Directional antenna prediction method and system based on base station position and route

Abstract

The invention discloses a directional antenna prediction method and system based on a base station position and an air route, and relates to the technical field of wireless communication network optimization, and the method comprises the steps: collecting base station information, building and updating a base station database, obtaining the position of an unmanned plane in real time, and matching the position of the unmanned plane with the position of the base station. The position of a real base station is approached through iterative calculation by using MDT data or network measurement data; and predicting a communication demand and RSRP, and obtaining an optimal antenna direction according to a directional antenna prediction result. According to the directional antenna prediction method based on the base station position and the route, the dynamic adaptive capacity of a system to the flight path of the unmanned aerial vehicle is improved by obtaining the position information and the route track of the unmanned aerial vehicle in real time, the quality and stability of data transmission are improved through a base station approximation algorithm, and the directional antenna prediction accuracy is improved by adjusting the pointing direction of the holder type antenna in real time. The reliability and efficiency of task execution of the unmanned aerial vehicle are improved, and better effects are achieved in the aspects of accuracy, stability and reliability.

Description

Technical Field

[0001] The present invention relates to the technical field of wireless communication network optimization, and in particular to a directional antenna prediction method based on base station positions and flight paths. Background Art

[0002] With the rapid development of 5G communication technology, drones have been widely used in various application scenarios, such as inspection, logistics distribution, emergency rescue, etc. 5G networks, with their high bandwidth and low latency, can support drones to transmit high-definition images or large amounts of data in real time in complex environments. Especially in drone flight missions, using 5G networks for data transmission can significantly improve mission efficiency and data transmission quality. However, due to the complexity of drone flight altitudes and areas, traditional directional antenna technology faces challenges and is difficult to adapt to the signal coverage requirements of rapidly moving drones. Therefore, how to maintain stable communication between drones and base stations in complex airspace environments has become a major research hotspot in the current wireless communication field.

[0003] Existing UAV communication systems usually rely on omnidirectional antennas or traditional directional antennas for data transmission. However, although omnidirectional antennas can cover a large range, they often cannot meet the needs of high-bandwidth data transmission in terms of signal strength and stability. Especially in the 5G network environment, when the UAV enters a weak signal area or moves quickly, the signal coverage capability of the omnidirectional antenna is obviously insufficient, resulting in communication interruption or data transmission delay. Although the traditional directional antenna can enhance the signal by adjusting the direction of the antenna, this method relies on real-time manual adjustment or pre-set fixed direction, which is difficult to cope with the rapid changes of UAVs in different flight states. In addition, since the deployment of 5G base stations is limited by geographical conditions and costs, signal coverage cannot guarantee full coverage of all airspaces, further exacerbating the limitations of existing technologies in practical applications. Therefore, existing technologies cannot achieve the best signal coverage range for UAVs during flight, and cannot meet the high stability requirements of real-time transmission of large amounts of data. This technical defect has become one of the key issues restricting the widespread application of UAVs, and also provides an urgent need for the research and development of new directional antenna prediction systems. Summary of the invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by the present invention is: the existing directional antenna prediction method has the problems of unstable signal coverage and large data transmission delay, as well as how to optimize the real-time high-quality data transmission during the flight of the UAV.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: a directional antenna prediction method based on base station location and route, including collecting base station information, establishing and updating a base station database, obtaining the position of a drone in real time, and matching the drone with the base station position; using MDT data or network measurement data to approximate the actual base station position through iterative calculation; predicting communication demand and RSRP, and obtaining the optimal antenna pointing according to the directional antenna prediction result.

[0007] As a preferred solution of the directional antenna prediction method based on base station location and route described in the present invention, the establishment and updating of the base station database includes collecting the location information, signal coverage range, and signal strength data of the base station through on-site measurement and network monitoring, and storing the collected data in the cloud base station database. As the base station construction and network environment change, the data in the base station database is regularly updated.

