment

Abstract

The invention discloses a method and device for determining an optimal beam and electronic equipment. The method comprises the steps that first state information of target equipment at a first moment is acquired, and the target equipment is a base station or an unmanned aerial vehicle; using a prediction model to predict the first state information to obtain an initial beam at a second moment, the second moment being a next moment of the first moment; determining a similar beam of the initial beam from a codebook of the target device, and determining an initial beam pair according to the initial beam and the similar beam; and determining an optimal beam of the target device in each radio frequency link according to the initial beam pair, the optimal beam being a beam configuration with the best communication performance. According to the method and the device, the technical problems that the initial communication quality after each time of beam switching is unstable due to random initialization of a traditional heuristic algorithm, and beam tracking and stable communication are not facilitated are solved.

Description

Technical Field

[0001] The present application relates to the field of wireless communications, and in particular, to a method, device, and electronic device for determining an optimal beam. Background Art

[0002] In low-altitude communication scenarios like drone inspections and live broadcasts, images must be transmitted back in real time. Especially when multiple drones are performing swarm missions, not only must communication between the drones and the base station be guaranteed, but data exchange between drones must also be ensured. This places high demands on network communication speed and stability. Due to the high speeds of drones, base stations must track and update communication beams in real time to avoid communication stalls or interruptions caused by high latency. Traditional heuristic algorithms, due to random initialization, result in unstable initial communication quality after each beam switch, hindering beam tracking and stable communication.

[0003] To address the above-mentioned problems, no effective solutions have been proposed so far. Summary of the Invention

[0004] The embodiments of the present application provide a method, device, and electronic device for determining an optimal beam, so as to at least solve the technical problem that the traditional heuristic algorithm has unstable initial communication quality after each beam switching due to random initialization, which is not conducive to beam tracking and stable communication.

[0005] According to one aspect of an embodiment of the present application, a method for determining an optimal beam is provided, including: obtaining first state information of a target device at a first moment, wherein the target device is a base station or a drone; using a prediction model to predict the first state information to obtain an initial beam at a second moment, wherein the second moment is the next moment of the first moment; determining a similar beam of the initial beam from a code book of the target device, and determining an initial beam pair based on the initial beam and the similar beam; determining an optimal beam of the target device in each radio frequency link based on the initial beam pair, wherein the optimal beam is a beam configuration with the best communication performance.

[0006] Optionally, the prediction model is trained in the following manner: obtaining a training set for training the prediction model, wherein the training set includes: first historical state information of the base station, second historical state information of the drone, and historical communication quality data between the base station and the drone; using the initial prediction model to predict the training set to obtain historical prediction results; determining a total loss function based on the historical prediction results, the first historical state information and the historical communication quality data; and iteratively training the initial prediction model based on the total loss function to obtain a prediction model.

[0007] Optionally, the second historical status information includes the first longitude, first latitude, first altitude of the UAV, the pitch angle, roll angle, yaw angle of the UAV, and the relative azimuth between the UAV and the base station, wherein the relative azimuth is determined by: obtaining the second longitude, second latitude and second altitude corresponding to the base station, and obtaining the equatorial radius and ellipsoidal flattening; determining a first value based on the first latitude, equatorial radius and ellipsoidal flattening, and determining a second value based on the second latitude, equatorial radius and ellipsoidal flattening; determining the first geocentric earth-fixed coordinate corresponding to the UAV based on the first value, the first longitude, the first latitude and the first altitude, and determining the second value based on the second value, the second longitude , the second latitude and the second altitude, determine the second geocentric earth-fixed coordinates corresponding to the base station, wherein the first geocentric earth-fixed coordinates include the coordinates of the UAV in the X-axis, Y-axis and Z-axis directions, and the second geocentric earth-fixed coordinates include the coordinates of the base station in the X-axis, Y-axis and Z-axis directions; determine the coordinate difference based on the first geocentric earth-fixed coordinates and the second geocentric earth-fixed coordinates, wherein the coordinate difference includes the coordinate difference between the UAV and the base station in the X-axis, Y-axis and Z-axis directions; determine the transformed coordinates of the UAV based on the coordinate difference and the base station rotation matrix, wherein the base station rotation matrix is ​​determined by the second longitude and the second latitude; determine the relative azimuth based on the transformed coordinates and the coordinate origin.

[0008] Optionally, based on the historical prediction results, the first historical state information and the historical communication quality data, a total loss function is determined, including: obtaining the first predicted azimuth, the first predicted downtilt, and the first predicted rate at time t in the historical prediction results, obtaining the true azimuth and the true downtilt in the first historical state information, and obtaining the true rate in the historical communication quality data, wherein t is a positive integer; determining the main loss function based on the first predicted azimuth, the first predicted downtilt, the true azimuth and the true downtilt; determining the rate loss function based on the predicted rate and the true rate; obtaining The second predicted azimuth and the second predicted downtilt at time t-1 in the historical prediction results; determine the beam smoothness constraint loss function based on the first predicted azimuth, the second predicted azimuth, the first predicted downtilt and the second predicted downtilt; determine the effective coverage constraint loss function based on the first predicted azimuth, the preset azimuth maximum value, the preset azimuth minimum value, the first predicted downtilt, the preset downtilt maximum value and the preset downtilt minimum value; determine the total loss function based on the main loss function, the rate loss function, the beam smoothness constraint loss function and the effective coverage constraint loss function.

[0009] Optionally, using a prediction model to predict the first state information to obtain an initial beam at a second moment includes: using a prediction model to predict the first state information to obtain second state information at a second moment; and determining the initial beam at a second moment based on the second state information.

[0010] Optionally, determining a similar beam to the initial beam from a codebook of the target device includes: determining a similarity between the initial beam and each beam in the codebook of the target device to obtain a similarity set; and determining a beam with the highest similarity in the similarity set as a similar beam to the initial beam.

[0011] Optionally, the optimal beam of the target device in each RF link is determined based on the initial beam pair, including: step 1: obtaining the first beam pair corresponding to the current moment, and obtaining the adjacent beam pair of the first beam pair; step 2: determining the first spectrum efficiency corresponding to the first beam pair, and determining the second spectrum efficiency corresponding to the adjacent beam pair; step 3: when the difference between the second spectrum efficiency and the first spectrum efficiency is greater than or equal to a preset threshold, determining the beam pair change rate based on the first spectrum efficiency, the second spectrum efficiency and the initial beam pair; step 4: determining the intermediate value based on the beam pair change rate, the attenuation coefficient, and the anti-zero coefficient; step 5: determining the second beam pair corresponding to the next moment after the current moment based on the intermediate value, the beam pair change rate, and the first beam pair, and determining the second beam pair as the first beam pair; step 6: repeating steps 1 to 5 until the difference between the second spectrum efficiency and the first spectrum efficiency is less than the preset threshold, stopping the iteration, and determining the second beam pair obtained for the last time as the optimal beam.

