Signal strength detection method and device, and storage medium

Abstract

The embodiment of the invention discloses a signal strength detection method and device, and a storage medium. The method comprises the following steps: receiving unmanned aerial vehicle pose data of an unmanned aerial vehicle in a to-be-detected area, and terminal position data collected by a ground test terminal without man-machine interaction; according to the unmanned aerial vehicle pose data and the terminal position data, establishing a plurality of position relation characteristics between the unmanned aerial vehicle and the ground test terminal; and inputting the plurality of position relation characteristics into the target signal prediction model to obtain the signal intensity of the to-be-detected area.

Description

Technical Field

[0001] The present application relates to the field of wireless technologies, and in particular, to a signal strength detection method, an apparatus, and a storage medium. Background Art

[0002] With the gradual expansion of the drone market, drones have great application prospects in multiple scenarios such as emergency rescue, transportation logistics, and agricultural plant protection. Good aerial network coverage planning is an important prerequisite for aerial operations such as drone emergency communication, and it is necessary to predict the coverage of the drone aerial wireless network and adjust the wireless network base station propagation model in a timely manner according to the prediction results.

[0003] In the related art, data is collected by setting up CW continuous waves on the ground in a specified scenario. The CW transmitter and receiver are fixedly set up at the ground end to receive the transmitted signal, and the received transmitted signal is manually exported and then the parameters of the wireless network base station propagation model are calibrated based on the existing ground propagation model (i.e., a mathematical formula model, such as OKumra-Hata, Uma, etc.), and the network signal is predicted according to the calibrated model.

[0004] Since the air-to-ground wireless channel is different from the wireless propagation channel of a ground stationary base station, when a drone is used for emergency rescue, the drone carrying the aerial base station continuously hovers and moves at high speed, and the air-to-ground wireless channel of a fixed-wing drone in an emergency state is a time-varying non-linear system, resulting in inaccurate model calibration and reducing the accuracy of network signal prediction. Summary of the Invention

[0005] To solve the above technical problems, embodiments of the present application are expected to provide a signal strength detection method, an apparatus, and a storage medium, which can improve the accuracy of network signal prediction.

[0006] The technical solution of the present application is implemented as follows:

[0007] Embodiments of the present application provide a signal strength detection method, and the signal strength detection method includes:

[0008] Receiving the drone pose data of the drone in the area to be detected, and the terminal position data collected by a ground test terminal interacting with the drone;

[0009] According to the drone pose data and the terminal position data, establishing a plurality of position relationship features between the drone and the ground test terminal;

[0010] Inputting the plurality of position relationship features into a target signal prediction model to obtain the signal strength of the area to be detected.

[0011] An embodiment of the present application provides a signal strength detection device, and the device includes:

[0012] A receiving unit, configured to receive the drone pose data of the drone in the area to be detected, and the terminal position data collected by a ground test terminal that interacts with the drone;

[0013] A establishing unit, configured to establish a plurality of position relationship features between the drone and the ground test terminal according to the drone pose data and the terminal position data;

[0014] An input unit, configured to input the plurality of position relationship features into a target signal prediction model to obtain the signal strength of the area to be detected.

[0015] An embodiment of the present application provides a signal strength detection system, and the system includes a drone, a ground test terminal, and a signal strength detection device:

[0016] The drone is configured to collect drone pose data;

[0017] The ground test terminal is configured to collect terminal position data;

[0018] The signal strength detection device is configured to establish a plurality of sample position relationship features between the drone and the ground test terminal according to the drone pose data and the terminal position data; train an initial signal prediction model by using the plurality of sample position relationship features to obtain a target signal prediction model; and predict the signal strength of the area to be detected by using the target signal prediction model.

[0019] An embodiment of the present application provides a signal strength detection device, and the device includes:

[0020] A memory, a processor, and a communication bus. The memory communicates with the processor through the communication bus. The memory stores a program for signal strength detection executable by the processor. When the program for signal strength detection is executed, the above-mentioned signal strength detection method is executed by the processor.

[0021] An embodiment of the present application provides a storage medium, on which a computer program is stored and applied to a signal strength detection device. The computer program, when executed by a processor, implements the above-mentioned signal strength detection method.

[0022] An embodiment of the present application further provides a computer program product, including a computer program, which can be executed by a processor of a signal strength detection device to complete the steps of any of the foregoing methods.