[0008] As a preferred solution of the directional antenna prediction method based on base station location and route described in the present invention, the matching of the UAV with the base station location includes obtaining the location information of the UAV in real time through the global positioning system of the UAV, determining the route trajectory, take-off point, landing point, and passing points of the UAV according to mission requirements, collecting the location information and route trajectory of the UAV in real time, and matching them with the base station location and signal coverage range data in the base station database to determine the communication needs of the UAV at each point on the flight path.

[0009] As a preferred solution of the directional antenna prediction method based on base station position and route described in the present invention, wherein: the approximation of the actual base station position by iterative calculation includes selecting an initial point close to the actual base station position as the iteration starting point according to the position information of the drone, estimating the initial point based on the positioning method of the reference signal received power, and defining the error function based on the difference between the predicted RSRP value estimated according to the current base station position and the actual measured value, and based on the relationship between the modeled RSRP and the base station position as:

[0010]

[0011] Among them, f(p i ,x) is the prediction function, which estimates the RSRP value based on the position of the drone and the base station, N is the number of drones, and p i is the position of different drones, x is the estimated position of the base station, r i is the RSRP value measured at different positions. In the iterative process, the base station position estimation is updated by the gradient descent method. The iterative formula of the gradient descent method is expressed as:

[0012]

[0013] Among them, αk is the learning rate, is the error function at x k The gradient at , derives the weighted least squares formula, jointly estimates the base station position and the reference path loss, and based on the position of each drone and the received signal strength, the path loss model is expressed as:

[0014]

[0015] Among them, d i is the distance from the drone to the base station, n is the path loss index, ∈ i is the random error term, S i is the signal strength, S0 is the signal strength at the reference distance d0, b is the base station location, and the path loss model is transformed into minimizing the weighted error sum of squares, expressed as:

[0016]

[0017] Among them, w i is the weight of each observation, x i is the x-dimensional position of different drones, y i is the y-dimension position of different drones.

[0018] As a preferred solution of the directional antenna prediction method based on base station position and route described in the present invention, wherein: the approximation of the true base station position by iterative calculation also includes stopping the iteration when the iterative calculation reaches the preset accuracy requirement, taking the average of the estimated base station positions of all cells as the final base station position, solving the weighted least squares problem for each cell by bisection, for each cell, setting the search range of the base station position according to preliminary information, and setting the search range or initial guess value for the reference path loss, iteratively optimizing each cell, selecting one from all cells for the current iteration, within the search range of the base station position, selecting a dimension for bisection, dividing the search range into two parts, and calculating the weighted least squares error of the center point of each part, selecting the interval with smaller error as the new search range according to the weighted least squares error, repeating the iteration until the error is less than the threshold or the maximum number of iterations is reached, saving the optimized base station position estimate of each cell, averaging the optimized base station positions of all cells, and obtaining the final base station position estimate expressed as:

[0019]

[0020] Among them, N1 is the number of cells, and is the optimized base station location estimate for the i-th cell.

[0021] As a preferred solution of the directional antenna prediction method based on base station position and route described in the present invention, wherein: the predicted communication demand and RSRP include predicting the communication demand according to the real-time position, speed and direction of the UAV, combined with the base station position information, predicting the bandwidth demand of the UAV at different time points according to the mission type of the UAV, predicting the signal strength and quality of the UAV at different time points through a signal propagation model or a prediction algorithm based on historical data, combined with the base station position and the UAV position, judging whether the base station meets the communication demand of the UAV in combination with the current load of the base station and the future load change, based on the principle of invariant spatial path loss, predicting the RSRP after the azimuth adjustment through the MDT data before the antenna azimuth adjustment, for each measurement point, using the spatial path loss model to calculate the path loss from the base station to the measurement point, the path loss value remains unchanged before and after the antenna azimuth adjustment, calculating the directional gain of the adjusted antenna in the direction of the measurement point according to the antenna parameters and the azimuth adjustment amount, and predicting the adjusted RSRP as:

[0022] RSRP = α j +β t -S i

[0023] Among them, α j is the base station transmission power, β t The RSRP before azimuth adjustment is obtained by fitting the BP neural network for the locations without MDT data. The number of neural network layers, the number of neurons in each layer, and the activation function are determined according to the complexity of the problem and the amount of data. The weights and biases of the network are randomly initialized. The data of the locations with MDT data are divided into training set, validation set, and test set. The training parameters are set, and the training set data is input into the neural network. The output value is calculated by forward propagation and compared with the true value to calculate the loss function. The weights and biases of the network are updated by the back propagation algorithm to minimize the loss function. This is repeated until the preset number of iterations is reached or the performance on the validation set is no longer improved.

[0024] As a preferred solution of the directional antenna prediction method based on base station location and route described in the present invention, wherein: obtaining the best antenna pointing includes predicting based on RSRP, SINR, and coverage area data, selecting one or more best antenna pointings, and if there is only one optimization target, selecting the best pointing that meets the target; if there are multiple optimization targets, using multi-objective optimization technology, weighted sum method, and Pareto optimal solution to balance the targets, selecting the best pointing according to business rules or network policies, and adjusting the pointing of the gimbal antenna in real time through the control system to place the drone within the best signal coverage range.

[0025] Another object of the present invention is to provide a directional antenna prediction system based on base station location and flight path, which can approximate the actual base station location through an iterative calculation module, thereby solving the problem of limited accuracy of current measurement methods.

[0026] As a preferred solution of the directional antenna prediction system based on base station location and route described in the present invention, it includes: a collection and matching module, an iterative calculation module, and a prediction and pointing module; the collection and matching module collects base station information, establishes and updates a base station database, obtains the drone location in real time, and matches the drone with the base station location; the iterative calculation module is used to use MDT data or network measurement data to approximate the actual base station location through iterative calculation; the prediction and pointing module is used to predict communication demand and RSRP, and obtain the optimal antenna pointing according to the directional antenna prediction result.

[0027] A computer device includes a memory and a processor, wherein the memory stores a computer program, and wherein the processor executes the computer program to implement a step of a directional antenna prediction method based on base station location and route.

[0028] 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 a directional antenna prediction method based on base station location and route are implemented.

[0029] Beneficial effects of the present invention: The directional antenna prediction method based on base station position and route provided by the present invention ensures the timeliness and accuracy of data by integrating and analyzing basic data, improves the system's dynamic adaptability to the flight path of the drone by acquiring the drone's position information and route trajectory in real time, improves the quality and stability of data transmission through the base station approximation algorithm and the position-based tracking algorithm, and improves the reliability and efficiency of the drone's mission execution by adjusting the pointing direction of the pan-tilt antenna in real time. The present invention achieves better results in terms of accuracy, stability and reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0031] Figure 1 An overall flow chart of a directional antenna prediction method based on base station location and flight path provided for the first embodiment of the present invention.

[0032] Figure 2An overall module diagram of a directional antenna prediction system based on base station location and flight path provided for the third embodiment of the present invention. DETAILED DESCRIPTION

[0033] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0034] Example 1

[0035] Reference Figure 1 , as an embodiment of the present invention, provides a directional antenna prediction method based on base station location and route, comprising:

[0036] S1: Collect base station information, establish and update the base station database, obtain the drone location in real time, and match the drone with the base station location.

[0037] Furthermore, establishing and updating the base station database includes through field measurements and network monitoring.

[0038] It should be noted that the location information, signal coverage, and signal strength data of base stations are collected and stored in the cloud base station database. As the base station construction and network environment change, the data in the base station database is updated regularly.

[0039] Furthermore, matching the drone to the base station location includes using the drone's global positioning system.

[0040] It should be noted that the location information of the UAV is obtained in real time, and the UAV's route trajectory, take-off point, landing point, and transit points are determined according to mission requirements. The location information and route trajectory of the UAV are collected in real time and matched with the base station location and signal coverage data in the base station database to determine the communication needs of the UAV at each point on the flight path.