[0012] According to another aspect of an embodiment of the present application, a device for determining an optimal beam is also provided, including: an acquisition module for acquiring first state information of a target device at a first moment, wherein the target device is a base station or a drone; a prediction module for using a prediction model to predict the first state information to obtain an initial beam at a second moment, wherein the second moment is the next moment of the first moment; a first determination module for determining a similar beam of the initial beam from a code book of the target device, and determining an initial beam pair based on the initial beam and the similar beam; a second determination module for determining the optimal beam of the target device in each radio frequency link based on the initial beam pair, wherein the optimal beam is the beam configuration with the best communication performance.

[0013] According to another aspect of the embodiments of the present application, an electronic device is also provided, including: a memory for storing program instructions; a processor, connected to the memory, for executing program instructions to implement the following functions: obtaining first state information of a target device at a first moment, wherein the target device is a base station or a drone; using a prediction model to predict the first state information to obtain an initial beam at a second moment, wherein the second moment is the next moment of the first moment; determining a similar beam of the initial beam from a code book of the target device, and determining an initial beam pair based on the initial beam and the similar beam; determining the optimal beam of the target device in each radio frequency link based on the initial beam pair, wherein the optimal beam is the beam configuration with the best communication performance.

[0014] According to another aspect of the embodiments of the present application, a non-volatile storage medium is provided, which includes a stored computer program, wherein the device where the non-volatile storage medium is located executes the above-mentioned method for determining the optimal beam by running the computer program.

[0015] According to another aspect of the embodiments of the present application, a computer program product is provided, including computer instructions, which implement the above-mentioned method for determining the optimal beam when executed by a processor.

[0016] In an embodiment of the present application, first state information of a target device at a first moment is obtained, wherein the target device is a base station or a drone; a prediction model is used to predict the first state information to obtain an initial beam at a second moment, wherein the second moment is the next moment of the first moment; a similar beam of the initial beam is determined from a code book of the target device, and an initial beam pair is determined based on the initial beam and the similar beam; an optimal beam of the target device in each radio frequency link is determined based on the initial beam pair, wherein the optimal beam is a beam configuration with the best communication performance, thereby achieving the purpose of fast and accurate beam tracking, thereby realizing the technical effect of reducing communication delay, and further solving the technical problem that the traditional heuristic algorithm has unstable initial communication quality after each beam switching due to random initialization, which is not conducive to beam tracking and stable communication. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0018] Figure 1 is a hardware structure block diagram of a computer terminal for implementing a method for determining an optimal beam according to an embodiment of the present application;

[0019] Figure 2 is a flowchart of a method for determining an optimal beam according to an embodiment of the present application;

[0020] Figure 3 1 is a memory unit structure diagram of a ConvLSTM according to an embodiment of the present application;

[0021] Figure 4 is a schematic diagram of a heuristic beam search according to an embodiment of the present application;

[0022] Figure 5 A comparison chart of the number of searches for this application and traversal searches;

[0023] Figure 6 is a performance curve comparison diagram according to an embodiment of the present application;

[0024] Figure 7 This is a structural diagram of a device for determining an optimal beam according to an embodiment of the present application. DETAILED DESCRIPTION

[0025] In order to enable those skilled in the art to better understand the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of this application.

[0026] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in a sequence other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0027] The information collected in the embodiments of the present application is information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data comply with the relevant laws, regulations and standards of the relevant regions, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entrances for users to choose to authorize or reject the automated decision results; if the user chooses to reject, the expert decision-making process will be entered.

[0028] First, some nouns or terms that appear in the process of explaining the embodiments of this application are subject to the following explanations:

[0029] Discrete Fourier Transform (DFT): Used to analyze the frequency components of discrete signals in time or space. The DFT converts a discrete, periodic, or finite-length signal into a series of complex coefficients that describe the amplitude and phase of the different frequency components in the signal.

[0030] DFT codebook: In the field of wireless communications, the DFT codebook is a set of beam weighting coefficients designed based on the DFT principle to guide the beamforming of the antenna array. Each weighting coefficient in the DFT codebook corresponds to a specific beam direction. By adjusting these coefficients, the direction of the beam can be controlled so that it is maximally directed toward the target receiver while suppressing interference in other directions. In this application, the DFT codebook is used as an efficient beam search library, combined with prediction and search algorithms to quickly find the optimal beam configuration suitable for communication between drones and base stations.

[0031] Long Short-Term Memory (LSTM) is a special type of recurrent neural network designed to overcome the vanishing gradient problem of traditional RNNs and is suitable for processing and predicting time series data. By introducing memory cells and three gating mechanisms—input gate, forget gate, and output gate—LSTM can selectively remember or forget information, effectively learning long-term dependencies. This makes it suitable for tasks that require understanding the temporal context of data, such as speech recognition and time series prediction.

[0032] Convolutional Neural Networks (CNNs): A deep learning model designed for processing grid-structured data, such as images. Convolutional layers capture local spatial features, pooling layers reduce data dimensionality, and fully connected layers perform classification or regression.

[0033] Mean-Square Error (MSE): It is a commonly used indicator for evaluating model prediction accuracy. It is used to measure the average square difference between the predicted value and the true value. It is sensitive to the size of the error and is suitable for scenarios where data needs to be precisely aligned.

[0034] Mean Absolute Error (MAE): It is a commonly used indicator for evaluating model prediction accuracy. It is used to calculate the average absolute value of the difference between the predicted value and the true value. It is less sensitive to outliers and provides an intuitive understanding of the error.

[0035] In related technologies, base station deployment frequency bands have high path loss, and millimeter wave frequency band sites have already been deployed. In the post-5G and 6G eras, base stations will need to rely on ultra-large-scale antenna arrays to achieve beamforming gain to combat path loss. The greater the number of antennas, the more numerous and narrower the beams, requiring base stations to spend more time tracking and searching for the optimal beam pair to maintain high-speed, low-latency communications. Related technologies are unable to meet the requirements for fast beam tracking between base stations and drones.

[0036] In order to achieve low-latency and fast beam tracking of the UAV by the base station and maintain stable high-speed communication between the base station and the UAV, the embodiment of the present application provides a method for determining the optimal beam, which can be run on Figure 1 Among the computer terminals shown, the computer terminal will be described below.