[0023] The embodiments of the present application provide a signal strength detection method, device, and storage medium. The signal strength detection method includes: receiving the UAV pose data of the UAV in the area to be detected, and the terminal position data collected by the ground test terminal interacting with the UAV; establishing multiple position relationship features between the UAV and the ground test terminal according to the UAV pose data and the terminal position data; inputting the multiple position relationship features into the target signal prediction model to obtain the signal strength of the area to be detected. By adopting the above method implementation scheme, the signal strength detection device establishes multiple position relationship features between the UAV and the ground test terminal according to the UAV pose data and the terminal position data. The multiple position relationship features include relationship features of multiple types such as distance, angle, and channel; it enables the signal parameters of the air-ground wireless channel in the emergency state of the fixed-wing UAV to be accurately described by using relationship features of multiple types when the UAV carrying the aerial base station continuously circles and moves at high speed, that is, the accuracy of network signal prediction is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is a flowchart of a signal strength detection method provided by an embodiment of the present application;

[0025] Figure 2 It is an exemplary air-space propagation model system architecture diagram provided by an embodiment of the present application;

[0026] Figure 3 It is an exemplary connection diagram of each module of the airborne part provided by an embodiment of the present application;

[0027] Figure 4 It is an exemplary distance calculation principle diagram provided by an embodiment of the present application;

[0028] Figure 5 It is an exemplary angle calculation principle diagram provided by an embodiment of the present application;

[0029] Figure 6 It is a flowchart of an exemplary propagation model calibration method provided by an embodiment of the present application;

[0030] Figure 7 It is a schematic composition structure of a signal strength detection device provided by an embodiment of the present application Figure 1 ;

[0031] Figure 8 It is a schematic composition structure of a signal strength detection device provided by an embodiment of the present application Figure 2 ;

[0032] Figure 9 It is a schematic composition structure diagram of a signal strength detection system provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0033] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0034] The embodiments of the present application provide a signal strength detection method. A signal strength detection method is applied to a signal strength detection device. Figure 1 It is a flowchart of a signal strength detection method provided by the embodiments of the present application. As Figure 1 shown, the signal strength detection method may include:

[0035] S101. Receive the drone pose data of the drone in the area to be detected, and the terminal position data collected by the ground test terminal interacting with the drone.

[0036] The signal strength detection method provided by the embodiments of the present application is applicable to the scenario of predicting the signal strength of the area to be detected according to the drone pose data and the terminal position data by using a target signal prediction model.

[0037] In the embodiments of the present application, the signal strength detection device may be implemented in various forms. For example, the signal strength detection device described in the present application may include devices such as mobile phones, cameras, tablet computers, laptop computers, palmtop computers, personal digital assistants (Personal Digital Assistant, PDA), portable media players (Portable Media Player, PMP), navigation devices, wearable devices, smart bracelets, pedometers, etc., and devices such as digital TVs, desktop computers, servers, etc.

[0038] It should be noted that the signal strength detection device may also be a cloud platform.

[0039] It should be noted that the drone pose data is the pose data of the drone, and the terminal position data is the position data of the ground test terminal. Specifically, the terminal position data includes information such as the reference signal receiving power (Reference Signal Receiving Power, RSRP) measured by the ground test terminal and the position of the ground test terminal; the drone pose data includes the flight attitude information of the drone, the position information of the drone, etc.

[0040] In the embodiments of the present application, the signal strength detection device may receive the drone pose data of the drone in the area to be detected transmitted by the satellite device.

[0041] In an embodiment of the present application, a transmission base station is provided on the drone. The drone can use the transmission base station to transmit the drone pose data to the satellite device, and then the satellite device transmits the drone pose data to the signal strength detection device through ground wireless relay. The ground test terminal is connected to the signal strength detection device, and the signal strength detection device can directly obtain the terminal position data in the ground test terminal that interacts with the drone.