[0041] S2: Use MDT data or network measurement data to approximate the actual base station location through iterative calculation.

[0042] Furthermore, the actual base station location is approximated through iterative calculation including the location information of the UAV.

[0043] It should be noted that the initial point close to the actual base station position is selected as the iteration starting point, and the initial point is estimated based on the positioning method of the reference signal received power. The difference between the predicted RSRP value estimated according to the current base station position and the actual measured value, and based on the relationship between the modeled RSRP and the base station position, the error function is defined as:

[0044]

[0045] Among them, f(p i ,x) is the prediction function, which estimates the RSRP value based on the position of the drone and the base station, N is the number of drones, and p i is the position of different drones, x is the estimated position of the base station, r i is the RSRP value measured at different positions. In the iterative process, the base station position estimation is updated by the gradient descent method. The iterative formula of the gradient descent method is expressed as:

[0046]

[0047] Among them, α k is the learning rate, is the error function at x k The gradient at , derives the weighted least squares formula, jointly estimates the base station position and the reference path loss, and based on the position of each drone and the received signal strength, the path loss model is expressed as:

[0048]

[0049]

[0050] Among them, d i is the distance from the drone to the base station, n is the path loss index, ∈ i is the random error term, S i is the signal strength, S0 is the signal strength at the reference distance d0, b is the base station location, and the path loss model is transformed into minimizing the weighted error sum of squares, expressed as:

[0051]

[0052] Among them, w i is the weight of each observation, x i is the x-dimensional position of different drones, y i is the y-dimension position of different drones.

[0053] Furthermore, approximating the actual base station position through iterative calculation also includes when the iterative calculation reaches a preset accuracy requirement.

[0054] It should be noted that the iteration is stopped, and the mean of the estimated base station positions of all cells is taken as the final base station position. The weighted least squares problem is solved for each cell by bisection. For each cell, the search range of the base station position is set according to the preliminary information, and the search range or initial guess value is set for the reference path loss. Each cell is iteratively optimized, and one cell is selected from all cells for the current iteration. Within the search range of the base station position, a dimension is selected for bisection, the search range is divided into two parts, and the weighted least squares error of the center point of each part is calculated. According to the weighted least squares error, the interval with smaller error is selected as the new search range, and the iteration is repeated until the error is less than the threshold or the maximum number of iterations is reached. The optimized base station position estimate of each cell is saved, and the optimized base station positions of all cells are averaged to obtain the final base station position estimate expressed as:

[0055]

[0056] Among them, N1 is the number of cells, and is the optimized base station location estimate for the i-th cell.

[0057] S3: Predict communication demand and RSRP, and obtain the optimal antenna direction based on the directional antenna prediction results.

[0058] Furthermore, the predicted communication demand and RSRP include the real-time position, speed and direction of the UAV.

[0059] It should be noted that the communication demand is predicted in combination with the base station location information. According to the mission type of the UAV, the bandwidth demand of the UAV at different time points is predicted. The signal strength and quality of the UAV at different time points are predicted by combining the base station location and the UAV location through the signal propagation model or the prediction algorithm based on historical data. In combination with the current load of the base station and the future load change, it is judged whether the base station meets the communication demand of the UAV. Based on the principle of invariance of spatial path loss, the RSRP after the azimuth adjustment is predicted through the MDT data before the antenna azimuth adjustment. For each measurement point, the spatial path loss model is used to calculate the path loss from the base station to the measurement point. The path loss value remains unchanged before and after the antenna azimuth adjustment. According to the antenna parameters and the azimuth adjustment amount, the directional gain of the adjusted antenna in the direction of the measurement point is calculated. The predicted adjusted RSRP is expressed as:

[0060] RSRP = α j +β t -S i

[0061] Among them, α j is the base station transmission power, β tThe RSRP before azimuth adjustment is obtained by fitting the BP neural network for the locations without MDT data. The number of neural network layers, the number of neurons in each layer, and the activation function are determined according to the complexity of the problem and the amount of data. The weights and biases of the network are randomly initialized. The data of the locations with MDT data are divided into training set, validation set, and test set. The training parameters are set, and the training set data is input into the neural network. The output value is calculated by forward propagation and compared with the true value to calculate the loss function. The weights and biases of the network are updated by the back propagation algorithm to minimize the loss function. This is repeated until the preset number of iterations is reached or the performance on the validation set is no longer improved.