[0037] The method for determining the optimal beam provided in the embodiments of the present application can be executed in a mobile terminal, a computer terminal, or a similar computing device. Figure 1 FIG1 shows a hardware structure block diagram of a computer terminal for implementing a method for determining an optimal beam. Figure 1 As shown, the computer terminal 10 may include one or more processors (illustrated as 102a, 102b, ..., 102n in the figure) (the processor may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 104 for storing data, and a transmission module 106 for communication functions via a wired and / or wireless network connection. In addition, it may also include: a display, a keyboard, a cursor control device, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the I / O interface), a network interface, and a BUS bus. Those skilled in the art will understand that Figure 1 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 1 More or fewer components than shown, or with Figure 1 Different configurations shown.

[0038] It should be noted that the one or more processors and / or other data processing circuits described above may generally be referred to herein as "data processing circuitry." The data processing circuitry may be embodied in whole or in part as software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuitry may be a single, independent processing module, or may be fully or partially integrated into any of the other components of the computer terminal 10. As discussed in the embodiments of the present application, the data processing circuitry functions as a processor control (e.g., the selection of a variable resistor terminal path connected to an interface).

[0039] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the method for determining the optimal beam in the embodiments of the present application. The processor executes the software programs and modules stored in the memory 104 to perform various functional applications and data processing, thereby implementing the above-mentioned method for determining the optimal beam. The memory 104 may include high-speed random access memory and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include memory remotely located relative to the processor, and these remote memories may be connected to the computer terminal 10 via a network. Examples of such networks include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0040] The transmission module 106 is configured to receive or transmit data via a network. A specific example of the aforementioned network may include a wireless network provided by the communications provider of the computer terminal 10. In one embodiment, the transmission module 106 includes a network interface controller (NIC), which can be connected to other network devices via a base station to enable communication with the Internet. In another embodiment, the transmission module 106 may be a radio frequency (RF) module, which is configured to communicate with the Internet wirelessly.

[0041] The display may be, for example, a touch screen liquid crystal display (LCD) that enables a user to interact with a user interface of the computer terminal 10 .

[0042] It should be noted that, in some optional embodiments, the above Figure 1 The computer terminal shown may include hardware elements (including circuits), software elements (including computer code stored on a computer-readable medium), or a combination of hardware elements and software elements. Figure 1 This is merely one example of a particular embodiment and is intended to illustrate the types of components that may be present in the computer terminal described above.

[0043] In the above-mentioned operating environment, an embodiment of the present application provides an embodiment of a method for determining an optimal beam. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0044] Figure 2 is a flow chart of a method for determining an optimal beam according to an embodiment of the present application, such as Figure 2As shown, the method includes the following steps:

[0045] Step S202: Acquire first status information of a target device at a first moment, where the target device is a base station or a drone.

[0046] In step S202 above, the first status information may include location information, attitude and orientation, and communication status. If the target device is a drone, the location information in the first status information includes the drone's current latitude, longitude, and altitude. The attitude and orientation include the drone's pitch, roll, and yaw angles, among others. The communication status includes key indicators of the health of the communication link between the base station and the drone, namely, parameters related to communication quality, such as reference signal received power (RSRP), signal-to-interference-plus-noise ratio (SINR), and data transmission rate (or uplink rate). If the target device is a base station, the location information in the first status information includes the base station's latitude, longitude, and altitude. The attitude and orientation include the antenna's azimuth and downtilt angles, indicating the direction of the base station's radiation. The communication status includes parameters related to communication quality.

[0047] The purpose of obtaining the above-mentioned first state information at the first moment is to enable the subsequent beam tracking algorithm to make predictions and searches based on the latest and most accurate data, ensuring that the communication between the drone and the base station is always maintained in the best state.

[0048] Step S204: Use the prediction model to predict the first state information to obtain an initial beam at a second moment, where the second moment is a moment next to the first moment.

[0049] In step S204, the prediction model is a trained ConvLSTM model that predicts the ideal beam direction based on temporal and spatial characteristics, serving as the starting point for subsequent beam searches. Specifically, the prediction model learns from historical data to understand the dynamic relationship between device state and beam direction. The prediction model receives the target device's first state at a first moment as input and, through its internal computational mechanisms (such as convolution operations to capture spatial features and recurrence mechanisms to understand temporal dependencies), predicts the communication conditions at a second moment. The predicted result is the initial beam (representing the theoretically optimal initial beam direction). This predicted initial beam serves as the starting point for a heuristic search, upon which the algorithm further adjusts the beam direction until the optimal beam is found, ensuring efficient and stable communication between the base station and the drone.

[0050] Step S206: Determine a similar beam to the initial beam from the codebook of the target device, and determine an initial beam pair based on the initial beam and the similar beam.

[0051] In step S206, the codebook is a set of predefined beam weighting coefficients used to generate different beam directions. In wireless communication systems, especially in scenarios using large-scale antenna arrays, codebooks provide an efficient, mathematically based beamforming method.

[0052] The above codebook can be a DFT codebook, which is a set of beam weighting coefficients designed based on the DFT principle. The first beam Weighting coefficient of the root antenna as follows:

[0053]

[0054] Among them, j represents an imaginary number, is the number of beams, is the number of antennas, the base station has beams, the drone has beams.

[0055] Due to the complexity of actual application scenarios and the limitations of model prediction, the initial beam predicted by the model may not exactly match the beam in the DFT codebook. By matching similar beams, we can find the beam configuration closest to the predicted result, obtain similar beams, and form an initial beam pair based on the initial beam and similar beams.

[0056] It should be noted that when determining the similar beam of the initial beam, the codebook used can be the DFT codebook of the target device, or other forms of codebooks corresponding to the target device. The above-mentioned DFT codebook is only used as an example of one of the codebooks and does not limit the scheme. In actual use, other codebooks can also be used to determine the similar beam of the initial beam.

[0057] Step S208 : determining the optimal beam of the target device in each radio frequency link based on the initial beam pair, wherein the optimal beam is the beam configuration with the best communication performance.

[0058] In step S208, based on the initial beam pair, a heuristic search algorithm is used to further optimize the beam configuration and determine the optimal beam for the target device on each RF link. An optimal beam is defined as the beam configuration that provides optimal communication performance (e.g., highest spectral efficiency, lowest latency, maximum data rate, etc.) under given communication conditions. This process essentially uses local search and iterative optimization based on model predictions to find the beam that truly maximizes communication quality.