[0042] Exemplarily, the positional relationship among the drone, the satellite device, the ground wireless relay, the ground test terminal, and the cloud platform is as Figure 2 shown: The satellite device (airborne wireless relay) is located in the space-air part, the drone is located in the airborne part, and the ground wireless relay, the ground test terminal, the core network, and the cloud platform (signal strength detection device) are located in the ground part. Specifically, the space-air part consists of relay devices such as satellite communication, provides communication traffic, and is used to realize the remote beyond-line-of-sight control of the drone and the backhaul of communication data information such as propagation model levels. By using the wireless coverage of satellite communication, the drone can maintain the smoothness of the cellular communication link during the flight at any time and any place, and can meet the requirements for the acquisition of propagation model environments and the air-ground height in various scenarios. The ground part consists of a ground wireless relay station, a core network, a ground test terminal, and a cloud platform. The ground wireless relay station is used to receive the aircraft control information, communication service data, and data information related to the drone attitude transmitted by relay devices such as space-air satellites. The ground wireless relay station accesses the mobile core network and the cloud platform in a wireless or wired manner. The ground test terminal is used to collect network data such as RSRP and SINR for propagation model calibration, and transmits it to the cloud platform for data fusion and model calibration with the drone attitude information data received by the cloud platform. The airborne part mainly includes a drone platform, an airborne relay device, and an airborne base station part. The connection schematic diagram among the modules is as Figure 3 shown: It includes a drone platform, an airborne relay device module, an airborne base station module, an aircraft interaction port, and a Global Positioning System (GPS). The space-air base station is mainly used to transmit the propagation signal source and perform the relay backhaul of control signal data, and consists of a baseband processing unit, a radio frequency unit, and an airborne antenna. The airborne antenna is mainly used to receive the transmitted signal from the base station; the drone platform is mainly used to provide flight tests and carry terminals and test antennas. The airborne relay device is mainly used to transmit and transfer the communication resources of the base station. The aircraft interaction port can be in the form of a serial port, a network port, etc., and is connected to the GPS information through the airborne base station.

[0043] In the embodiments of the present application, the signal strength detection device can transmit data acquisition instructions to the ground test terminal and the drone respectively. The ground test terminal sends terminal position data to the signal strength detection device according to the data acquisition instructions; the drone sends the drone pose data to the data transmission device through air wireless relay and ground wireless relay according to the data acquisition instructions.

[0044] In the embodiments of the present application, the ground wireless relay can send all the pose data of the received drone to the signal strength detection device, or the ground wireless relay can also send the pose data of the drone within a preset time period to the signal strength detection device; specifically, it can be determined according to the actual situation, and the embodiments of the present application do not limit this.

[0045] It should be noted that the preset time period can be the time period configured in the ground wireless relay, or the time period carried in the data acquisition instruction transmitted by the signal strength detection device, or the time period obtained by the ground wireless relay in other ways. Specifically, the way for the ground wireless relay to obtain the preset time period can be determined according to the actual situation, and the embodiments of the present application do not limit this.

[0046] It should be noted that the preset time period can be a data acquisition cycle, that is, the time period between the data acquisition instruction sent by the signal strength detection device and the previous data acquisition instruction sent, or other time periods. Specifically, the preset time period can be determined according to the actual situation, and the embodiments of the present application do not limit this.

[0047] S102. Establish multiple position relationship features between the drone and the ground test terminal according to the drone pose data and the terminal position data.

[0048] In the embodiments of the present application, after the signal strength detection device receives the drone pose data of the drone in the area to be detected and the terminal position data collected by the ground test terminal interacting with the drone, it establishes multiple position relationship features between the drone and the ground test terminal according to the drone pose data and the terminal position data.

[0049] In the embodiments of the present application, the multiple position relationship features include relationship features of multiple types such as distance, angle, and channel.

[0050] It should be noted that the multiple position relationship features established are shown in Table 1:

[0051] Table 1

[0052]

[0053]

[0054] In the embodiment of the present application, the process of the signal strength detection device establishing multiple position relationship features between the unmanned aerial vehicle and the ground test terminal according to the unmanned aerial vehicle pose data and the terminal position data includes: preprocessing the unmanned aerial vehicle pose data to obtain the preprocessed unmanned aerial vehicle pose data; preprocessing the terminal position data to obtain the preprocessed terminal position data; and establishing multiple position relationship features according to the preprocessed unmanned aerial vehicle pose data and the preprocessed terminal position data.

[0055] In the embodiment of the present application, the signal strength detection device preprocesses the unmanned aerial vehicle pose data in the same way as it preprocesses the terminal position data.

[0056] In the embodiment of the present application, the preprocessing method includes removing points with too large or too small signals (such as points where the RSRP exceeds -140 and is less than -40) in the unmanned aerial vehicle pose data or the terminal position data, removing position abnormal points (such as data points without longitude and latitude positions in the GPS record, or data points where the longitude and latitude drift out of the preset area, etc.) in the unmanned aerial vehicle pose data or the terminal position data, deleting duplicate points, etc.

[0057] In the embodiment of the present application, the multiple position relationship features of the signal strength detection device include: the horizontal distance between the unmanned aerial vehicle and the ground test terminal, the vertical distance between the unmanned aerial vehicle and the ground test terminal, the straight-line distance between the unmanned aerial vehicle and the ground test terminal, the first azimuth angle of the ground test terminal relative to the unmanned aerial vehicle, the first elevation angle of the ground test terminal relative to the unmanned aerial vehicle, the second azimuth angle of the ground test terminal relative to the antenna, the second elevation angle of the ground test terminal relative to the antenna, and the multi-dimensional dynamic antenna gain.