[0062] It should also be noted that a multi-layer perceptron (MLP) structure can usually be used, and a suitable activation function (such as ReLU, Sigmoid or Tanh) can be selected to improve the nonlinear mapping ability of the network.

[0063] Furthermore, obtaining the best antenna pointing includes prediction based on RSRP, SINR, and coverage area data.

[0064] It should be noted that one or more optimal antenna directions are selected. If there is only one optimization target, the best direction that meets the target is selected. If there are multiple optimization targets, the multi-objective optimization technology, weighted sum method, and Pareto optimal solution are used to balance the targets. The best direction is selected according to business rules or network strategies. The control system is used to adjust the direction of the gimbal antenna in real time to ensure that the drone is within the optimal signal coverage range.

[0065] Example 2

[0066] An embodiment of the present invention provides a directional antenna prediction method based on base station location and flight path. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0067] In order to verify the performance advantages of the directional antenna prediction system based on base station location and route in UAV communication, rigorous experimental tests were carried out. The experiment selected a complex airspace covering high-rise buildings, mountains and open plains to simulate the flight environment of UAVs under various terrain conditions. The UAV used in the test has the ability to transmit high-definition images and point cloud data in real time. The flight path is from starting point A to end point B, passing through multiple signal weak areas. In the experimental preparation stage, multiple 5G base stations were first deployed in the test area. The locations of these base stations covered the ground and high-rise buildings. The base station database stored key data such as the location information of the base station, signal coverage range and signal strength. The flight path of the UAV was pre-planned to ensure signal coverage and transmission quality under different terrain and altitude conditions. At the beginning of the experiment, the system used the preset base station database and the UAV's The positioning system obtains the location information of the drone in real time and matches it with the base station location data. Through the approximation algorithm, the system uses the minimized drive test technology (MDT) data to iteratively calculate the actual base station location to improve the positioning accuracy. One of the core algorithms in the experiment is the base station approximation algorithm, which can gradually and accurately determine the base station location and ensure the accuracy of the antenna pointing. During the flight of the drone, the system uses a location-based tracking algorithm to predict the communication needs of the drone in flight in real time and dynamically adjust the pointing direction of the pan-tilt antenna. To verify the effectiveness of the present invention, two groups of comparative tests were conducted in the experiment, one group was conducted with all functions of the system enabled, and the other group was conducted with some functions disabled to simulate the application scenarios of traditional technologies. The key data such as signal strength, data transmission rate, signal coverage, and number of communication interruptions were continuously recorded throughout the experiment to analyze the performance of the system.

[0068] As shown in Table 1 and Table 2, when all functions are enabled, the system shows higher signal coverage and data transmission rate. When all functions are enabled, the average strength of the base station signal strength is maintained at around -75dBm, while when some functions are disabled, the average signal strength drops to below -80dBm, which directly affects the data transmission rate. When all functions are enabled, the average transmission rate is about 128Mbps, while after disabling some functions, the rate drops to below 110Mbps. In addition, the system with all functions enabled also shows significant advantages in antenna pointing accuracy and communication interruption times. The best antenna pointing deviation is usually within 1 degree when all functions are enabled, while it is within 1 degree when all functions are disabled. When some functions are enabled, the deviation expands to more than 3 degrees, resulting in a significant increase in the number of communication interruptions. The number of communication interruptions of the UAV at positions A and B increased by 3 times respectively after disabling some functions, showing the defect that traditional technology cannot maintain a stable connection in a dynamic environment. When the present invention enables all functions, it significantly improves the communication performance of the UAV through precise base station positioning, real-time antenna pointing adjustment and efficient communication demand prediction, and shows significant innovation and superiority in signal coverage, data transmission rate and antenna pointing accuracy. In comparison, the performance of traditional technology in a dynamic environment is obviously inferior to that of the present invention, and there is a greater risk of communication interruption and data transmission instability.