[0059] Through the above steps S202 to S208, the purpose of fast and accurate beam tracking is achieved. In the scenario of a rapidly moving drone, this method can quickly respond to changes in the drone's position and reduce the beam search time, thereby significantly reducing communication latency and improving communication stability. Not only does it optimize the communication between the drone and the base station, it also lays the foundation for efficient communication of multiple subsequent RF links, avoiding communication interruptions or quality degradation caused by slow beam search processes. It effectively solves the problem of unstable communication performance caused by random initialization in traditional heuristic algorithms. By predicting and optimizing the beam configuration, it ensures the communication quality after each beam switch, providing a more stable and high-speed communication experience for mobile devices such as drones. The following is an explanation.

[0060] In the above-mentioned method for determining the optimal beam, the prediction model is trained in the following manner: obtaining a training set for training the prediction model, wherein the training set includes: first historical state information of the base station, second historical state information of the drone, and historical communication quality data between the base station and the drone; using the initial prediction model to predict the training set to obtain historical prediction results; determining a total loss function based on the historical prediction results, the first historical state information, and the historical communication quality data; and iteratively training the initial prediction model based on the total loss function to obtain a prediction model.

[0061] In some embodiments of the present application, LSTM, as a commonly used time series prediction model, can process time series data well, but the local features of spatial data are strong, and LSTM cannot characterize this feature, so it is not suitable for scenarios where the data has spatial characteristics. Similarly, CNN, as a commonly used model for processing spatial data, can extract local features of spatial data well, but cannot effectively capture the dependencies of time series data, and is not suitable for scenarios where the data has temporal characteristics. In scenarios where the communication data between base stations and drones has both temporal and spatial characteristics, ConvLSTM, which combines LSTM and CNN architectures, is used for beam prediction. ConvLSTM abandons the fully connected layer, introduces convolution in space, and loops in time, while retaining the memory unit and gating mechanism of LSTM, which can capture temporal and spatial characteristics, taking into account both the long-term dependencies of time series data and the local features of spatial structure, and can effectively process spatiotemporal data. Figure 3 This is the memory unit structure diagram of ConvLSTM. The data is processed in the following order:

[0062] (1)

[0063] (2)

[0064] (3)

[0065] (4)

[0066] (5)

[0067] (6)

[0068] in, is convolution, Is the convolution kernel, different subscripts in W represent the convolution kernels used in different processes, is the bias vector, and different subscripts in b represent the bias vectors used in different processes. / is the hidden layer state at the previous moment / current moment, is the current input data, is the sigmoid function, is the hyperbolic tangent function, is a temporary cell state, is the cell state at the current time t, is the cell state at the previous moment t-1, 、 、 、 、 All calculated The intermediate value of . Equations (1), (2) to (4), (5) to (6) are the calculation processes of the forget gate, input gate, and output gate, respectively. Among them, the forget gate determines the information retained at the previous moment, the input gate determines the new information, and the output gate determines the output information.

[0069] When training the prediction model, a training set is first obtained. The training set contains historical base station state information, including its location (longitude, latitude, and altitude), antenna configuration parameters (such as azimuth and downtilt, and beamwidth). The training set also contains historical drone state information, including its location (longitude, latitude, and altitude), flight attitude information (such as pitch, roll, yaw, and azimuth relative to the base station). Historical communication quality data includes reference signal received power (RSRP), signal-to-interference-plus-noise ratio (SINR), and data rate. An untrained initial prediction model is used to predict the data in the training set. The model attempts to predict the communication quality or beam direction between the base station and drone based on the historical state information of the base station and drone, generating historical prediction results. A total loss function is defined based on the difference between the historical prediction results and the actual historical communication quality data and historical state information. Minimizing this total loss function guides model training, ensuring that the model closely matches the actual data. Using a backpropagation algorithm and an optimizer, the model parameters are iteratively adjusted based on the gradient of the total loss function, enabling continuous learning and optimization. This process continues until the model's predictive ability reaches a satisfactory level or the loss function converges to a stable value. In this way, after multiple iterations of training, the initial prediction model is trained to obtain a predictive model. For example, after 35 rounds of training using 3 million data points, the total loss function value of the initial prediction model gradually decreases, and the prediction accuracy ultimately increases to 86.7%.

[0070] The entire training process aims to use a large amount of historical data to enable the prediction model to understand the complex interactions between base stations and drones and predict results that conform to physical laws and environmental characteristics. In actual applications, when receiving new base station and drone status information, the model can quickly give relatively accurate predictions to guide rapid beam tracking and optimize communication quality.

[0071] In the above-mentioned method for determining the optimal beam, the second historical state information includes the first longitude, the first latitude, the first altitude of the UAV, the pitch angle, the roll angle, the yaw angle of the UAV, and the relative azimuth between the UAV and the base station, wherein the relative azimuth is determined by: obtaining the second longitude, the second latitude, and the second altitude corresponding to the base station, and obtaining the equatorial radius and the ellipsoidal flattening; determining a first value based on the first latitude, the equatorial radius, and the ellipsoidal flattening, and determining a second value based on the second latitude, the equatorial radius, and the ellipsoidal flattening; determining the first geocentric earth-fixed coordinate corresponding to the UAV based on the first value, the first longitude, the first latitude, and the first altitude, and determining the second value based on the second value , the second longitude, the second latitude and the second altitude, determine the second geocentric earth-fixed coordinates corresponding to the base station, wherein the first geocentric earth-fixed coordinates include the coordinates of the UAV in the X-axis, Y-axis and Z-axis directions, and the second geocentric earth-fixed coordinates include the coordinates of the base station in the X-axis, Y-axis and Z-axis directions; determine the coordinate difference based on the first geocentric earth-fixed coordinates and the second geocentric earth-fixed coordinates, wherein the coordinate difference includes the coordinate difference between the UAV and the base station in the X-axis, Y-axis and Z-axis directions; determine the transformed coordinates of the UAV based on the coordinate difference and the base station rotation matrix, wherein the base station rotation matrix is ​​determined by the second longitude and the second latitude; determine the relative azimuth based on the transformed coordinates and the coordinate origin.

[0072] In some embodiments of the present application, in order to extract spatial features, the coordinates of the drone need to be converted to a local coordinate system with the base station as the origin. Specifically, let the second longitude, second latitude, and second altitude of the base station be 、 、 , the first longitude, first latitude, and first altitude of the drone are 、 、 First, convert the latitude, longitude, and altitude (i.e., longitude, latitude, and altitude, referred to as latitude and altitude) coordinates of the base station and the drone into Earth-centered Earth-fixed coordinates, and then convert them into a station-centered coordinate system with the base station as the origin. The east (E) axis of the station-centered coordinate system points eastward along the latitude circle, the north (N) axis points northward along the longitude circle, and the celestial (U) axis points vertically upward from the ground. Calculate the radius of curvature of the celestial circle using the following formula: :

[0073] (7)

[0074] (8)

[0075] in, is the equatorial radius, is the ellipsoid flattening. When , the formula (7) Indicates the first latitude, and the obtained is the first value mentioned above, and the corresponding When , the formula (7) Indicates the second latitude, and the obtained is the second value mentioned above.