[0058] In the embodiment of the present application, the process of the signal strength detection device establishing multiple position relationship features based on the preprocessed UAV pose data and the preprocessed terminal position data includes: obtaining the first longitude and latitude of the UAV and the first altitude of the UAV from the preprocessed UAV pose data; obtaining the second longitude and latitude of the ground test terminal, the second altitude of the ground test terminal, and the roll angle and heading angle of the UAV from the preprocessed terminal position data; determining the horizontal distance between the UAV and the ground test terminal according to the first longitude and latitude, the second longitude and latitude, and the radius of the earth; determining the vertical distance between the UAV and the ground test terminal according to the first altitude and the second altitude; determining the straight-line distance between the UAV and the ground test terminal according to the horizontal distance and the vertical distance; determining the first azimuth angle of the ground test terminal relative to the UAV according to the first longitude and latitude and the second longitude and latitude; determining the first elevation angle of the ground test terminal relative to the UAV according to the horizontal distance and the vertical distance; determining the second azimuth angle of the ground test terminal relative to the antenna according to the first azimuth angle and the roll angle; determining the second elevation angle of the ground test terminal relative to the antenna according to the first elevation angle and the heading angle; determining the multi-dimensional dynamic antenna gain matching the second azimuth angle and the second elevation angle; and determining the horizontal distance, the vertical distance, the straight-line distance, the first azimuth angle, the first elevation angle, the second azimuth angle, the second elevation angle, and the multi-dimensional dynamic antenna gain as multiple position relationship features.

[0059] It should be noted that an antenna is provided on the UAV, which is also called the UAV antenna. The second azimuth angle of the ground test terminal relative to the UAV antenna is determined according to the first azimuth angle and the roll angle.

[0060] It should be noted that the first longitude and latitude of the UAV include the first longitude and the first latitude; the second longitude and latitude of the ground test terminal include the second longitude and the second latitude.

[0061] In the embodiment of the present application, if the horizontal distance between the UAV base station and the ground receiving terminal is d 2D , as Figure 4 shown, point A represents the position of the ground test terminal, point B represents the position of the UAV, and R represents the average radius of the earth. Aj and Bj represent the longitudes of points A and B respectively. Aw and Bw represent the latitudes of points A and B respectively. c represents the angle between the lines connecting the two endpoints of the "arc" corresponding to point C and the center of the earth.

[0062] Derived from the spherical cosine formula, the calculation formula for the angle c is as shown in formula (1):

[0063] c = arccos(cos(90 - Bw)cos(90 - Aw) + sin(90 - Bw)sin(90 - Aw)cos(Bj - Aj)) (1)

[0064] According to formula (2), the horizontal distance d between the UAV and the ground test terminal can be determined 2D (i.e., Figure 4 Dis_2d in Figure 5 ). As shown in 3D (i.e., Figure 5 Dis_3d in

[0065] d 2D = R * c (2)

[0066]

[0067] It should be noted that in formula (3), H UT is the height of the ground test terminal from the ground (i.e., the second height), and H_drone is the height of the UAV from the ground (i.e., the first height).

[0068] The vertical distance between the UAV and the ground test terminal, i.e., the signal propagation height difference H d is as shown in formula (4):

[0069] H d = H_drone - H UT (4)

[0070] Furthermore, formula (3) can be expressed as shown in formula (5):

[0071]

[0072] The first azimuth angle between the base station and the UAV can be obtained from formula (6):

[0073]

[0074] The first elevation angle θ between the UAV and the test terminal relative to the antenna can be determined using formula (7) ue :

[0075] θ ue = arctan(d 2d / H d ) (7)

[0076] In the embodiment of the present application, the second azimuth angle of the ground test terminal relative to the 3D antenna gain (i.e., the multi-dimensional dynamic antenna gain) The second elevation angle θ is affected by the dynamic transformation of the roll angle (roll angle ) and the heading angle (θ uav ) during the flight of the UAV. The second azimuth angle of the terminal relative to the antenna and the second elevation angle θ is the first azimuth angle of the terminal relative to the UAV The first elevation angle θ ue and the roll angle of the UAV The heading angle θ uav The difference and the modulo value of 2π. The specific calculation method is shown in formulas (8)-(9):

[0077]

[0078] θ = (θ ue - θ uav ) mod (2π) (9)

[0079] In the embodiments of the present application, through the angles obtained in formulas (8)-(9) θ, by looking up the simulated values of the 3D antenna of the constructed system (the corresponding relationship between the preset second elevation angle, the preset second azimuth angle and the preset gain), the dynamic 3D antenna gain value at the corresponding position can be obtained, that is, the multi-dimensional dynamic antenna gain G is obtained Tx3D .