[0069] Table 1 Enable full-function experimental data table

[0070]

[0071]

[0072] Table 2 Experimental data table of disabling some functions

[0073]

[0074]

[0075] Example 3

[0076] Reference Figure 2 , as an embodiment of the present invention, provides a directional antenna prediction system based on base station location and route, including: a collection and matching module, an iterative calculation module, and a prediction and pointing module.

[0077] The collection and matching module collects base station information, establishes and updates the base station database, obtains the drone position in real time, and matches the drone with the base station position; the iterative calculation module is used to use MDT data or network measurement data to approximate the actual base station position through iterative calculation; the prediction and pointing module is used to predict communication demand and RSRP, and obtain the optimal antenna pointing according to the directional antenna prediction results.

[0078] If the function 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 invention, or the part that contributes to the prior art, or the 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, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0079] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0080] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering, or processing in another suitable manner as necessary, and then stored in a computer memory.

[0081] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc. It should be noted that the above embodiments are only used to illustrate the technical solution of the present invention and are not limited. Although the present invention is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention, which should be included in the scope of the claims of the present invention.

[0082] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A directional antenna prediction method based on base station location and flight path, characterized in that: include: Collect base station information, establish and update base station database, obtain drone location in real time, and match drone with base station location; Using MDT data or network measurement data, the actual base station location is approximated through iterative calculation; Predict communication demand and RSRP, and obtain the optimal antenna pointing based on the directional antenna prediction results.

2. The directional antenna prediction method based on base station location and flight path according to claim 1, characterized in that: The establishment and updating of the base station database includes collecting the location information, signal coverage, and signal strength data of the base stations through on-site measurement and network monitoring, storing the collected data in the cloud base station database, and regularly updating the data in the base station database as the base station construction and network environment change.

3. The directional antenna prediction method based on base station location and flight path as claimed in claim 2, characterized in that: The matching of the drone with the base station location includes obtaining the drone's location information in real time through the drone's global positioning system, determining the drone's route trajectory, take-off point, landing point, and waypoints according to mission requirements, collecting the drone's location information and route trajectory in real time, and matching them with the base station location and signal coverage range data in the base station database to determine the drone's communication needs at each point on the flight path.

4. The directional antenna prediction method based on base station location and flight path as claimed in claim 3, characterized in that: The method of approximating the actual base station position by iterative calculation includes selecting an initial point close to the actual base station position as an iteration starting point according to the position information of the drone, estimating the initial point based on the positioning method of the reference signal received power, and defining the error function based on the difference between the predicted RSRP value estimated by the current base station position and the actual measured value and the relationship between the modeled RSRP and the base station position as follows: Among them, f(p i , x) is the prediction function, which estimates the RSRP value based on the position of the drone and the base station, N is the number of drones, p i is the position of different drones, x is the estimated position of the base station, r i is the RSRP value measured at different positions. In the iterative process, the base station position estimation is updated by the gradient descent method. The iterative formula of the gradient descent method is expressed as: Among them, α k is the learning rate, is the error function at x k The gradient at , derives the weighted least squares formula, jointly estimates the base station position and the reference path loss, and based on the position of each drone and the received signal strength, the path loss model is expressed as: Among them, d i is the distance from the drone to the base station, n is the path loss index, ∈ i is the random error term, S i is the signal strength, S0 is the signal strength at the reference distance d0, b is the base station location, and the path loss model is transformed into minimizing the weighted error sum of squares, expressed as: Among them, w i is the weight of each observation, x i is the x-dimensional position of different drones, y i is the y-dimension position of different drones.