[0076] The base station and the UAV calculate their respective Earth-centered Earth-fixed coordinates according to equations (9) to (11):

[0077] (9)

[0078] (10)

[0079] (11)

[0080] Among them, when calculating the first earth-centered earth-fixed coordinates corresponding to the UAV, the formulas (9)-(11) represents the first value, Indicates the first height, represents the first latitude, Indicates the first longitude. Substituting the data corresponding to each letter into formula (9)-(11), the X, Y, and Z obtained are the above-mentioned first geocentric earth-fixed coordinates. When calculating the second geocentric earth-fixed coordinates corresponding to the base station, the represents the second value, Indicates the second height, represents the second latitude, Represents the second longitude. Substituting the data corresponding to each letter into formula (9)-(11), the obtained X, Y, and Z are the above-mentioned second geocentric Earth-fixed coordinates.

[0081] According to the first Earth-centered Earth-fixed coordinate and the second Earth-centered Earth-fixed coordinate, the coordinate difference is determined. Specifically, let 、 、 is the difference in the Earth-centered and Earth-fixed coordinates between the UAV and the base station, 、 、 They represent the X-axis, Y-axis, and Z-axis coordinates of the first Earth-fixed coordinate system corresponding to the drone. 、 、 Respectively represent the coordinates of the X-axis, Y-axis, and Z-axis in the second earth-centered earth-fixed coordinate system corresponding to the base station. After constructing the base station rotation matrix (i.e., the first matrix on the right side of equation 12), the coordinate difference is processed according to equation (12) to obtain the ENU coordinates of the drone in the station-centered coordinate system:

[0082] (12)

[0083] After obtaining the ENU coordinates of the UAV in the station center coordinate system (i.e. the converted coordinates mentioned above), the ENU coordinates of the base station in the station center coordinate system are used as the coordinate origin. Based on the converted coordinates and the coordinate origin, the relative azimuth angle between the UAV and the base station can be obtained.

[0084] In the above steps, the total loss function is determined based on the historical prediction results, the first historical state information and the historical communication quality data, including: obtaining the first predicted azimuth, the first predicted downtilt, and the first predicted rate at time t in the historical prediction results, obtaining the true azimuth and the true downtilt in the first historical state information, and obtaining the true rate in the historical communication quality data, wherein t is a positive integer; determining the main loss function based on the first predicted azimuth, the first predicted downtilt, the true azimuth and the true downtilt; determining the rate loss function based on the predicted rate and the true rate; and obtaining Take the second predicted azimuth and the second predicted downtilt at time t-1 in the historical prediction results; determine the beam smoothness constraint loss function based on the first predicted azimuth, the second predicted azimuth, the first predicted downtilt and the second predicted downtilt; determine the effective coverage constraint loss function based on the first predicted azimuth, the preset maximum azimuth, the preset minimum azimuth, the first predicted downtilt, the preset maximum downtilt and the preset minimum downtilt; determine the total loss function based on the main loss function, the rate loss function, the beam smoothness constraint loss function and the effective coverage constraint loss function.

[0085] In some embodiments of the present application, in order to improve the beam prediction accuracy of ConvLSTM, the total loss function used in the prediction model is Including: main loss function, rate loss function , physical constraint loss function, environment adaptive weight , the main loss function includes the angle smoothness loss function And cosine similarity loss function , the physical constraint loss function includes the beam smoothness constraint loss function and effective coverage constraint loss function The details are as follows:

[0086] (13)

[0087] (14)

[0088] (15)

[0089] (16)

[0090] (17)

[0091] (18)

[0092] (19)

[0093] (20)

[0094] in, / is the true azimuth / true downtilt, / is the predicted azimuth / predicted downtilt angle, which can be the first predicted azimuth / first predicted downtilt angle corresponding to time t, is the difference between the true azimuth and the predicted azimuth, is the difference between the actual downtilt angle and the predicted downtilt angle, is the error threshold, is the cosine balance coefficient, is the environmental weight, It is the environmental threshold for adjusting the angle prediction weight, SINR represents the signal to interference plus noise ratio, is the rate coefficient to prevent the main loss from being disturbed, / is the true / predicted rate, Here, it can be the first predicted rate corresponding to time t, / and / They are The first predicted azimuth angle / first predicted downtilt angle at time t-1, the second predicted azimuth angle / second predicted downtilt angle at time t-1, / It is the smoothing coefficient to suppress beam hopping and ensure communication stability. / is the penalty coefficient for exceeding the physical limit, Indicates the preset minimum azimuth angle. Indicates the preset maximum azimuth angle. Indicates the preset minimum downtilt angle. Represents the preset maximum downtilt angle, and RELU represents the activation function. It combines MSE and MAE, using MSE for small errors and MAE for large errors to reduce the impact of outliers; Adjusting the angle prediction weight according to the environment can reduce the impact of interference on the model.

[0095] In step S204 of the above-mentioned method for determining the optimal beam, a prediction model is used to predict the first state information to obtain an initial beam at the second moment, including: using the prediction model to predict the first state information to obtain second state information at the second moment; and determining the initial beam at the second moment based on the second state information.

[0096] In some embodiments of the present application, the first state information is input into a trained prediction model (such as a ConvLSTM model). Based on its internal learning mechanism and previously trained data, the model predicts the target device's likely state at a second moment, i.e., the second state information. Based on the second state information, the model calculates or selects a preliminary beam direction that is deemed most likely to provide optimal communication quality under the current predicted environment, thereby determining the initial beam at the second moment.

[0097] In step S206 of the above-mentioned method for determining the optimal beam, a similar beam to the initial beam is determined from the codebook of the target device, including: determining the similarity between the initial beam and each beam in the codebook of the target device to obtain a similarity set; and determining the beam with the highest similarity in the similarity set as the similar beam to the initial beam.

[0098] In some embodiments of the present application, the base station and the drone use their own codebooks (such as DFT codebooks) to communicate. Assuming that the target device is the base station, let the initial beam at the second moment predicted by the model be , the beam in the DFT codebook is .because Limited accuracy cannot be compared with Complete match, beam similarity needs to be calculated , use with Most similar As a similar beam. When the target device is a drone, the processing method is the same as the base station side. The closer it is to 1, the higher the beam similarity.