[0080] S103. Input multiple position relationship features into the target signal prediction model to obtain the signal strength of the area to be detected.

[0081] In the embodiments of the present application, after the signal strength detection device establishes multiple position relationship features between the UAV and the ground test terminal according to the UAV pose data and the terminal position data, it inputs the multiple position relationship features into the target signal prediction model to obtain the signal strength of the area to be detected.

[0082] In the embodiments of the present application, the process of the signal strength detection device inputting multiple position relationship features into the target signal prediction model to obtain the signal strength of the area to be detected includes: screening at least one target feature from the multiple position relationship features; inputting the at least one target feature into the target signal prediction model to obtain the target path loss of the area to be detected; and determining the signal strength according to the target path loss.

[0083] In the embodiments of the present application, the process of determining the signal strength according to the target path loss includes: obtaining the signal transmission power of the UAV corresponding to the area to be detected, the receiving antenna gain of the ground test terminal corresponding to the area to be detected, and the receiving power when the ground test terminal receives the signal transmitted by the UAV; and determining the signal strength according to the receiving power, the signal transmission power, the receiving antenna gain and the target path loss.

[0084] In the embodiments of the present application, at least one target feature corresponding to the area to be detected processed by the signal strength detection device {such as the target straight-line distance, the target horizontal distance, the target first elevation angle, the target heading angle, the target first azimuth angle}, that is Input the trained XGboost model (i.e., the target signal prediction model), and the target path loss Loss_test can be predicted. Using formula (10), the signal strength in the area to be detected can be further determined, that is, the level P of the network signal. test :

[0085] P test = P Tx_test + G Tx3D_test + G rx_test - Loss_test (10)

[0086] Wherein, {G Tx3D_test , P Tx_test , G rx_test} are all actually known parameters of the test system. G Tx3D_test is the dynamic transmission 3D gain of the antenna in the scene to be tested, and the corresponding gain value can be queried through the constructed 3D gain (i.e., determined using formula 13); P Tx_test is the transmission power of the base station cell in the scene to be tested, which is a value that can be measured by the ground test terminal; G rx_test is the receiving antenna gain in the scene to be tested, which is the device's own parameter configuration value; P test is the power value collected by the ground test terminal in the scene to be tested.

[0087] In the embodiment of the present application, P test is the predicted value of the signal strength in the area to be tested given by the target signal prediction model. The signal strength of the area to be tested can be predicted in combination with the attitude of the drone, and the optimal flight plan can be given to save flight costs.

[0088] In the embodiment of the present application, the process of the signal strength detection device screening at least one target feature from multiple position relationship features includes: determining the first path loss; respectively determining the correlation coefficients between multiple position relationship features and the first path loss to obtain multiple correlation coefficients; sorting the multiple correlation coefficients in descending order of values to obtain a sorted sequence; starting from the first sorting position of the sorted sequence, screening a preset number of correlation coefficients to obtain at least one target correlation coefficient; among the multiple position relationship features, obtaining at least one target feature corresponding to the at least one target correlation coefficient.

[0089] In the embodiment of the present application, the first path loss is the path loss during signal transmission between the drone and the ground test terminal corresponding to multiple position relationship features.

[0090] In the embodiments of the present application, the Pearson correlation coefficient calculation method can be used to determine the correlation coefficients between multiple positional relationship features and the first path loss respectively, obtaining multiple correlation coefficients; the Spearman rank correlation coefficient calculation method can also be used to determine the correlation coefficients between multiple positional relationship features and the first path loss respectively, obtaining multiple correlation coefficients; other correlation coefficient calculation methods can also be used to determine the correlation coefficients between multiple positional relationship features and the first path loss, obtaining multiple correlation coefficients; the specific implementation manner can be determined according to the actual situation, and the embodiments of the present application do not limit this.

[0091] Exemplarily, the correlation coefficient determination method in formula (11) can be used to determine the correlation coefficients between multiple positional relationship features and the first path loss respectively, obtaining multiple correlation coefficients:

[0092]

[0093] It should be noted that X represents the feature vector corresponding to multiple positional relationship features, Y represents the first path loss, cov(X, Y) represents the covariance between each positional relationship feature and the first path loss, μ X represents the mean value of the positional relationship features, μ Y represents the mean value of the first path loss value, σ X represents the standard deviation of the positional relationship features, σ Y represents the standard deviation of the first path loss value.