5. The directional antenna prediction method based on base station location and flight path as claimed in claim 4, characterized in that: The method of approximating the true base station position by iterative calculation also includes stopping the iteration when the iterative calculation reaches a preset accuracy requirement, taking the average of the estimated base station positions of all cells as the final base station position, solving the weighted least squares problem for each cell by bisection, setting the search range of the base station position for each cell according to preliminary information, and setting the search range or initial guess value for the reference path loss, iteratively optimizing each cell, selecting one from all cells for the current iteration, selecting a dimension within the search range of the base station position for bisection, dividing the search range into two parts, and calculating the weighted least squares error of the center point of each part, selecting an interval with a smaller error as a new search range according to the weighted least squares error, repeating the iteration until the error is less than a threshold or the maximum number of iterations is reached, saving the optimized base station position estimate for each cell, averaging the optimized base station positions of all cells, and obtaining the final base station position estimate expressed as: Among them, N1 is the number of cells, and is the optimized base station location estimate for the i-th cell.

6. The directional antenna prediction method based on base station location and flight path according to claim 5, characterized in that: The communication demand and RSRP prediction includes predicting the communication demand according to the real-time position, speed and direction of the UAV in combination with the base station position information, predicting the bandwidth demand of the UAV at different time points according to the mission type of the UAV, predicting the signal strength and quality of the UAV at different time points in combination with the base station position and the UAV position through a signal propagation model or a prediction algorithm based on historical data, judging whether the base station meets the communication demand of the UAV in combination with the current load of the base station and the future load change, predicting the RSRP after the azimuth adjustment based on the principle of invariance of spatial path loss through the MDT data before the antenna azimuth adjustment, for each measurement point, using the spatial path loss model to calculate the path loss from the base station to the measurement point, the path loss value remains unchanged before and after the antenna azimuth adjustment, calculating the directional gain of the adjusted antenna in the direction of the measurement point according to the antenna parameters and the azimuth adjustment amount, and predicting the adjusted RSRP as follows: RSRP=α j +β t -S i Among them, α j is the base station transmission power, β t The RSRP before azimuth adjustment is obtained by fitting the BP neural network for the locations without MDT data. The number of neural network layers, the number of neurons in each layer, and the activation function are determined according to the complexity of the problem and the amount of data. The weights and biases of the network are randomly initialized. The data of the locations with MDT data are divided into training set, validation set, and test set. The training parameters are set, and the training set data is input into the neural network. The output value is calculated by forward propagation and compared with the true value to calculate the loss function. The weights and biases of the network are updated by the back propagation algorithm to minimize the loss function. This is repeated until the preset number of iterations is reached or the performance on the validation set is no longer improved.

7. The directional antenna prediction method based on base station location and flight path according to claim 6, characterized in that: The method of obtaining the optimal antenna pointing includes making predictions based on RSRP, SINR, and coverage area data, selecting one or more optimal antenna pointings, and if there is only one optimization target, selecting the optimal pointing that meets the target; if there are multiple optimization targets, balancing the targets through multi-target optimization technology, weighted sum method, and Pareto optimal solution, selecting the optimal pointing according to business rules or network policies, and adjusting the pointing of the gimbal antenna in real time through the control system to place the drone within the optimal signal coverage range.

8. A system using the directional antenna prediction method based on base station location and flight path as claimed in any one of claims 1 to 7, characterized in that: It includes collection and matching module, iterative calculation module and prediction and pointing module; The collection and matching module collects base station information, establishes and updates a base station database, obtains the position of the drone in real time, and matches the drone with the base station position; The iterative calculation module is used to use MDT data or network measurement data to approximate the actual base station position through iterative calculation; The prediction and pointing module is used to predict communication demand and RSRP, and obtain the optimal antenna pointing according to the prediction result of the directional antenna.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the directional antenna prediction method based on base station position and route described in any one of claims 1 to 7 are implemented.

10. 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 directional antenna prediction method based on base station position and route described in any one of claims 1 to 7 are implemented.

Images