[0099] (twenty one)

[0100] The similarity between the initial beam and each beam in the codebook (such as DFT codebook) of the target device is determined according to formula (21), a similarity set is obtained, and the beam with the highest similarity in the similarity set is determined as the similar beam of the initial beam.

[0101] In step S208 of the above-mentioned method for determining the optimal beam, the optimal beam of the target device in each radio frequency link is determined based on the initial beam pair, including: step 1: obtaining a first beam pair corresponding to the current moment, and obtaining an adjacent beam pair of the first beam pair; step 2: determining a first spectral efficiency corresponding to the first beam pair, and determining a second spectral efficiency corresponding to the adjacent beam pair; step 3: when the difference between the second spectral efficiency and the first spectral efficiency is greater than or equal to a preset threshold, determining a beam pair change rate based on the first spectral efficiency, the second spectral efficiency, and the initial beam pair; step 4: determining an intermediate value based on the beam pair change rate, an attenuation coefficient, and an anti-zero coefficient; step 5: determining a second beam pair corresponding to a moment next to the current moment based on the intermediate value, the beam pair change rate, and the first beam pair, and determining the second beam pair as the first beam pair; step 6: repeating steps 1 to 5 until the difference between the second spectral efficiency and the first spectral efficiency is less than the preset threshold, stopping the iteration, and determining the second beam pair obtained last as the optimal beam.

[0102] In some embodiments of the present application, the base station and the drone open the antennas of the first radio frequency link, the base station uses the transmitting beam of the predicted beam pair to send the reference signal, and the drone uses the receiving beam of the predicted beam pair to receive the reference signal. Figure 4 As shown in the figure, each point is a beam pair, and a square red dot beam search area is formed with the blue dot predicted beam pair as the center. The drone uses the blue dot as the starting point and uses the heuristic algorithm to determine the beam pair for the next communication and feedback to the base station. After receiving the information, the base station uses the beam to send a reference signal. The drone uses the heuristic algorithm to determine the beam pair for the next communication and feedback to the base station. The above process is repeated until the optimal beam for the first RF link between the base station and the drone is determined, that is, and The codeword used by the first sub-matrix is, / The base station and the drone open the second RF link and determine and The codeword used by the second sub-array is repeated until the and The codewords used by all sub-matrices in . The heuristic algorithm is as follows:

[0103] (twenty two)

[0104] (twenty three)

[0105] (twenty four)

[0106] (25)

[0107] in, is the first beam pair corresponding to the current moment, is the beam pair change rate, is the adjacent beam pair of the first beam pair, , is the initial beam pair, represents the first spectrum efficiency, represents the second spectrum efficiency, is the attenuation coefficient, which takes a value of 0.85. Is the anti-zero coefficient, the value , 、 、 represents the middle value, Iterate until the spectral efficiency of the beam pair is less than the preset threshold (such as ), the second beam pair obtained last time is determined as the optimal beam, and the optimal beam pair is used for communication.

[0108] Spectral efficiency (abbreviated as spectrum efficiency) is used to evaluate communication quality. The calculation formula is as follows:

[0109] (26)

[0110] in, / is the (total) number of RF links of the base station / drone, is the signal power, is the noise variance, / is the simulated beam matrix of the base station / drone, H represents the channel matrix, I represents the unit matrix, and the dimension of the unit matrix is ​​based on Sure.

[0111] Since the scheme in the embodiment of the present application has no closed-form expression for complexity, the algorithm complexity of the actual simulation is given. Figure 5 This figure compares the number of searches performed by this application and traversal search, with the search count averaged over 5,000 independent searches. The simulation configurations are: a base station with 256 antennas, 4 RF links, and a fully connected architecture; a drone with 16 antennas, 2 links, and a fully connected architecture. The figure shows approximately 110 billion traversal searches, while the solution in this application only performed 15 searches. Compared to the optimal traversal search, the beam switching solution proposed in this embodiment reduces complexity by 99.9999%. Figure 6 This is a performance curve comparison diagram according to an embodiment of the present application. Figure 6 It can be seen that the performance of the solution of this application is about 99% of the optimal performance, which has obvious advantages.

[0112] Figure 7 is a structural diagram of an optimal beam determination device according to an embodiment of the present application, such as Figure 7 As shown, the device includes:

[0113] an acquisition module 40, configured to acquire first state information of a target device at a first moment, wherein the target device is a base station or a drone;

[0114] A prediction module 42 is configured to predict the first state information using a prediction model to obtain an initial beam at a second moment, where the second moment is a moment subsequent to the first moment;

[0115] A first determining module 44 is configured to determine a similar beam to the initial beam from a codebook of the target device, and determine an initial beam pair based on the initial beam and the similar beam;

[0116] The second determination module 46 is configured to determine the optimal beam of the target device in each radio frequency link according to the initial beam pair, wherein the optimal beam is a beam configuration with the best communication performance.

[0117] Through the acquisition module, prediction module, first determination module and second determination module in the above-mentioned optimal beam determination device, the purpose of fast and accurate beam tracking is achieved, thereby realizing the technical effect of reducing communication delay, and further solving the technical problem that the traditional heuristic algorithm has unstable initial communication quality after each beam switching due to random initialization, which is not conducive to beam tracking and stable communication.

[0118] The above-mentioned device for determining the optimal beam also includes a training module 48, which is used to train the prediction model. Specifically, the prediction model is trained in the following manner: obtaining a training set for training the prediction model, wherein the training set includes: the first historical state information of the base station, the second historical state information of the drone, and the historical communication quality data between the base station and the drone; using the initial prediction model to predict the training set to obtain a historical prediction result; determining the total loss function based on the historical prediction result, the first historical state information and the historical communication quality data; and iteratively training the initial prediction model based on the total loss function to obtain a prediction model.