[0094] In the embodiments of the present application, the multiple correlation coefficients can be sorted in descending order of values to obtain a sorted sequence; they can also be sorted in ascending order of values to obtain a sorted sequence; the specific sorting method can be determined according to the actual situation, and the embodiments of the present application do not limit this.

[0095] In the embodiments of the present application, if the multiple correlation coefficients are sorted in descending order of values to obtain a sorted sequence; then starting from the first sorting position of the sorted sequence, a preset number of correlation coefficients are screened to obtain at least one target correlation coefficient; if the multiple correlation coefficients are sorted in ascending order of values to obtain a sorted sequence; then starting from the last sorting position of the sorted sequence, a preset number of correlation coefficients are screened to obtain at least one target correlation coefficient.

[0096] It should be noted that the preset number can be the number configured in the signal strength detection device, or the number transmitted from other devices to the signal strength detection device, or the number obtained by the signal strength detection device in other ways; the specific way for the signal strength detection device to obtain the preset number can be determined according to the actual situation, and the embodiments of the present application do not limit this.

[0097] Exemplarily, m (preset quantity) feature variables with the largest correlation coefficient related to path loss can be selected in descending order of the correlation values. For example: {linear distance, horizontal distance, first elevation angle, course angle, first azimuth angle} are used as feature vectors (at least one target feature) to construct the propagation model.

[0098] In the embodiment of the present application, the quantity of at least one target feature is less than or equal to the quantity of multiple position relationship features.

[0099] In the embodiment of the present application, the process for the signal strength detection device to determine the first path loss includes: determining the multi-dimensional dynamic antenna gain; obtaining the cell transmission power of the UAV transmitting a signal to the ground test terminal, the receiving antenna gain of the ground test terminal, and the received power of the ground test terminal receiving the signal transmitted by the UAV; and determining the first path loss according to the multi-dimensional dynamic antenna gain, the cell transmission power, the receiving antenna gain, and the received power.

[0100] In the embodiment of the present application, the cell transmission power of the UAV transmitting a signal to the ground test terminal, the receiving antenna gain of the ground test terminal, and the received power of the ground test terminal receiving the signal transmitted by the UAV can be obtained from a database, or the cell transmission power can be obtained from the UAV, and the receiving antenna gain and the received power can be obtained from the ground test terminal; specifically, the manner of obtaining the cell transmission power, the receiving antenna gain, and the received power can be determined according to the actual situation, and the embodiment of the present application does not limit this.

[0101] In the embodiment of the present application, the process of determining the first path loss according to the multi-dimensional dynamic antenna gain, the cell transmission power, the receiving antenna gain, and the received power can be to determine the sum of the multi-dimensional dynamic antenna gain, the cell transmission power, and the receiving antenna gain, and then determine the difference from the received power, so as to obtain the first path loss.

[0102] Exemplarily, as shown in formula (12):

[0103] P L =P Tx +P Tx3D +G rx -P Rx (12)

[0104] It should be noted that P L is the first path loss; P Tx is the cell transmission power; P Tx3D is the multi-dimensional dynamic antenna gain; G rx is the receiving antenna gain; P Rx is the received power.

[0105] In an embodiment of the present application, the process for the signal strength detection device to determine the multi-dimensional dynamic antenna gain includes: obtaining the correspondence between the preset second elevation angle, the preset second azimuth angle, and the preset gain; and determining the multi-dimensional dynamic antenna gain according to the second elevation angle, the second azimuth angle, and the correspondence.

[0106] In an embodiment of the present application, the correspondence between the preset second elevation angle, the preset second azimuth angle, and the preset gain can be configured in the signal strength detection device, or transmitted to the signal strength detection device by other devices, or obtained by the signal strength detection device through other means. The specific method for the signal strength detection device to obtain the correspondence between the preset second elevation angle, the preset second azimuth angle, and the preset gain can be determined according to the actual situation, and the embodiments of the present application do not limit this.

[0107] Exemplarily, the correspondence between the preset second elevation angle, the preset second azimuth angle, and the preset gain is shown in Formula (13):

[0108]

[0109] It should be noted that Φ s is the horizontal angle (preset second azimuth angle) of the ground test terminal under the base station coordinates, θ s is the elevation angle (preset second elevation angle) of the terminal under the base station coordinates. The horizontal gain and vertical gain of the terminal position are H(Φ s ) and V(θ s ), respectively. H(0) is the horizontal gain value of the antenna in the direction of 0 degrees of the horizontal angle, and H(π) refers to the horizontal gain value of the antenna in the direction of 180 degrees of the horizontal angle. Usually, the antenna only has a two-dimensional gain chart (the horizontal angle is 0, the vertical angle is 0 - 360, that is, H(0); and the vertical angle is 0, the horizontal angle is 0 - 360, that is, V(0), and the corresponding gain table), which can be obtained by reading the given two-dimensional gain table. V(θ s ) is the elevation angle of the terminal under the base station coordinates, and (π - θ s ) is the elevation angle value of 180° of the terminal under the base station coordinates. These can all be obtained by reading the original two-dimensional table of the antenna. Gain(Φ s , θ s ) is the three-dimensional antenna gain value corresponding to the interpolation at the angles of Φ s and θ s , that is, the preset gain. The value range of each angle in Formula 13 is 0 - 360.

[0110] In an embodiment of the present application, the process of determining the multi-dimensional dynamic antenna gain according to the second elevation angle, the second azimuth angle, and the correspondence can be to use the second elevation angle and the second azimuth angle as Φ s and θ s, the multi-dimensional dynamic antenna gain determined when other parameter values in Formula 13 remain unchanged.

[0111] In the embodiment of the present application, before the signal strength detection inputs multiple position relationship features into the target signal prediction model to obtain the signal strength of the area to be detected, it will also obtain the sample UAV pose data transmitted by the UAV and the sample terminal position data collected by the ground test terminal interacting with the UAV; according to the sample UAV pose data and the sample terminal position data, multiple sample position relationship features between the UAV and the ground test terminal are established; the initial signal prediction model is trained using the multiple sample position relationship features to obtain the target signal prediction model.

[0112] In the embodiment of the present application, the information processing device can obtain the sample UAV pose data transmitted by the UAV based on the satellite device. The sample UAV pose data transmitted by the UAV based on the satellite device can be historical UAV pose data, including the historical flight attitude information of the UAV, the historical position information of the UAV, etc. The sample terminal position data can be the historical terminal position data collected by the ground test terminal, including the historical reference signal received power (RSRP) measured by the ground test terminal, the historical position of the ground test terminal, and other information.

[0113] In the embodiment of the present application, the signal strength detection can obtain the sample UAV pose data transmitted by the UAV based on the satellite device and the sample terminal position data collected by the ground test terminal interacting with the UAV in the database; it can also obtain the sample UAV pose data transmitted by the UAV based on the satellite device and the sample terminal position data collected by the ground test terminal interacting with the UAV through other means; the specific method for obtaining the sample UAV pose data transmitted by the UAV based on the satellite device and the sample terminal position data collected by the ground test terminal interacting with the UAV can be determined according to the actual situation, and the embodiment of the present application does not limit this.

[0114] In the embodiment of the present application, since a transmitting base station is provided on the UAV, the UAV can use the transmitting base station to transmit the sample UAV pose data to the satellite device, and then the satellite device transmits the sample UAV pose data to the signal strength detection device through ground wireless relay. Since the ground test terminal is connected to the signal strength detection device, the signal strength detection device can directly obtain the sample terminal position data in the ground test terminal interacting with the UAV.

[0115] In an embodiment of the present application, the implementation manner of establishing multiple sample position relationship features between the drone and the ground test terminal according to the sample drone pose data and the sample terminal position data is the same as the implementation manner of establishing multiple position relationship features between the drone and the ground test terminal according to the drone pose data and the terminal position data. The specific implementation process can refer to the implementation process of establishing multiple position relationship features between the drone and the ground test terminal according to the drone pose data and the terminal position data.

[0116] In an embodiment of the present application, the process of training an initial signal prediction model using multiple sample position relationship features to obtain a target signal prediction model in signal strength detection includes: screening at least one sample target feature from the multiple sample position relationship features; using the at least one sample target feature to train the initial signal prediction model to obtain the target signal prediction model.

[0117] In an embodiment of the present application, the method of screening at least one sample target feature from the multiple sample position relationship features is the same as the method of screening at least one target feature from the multiple position relationship features. Specifically, reference can be made to the implementation process of screening at least one target feature from the multiple position relationship features.

[0118] In an embodiment of the present application, the process of training an initial signal prediction model using at least one sample target feature to obtain a target signal prediction model by the signal strength detection device includes: inputting the at least one sample target feature into the initial signal prediction model to obtain an output path loss; determining a model loss of the initial signal prediction model according to the output path loss and the sample path loss; in the case where the model loss is greater than or equal to a preset loss threshold, continuing to train the initial signal prediction model using the at least one sample target feature and the sample path loss to obtain a training model; in the case where the training model loss of the training model is less than the preset loss threshold, determining the training model as the target signal prediction model.