[0119] In the training module of the above-mentioned optimal beam determination device, the second historical state information includes the first longitude, first latitude, first altitude of the UAV, the pitch angle, roll angle, yaw angle of the UAV, and the relative azimuth between the UAV and the base station, wherein the training module is also used to determine the relative azimuth. Specifically, the relative azimuth is determined by: obtaining the second longitude, second latitude and second altitude corresponding to the base station, and obtaining the equatorial radius and ellipsoidal flattening; determining a first value based on the first latitude, equatorial radius and ellipsoidal flattening, and determining a second value based on the second latitude, equatorial radius and ellipsoidal flattening; determining the first value corresponding to the UAV based on the first value, the first longitude, the first latitude and the first altitude. Geocentric Earth-fixed coordinates, determining the second geocentric Earth-fixed coordinates corresponding to the base station based on the second value, the second longitude, the second latitude and the second altitude, wherein the first geocentric Earth-fixed coordinates include the coordinates of the UAV in the X-axis, Y-axis and Z-axis directions, and the second geocentric Earth-fixed coordinates include the coordinates of the base station in the X-axis, Y-axis and Z-axis directions; determining the coordinate difference based on the first geocentric Earth-fixed coordinates and the second geocentric Earth-fixed coordinates, wherein the coordinate difference includes the coordinate difference between the UAV and the base station in the X-axis, Y-axis and Z-axis directions; determining the transformed coordinates of the UAV based on the coordinate difference and the base station rotation matrix, wherein the base station rotation matrix is ​​determined by the second longitude and the second latitude; determining the relative azimuth based on the transformed coordinates and the coordinate origin.

[0120] In the training module in the above-mentioned optimal beam determination device, the training module is also used to obtain the first predicted azimuth, the first predicted downtilt, and the first predicted rate at time t in the historical prediction results, obtain the true azimuth and the true downtilt in the first historical state information, and obtain the true rate in the historical communication quality data, where t is a positive integer; determine the main loss function based on the first predicted azimuth, the first predicted downtilt, the true azimuth, and the true downtilt; determine the rate loss function based on the predicted rate and the true rate; obtain the second predicted azimuth and the second predicted downtilt at time t-1 in the historical prediction results; determine the beam smoothness constraint loss function based on the first predicted azimuth, the second predicted azimuth, the first predicted downtilt, and the second predicted downtilt; determine the effective coverage constraint loss function based on the first predicted azimuth, the preset azimuth maximum value, the preset azimuth minimum value, the first predicted downtilt, the preset downtilt maximum value, and the preset downtilt minimum value; determine the total loss function based on the main loss function, the rate loss function, the beam smoothness constraint loss function, and the effective coverage constraint loss function.

[0121] In the prediction module in the above-mentioned optimal beam determination device, the prediction module is further used to use the prediction model to predict the first state information to obtain second state information at the second moment; and determine the initial beam at the second moment based on the second state information.

[0122] In the first determination module in the above-mentioned optimal beam determination device, the first determination module is further used to determine the similarity between the initial beam and each beam in the codebook of the target device to obtain a similarity set; and determine the beam with the highest similarity in the similarity set as the similar beam of the initial beam.

[0123] In the second determination module in the above-mentioned optimal beam determination device, the second determination module is further used to implement the following steps: Step 1: Obtain the first beam pair corresponding to the current moment, and obtain the adjacent beam pair of the first beam pair; Step 2: Determine the first spectral efficiency corresponding to the first beam pair, and determine the second spectral efficiency corresponding to the adjacent beam pair; Step 3: When the difference between the second spectral efficiency and the first spectral efficiency is greater than or equal to a preset threshold, determine the beam pair change rate based on the first spectral efficiency, the second spectral efficiency and the initial beam pair; Step 4: Determine the intermediate value based on the beam pair change rate, the attenuation coefficient, and the anti-zero coefficient; Step 5: Determine the second beam pair corresponding to the next moment after the current moment based on the intermediate value, the beam pair change rate, and the first beam pair, and determine the second beam pair as the first beam pair; Step 6: Repeat steps 1 to 5 until the difference between the second spectral efficiency and the first spectral efficiency is less than the preset threshold, and stop the iteration, and determine the second beam pair obtained for the last time as the optimal beam.

[0124] It should be noted that Figure 7 The optimal beam determination device shown is used to perform Figure 2 The method for determining the optimal beam is shown in the figure, so the relevant explanations in the above method for determining the optimal beam are also applicable to the device for determining the optimal beam, and will not be repeated here.

[0125] An embodiment of the present application also provides an electronic device, which includes a memory and a processor, wherein the memory is used to store program instructions; the processor is connected to the memory and is used to execute program instructions to implement the following functions: obtaining first state information of a target device at a first moment, wherein the target device is a base station or a drone; using a prediction model to predict the first state information to obtain an initial beam at a second moment, wherein the second moment is the next moment of the first moment; determining a similar beam of the initial beam from a code book of the target device, and determining an initial beam pair based on the initial beam and the similar beam; determining the optimal beam of the target device in each radio frequency link based on the initial beam pair, wherein the optimal beam is the beam configuration with the best communication performance.

[0126] It should be noted that the above electronic equipment is used to perform Figure 2 The method for determining the optimal beam is shown in the figure, so the relevant explanations in the above method for determining the optimal beam are also applicable to the electronic device and will not be repeated here.

[0127] An embodiment of the present application also provides a non-volatile storage medium, which includes a stored computer program, wherein the device where the non-volatile storage medium is located executes the following method for determining the optimal beam by running the computer program: obtaining first state information of a target device at a first moment, wherein the target device is a base station or a drone; using a prediction model to predict the first state information to obtain an initial beam at a second moment, wherein the second moment is the next moment of the first moment; determining a similar beam of the initial beam from a code book of the target device, and determining an initial beam pair based on the initial beam and the similar beam; determining the optimal beam of the target device in each radio frequency link based on the initial beam pair, wherein the optimal beam is the beam configuration with the best communication performance.

[0128] It should be noted that the above non-volatile storage medium is used to execute Figure 2 The method for determining the optimal beam is shown in the figure, so the relevant explanations in the above method for determining the optimal beam are also applicable to the non-volatile storage medium and will not be repeated here.

[0129] An embodiment of the present application further provides a computer program product, comprising computer instructions, which, when executed by a processor, implement the steps of the method for determining the optimal beam in each embodiment of the present application.

[0130] An embodiment of the present application further provides a computer program, which, when executed by a processor, implements the steps of the method for determining the optimal beam in each embodiment of the present application.

[0131] The serial numbers of the above embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0132] In the above embodiments of the present application, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0133] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only exemplary. For example, the division of the units can be a logical function division. In actual implementation, there may be other division methods, such as 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 mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0134] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0135] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0136] If the 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 the present application, or the part that contributes to the existing technology, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for enabling a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk, etc. Various media that can store program code.

[0137] The above is only a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for determining an optimal beam, characterized in that: include: Acquire first state information of a target device at a first moment, wherein the target device is a base station or a drone; Using a prediction model to predict the first state information, to obtain an initial beam at a second moment, wherein the second moment is a moment subsequent to the first moment; Determining a similar beam to the initial beam from a codebook of the target device, and determining an initial beam pair based on the initial beam and the similar beam; An optimal beam of the target device in each radio frequency link is determined based on the initial beam pair, wherein the optimal beam is a beam configuration with the best communication performance.

2. The method according to claim 1, characterized in that The prediction model is trained in the following way: Acquire a training set for training the prediction model, wherein the training set includes: first historical state information of the base station, second historical state information of the drone, and historical communication quality data between the base station and the drone; Using the initial prediction model to predict the training set to obtain historical prediction results; Determining a total loss function based on the historical prediction result, the first historical state information, and the historical communication quality data; The initial prediction model is iteratively trained according to the total loss function to obtain the prediction model.

3. The method according to claim 2, characterized in that The second historical state information includes a first longitude, a first latitude, a first altitude of the UAV, a pitch angle, a roll angle, a yaw angle of the UAV, and a relative azimuth angle between the UAV and the base station, wherein the relative azimuth angle is determined by: Obtaining a second longitude, a second latitude, and a second altitude corresponding to the base station, as well as an equatorial radius and an ellipsoidal flattening; Determine a first value based on the first latitude, the equatorial radius, and the ellipsoidal flattening, and determine a second value based on the second latitude, the equatorial radius, and the ellipsoidal flattening; Determine, based on the first value, the first longitude, the first latitude, and the first altitude, a first geocentric, earth-fixed coordinate corresponding to the drone; and determine, based on the second value, the second longitude, the second latitude, and the second altitude, a second geocentric, earth-fixed coordinate corresponding to the base station, wherein the first geocentric, earth-fixed coordinate includes coordinates of the drone in the X-axis, Y-axis, and Z-axis directions, and the second geocentric, earth-fixed coordinate includes coordinates of the base station in the X-axis, Y-axis, and Z-axis directions; Determining a coordinate difference based on the first Earth-centered Earth-fixed coordinate and the second Earth-centered Earth-fixed coordinate, wherein the coordinate difference includes a coordinate difference between the UAV and the base station in the X-axis, Y-axis, and Z-axis directions; Determining the transformed coordinates of the UAV based on the coordinate difference and a base station rotation matrix, wherein the base station rotation matrix is ​​determined by the second longitude and the second latitude; The relative azimuth angle is determined based on the transformed coordinates and the coordinate origin.

4. The method according to claim 2, characterized in that Determining a total loss function based on the historical prediction result, the first historical state information, and the historical communication quality data includes: Obtaining a first predicted azimuth, a first predicted downtilt, and a first predicted rate at time t in the historical prediction result, obtaining a true azimuth and a true downtilt in the first historical state information, and obtaining a true rate in the historical communication quality data, where t is a positive integer; determining a main loss function based on the first predicted azimuth, the first predicted downtilt, the true azimuth, and the true downtilt; Determining a rate loss function based on the predicted rate and the actual rate; Obtaining a second predicted azimuth angle and a second predicted downtilt angle at time t-1 from the historical prediction result; Determining a beam smoothness constraint loss function according to the first predicted azimuth angle, the second predicted azimuth angle, the first predicted downtilt angle, and the second predicted downtilt angle; Determining an effective coverage constraint loss function based on the first predicted azimuth, the preset maximum azimuth, the preset minimum azimuth, the first predicted downtilt, the preset maximum downtilt, and the preset minimum downtilt; The total loss function is determined according to the main loss function, the rate loss function, the beam smoothness constraint loss function and the effective coverage constraint loss function.

5. The method according to claim 1, characterized in that Predicting the first state information using a prediction model to obtain an initial beam at a second moment includes: Using the prediction model to predict the first state information, to obtain second state information at the second moment; An initial beam at the second moment is determined according to the second state information.

6. The method according to claim 1, characterized in that Determining a similar beam to the initial beam from a codebook of the target device includes: Determine a similarity between the initial beam and each beam in a codebook of the target device to obtain a similarity set; The beam with the highest similarity in the similarity set is determined as a similar beam to the initial beam.

7. The method according to claim 1, characterized in that Determining an optimal beam of the target device in each radio frequency link according to the initial beam pair includes: Step 1: Obtain a first beam pair corresponding to the current moment, and obtain adjacent beam pairs of the first beam pair; Step 2: Determine a first spectrum efficiency corresponding to the first beam pair, and determine a second spectrum efficiency corresponding to the adjacent beam pair; Step 3: When the difference between the second spectrum efficiency and the first spectrum efficiency is greater than or equal to a preset threshold, determine a beam pair change rate based on the first spectrum efficiency, the second spectrum efficiency, and the initial beam pair; Step 4: Determine an intermediate value based on the beam pair change rate, attenuation coefficient, and anti-zero coefficient; Step 5: Determine a second beam pair corresponding to a moment next to the current moment based on the intermediate value, the beam pair change rate, and the first beam pair, and determine the second beam pair as the first beam pair; Step 6: Repeat steps 1 to 5 until the difference between the second spectrum efficiency and the first spectrum efficiency is less than the preset threshold, then stop the iteration, and determine the second beam pair obtained last time as the optimal beam.

8. A device for determining an optimal beam, characterized in that: include: an acquisition module, configured to acquire first state information of a target device at a first moment, wherein the target device is a base station or a drone; a prediction module, configured to predict the first state information using a prediction model to obtain an initial beam at a second moment, wherein the second moment is a moment subsequent to the first moment; a first determining module, configured to determine a similar beam to the initial beam from a codebook of the target device, and determine an initial beam pair based on the initial beam and the similar beam; The second determination module is configured to determine an optimal beam of the target device in each radio frequency link based on the initial beam pair, wherein the optimal beam is a beam configuration with the best communication performance.

9. An electronic device, characterized in that: include: a memory for storing program instructions; A processor, connected to the memory, is configured to execute program instructions that implement the following functions: obtaining first state information of a target device at a first moment, wherein the target device is a base station or a drone; using a prediction model to predict the first state information to obtain an initial beam at a second moment, wherein the second moment is a moment next to the first moment; determining a similar beam of the initial beam from a codebook of the target device, and determining an initial beam pair based on the initial beam and the similar beam; and determining an optimal beam of the target device in each radio frequency link based on the initial beam pair, wherein the optimal beam is a beam configuration with the best communication performance.

10. A non-volatile storage medium, characterized in that: The non-volatile storage medium includes a stored computer program, wherein the device where the non-volatile storage medium is located executes the method for determining the optimal beam according to any one of claims 1 to 7 by running the computer program.

11. A computer program product comprising computer instructions, characterized in that When the computer instructions are executed by a processor, the method for determining the optimal beam according to any one of claims 1 to 7 is implemented.

Images