[0119] It should be noted that the sample path loss is the path loss when signal transmission occurs between the drone corresponding to the multiple sample position relationship features and the ground test terminal. The determination method of the sample path loss is the same as the determination method of the first path loss. The specific method of determining the sample path loss can refer to the implementation process of determining the first path loss.

[0120] In an embodiment of the present application, the preset loss threshold can be a loss value configured in the signal strength detection device, or a loss value transmitted from other devices to the signal strength detection device, or a loss value obtained by the signal strength detection device through other means. The specific way for the signal strength detection device to obtain the preset loss threshold can be determined according to the actual situation, and the present application embodiment does not limit this.

[0121] In the embodiments of the present application, the initial signal prediction model can be a model established using the XGBoost algorithm, or a model established by other means. Specifically, the initial signal prediction model can be determined according to the actual situation, and the embodiments of the present application do not limit this.

[0122] In the embodiments of the present application, the process of the signal strength detection device determining the model loss of the initial signal prediction model according to the output path loss and the sample path loss includes: determining the training loss according to the output path loss and the sample path loss; obtaining the gradient update parameters of the initial signal prediction model; determining the regularization loss of the initial signal prediction model according to the gradient update parameters; and determining the model loss of the initial signal prediction model according to the training loss and the regularization loss.

[0123] In the embodiments of the present application, the process of determining the model loss of the initial signal prediction model according to the training loss and the regularization loss can be to determine the sum of the training loss and the regularization loss, so as to obtain the model loss.

[0124] In the embodiments of the present application, the XGBoost algorithm is used as the core algorithm for inverting the air-to-ground path loss in the high-dynamic UAV emergency scenario. The loss function Loss of the algorithm is shown in formula (14), which symbolizes the fitting of the model to the data. y i and represent the actual and predicted path loss values respectively. y i is the path loss value P calculated by combining the measured data with formula (12) L , is the path loss value predicted by the XGBoost algorithm. x represents the input features {linear distance, horizontal distance, first elevation angle, heading angle, first azimuth angle}, that is

[0125]

[0126]

[0127] Among them, the loss function (i.e., the model loss) consists of two parts: the training error and the regularization term. The regularization term is represented by formula (15). It is the sum of the regularization terms of i trees and represents the penalty parameter of the model, which is used to solve the overfitting problem. The parameters γ, T, λ, ω j represent the gradient update parameters of the model. f i represents the i-th tree, represents the path loss sample training error.

[0128] When calling the XGboost data packet interface to solve the optimal parameter values, steps (1)-(4) need to be passed through:

[0129] (1) First, the test dataset obtained needs to be cross-validated and divided into a training set and a test set: {x_train, y_trian; x_test, y_test}

[0130] (2) The underlying of Xgboost is based on decision trees and is optimized through the optimal splitting point. The adjustable parameters of the model are: Params = {objective, lamda, gamma, max_depth}, where:

[0131] objective: Solve the optimal regression tree based on this function.

[0132] lamda: The coefficient of the regularization weight term. Increasing this value makes the model more conservative.

[0133] gamma: The penalty term coefficient, specifying the minimum loss function decrease required for node splitting.

[0134] max_depth: Specify the maximum depth of the tree and find the optimal value to prevent overfitting.

[0135] (3) Use the test set data to fit the model XGBRgressor(params).fit(X_train, y_train), and judge the quality and compliance of the model according to the fitting results MSE (mean squared error) or MAE (mean absolute error). If the error does not meet the standard, continuously iterate to correct the parameters to find the optimal value.

[0136] (4) After training, obtain the optimal parameters Params of the required model and save the model.

[0137] In the embodiment of this application, the signal strength detection device is a ground-air propagation model calibration system and algorithm based on a fixed-wing unmanned aerial vehicle. The system installs a transmitting base station on a high-altitude unmanned aerial vehicle, collects propagation model data at a specified height and scene based on a large fixed-wing carrying an aerospace base station, transmits the collected data to the cloud platform center. The cloud platform center combines the received flight attitude data to construct feature vectors such as 3D dynamic antenna gain in real time, and constructs a propagation model and calibrates parameters based on a machine learning model of the XGboost algorithm on the cloud platform to obtain a calibrated system, and then can estimate the ground-air wireless coverage signal. The specific processing flow chart of the signal strength detection device is as Figure 6As shown: First, use drones, satellite equipment, ground test terminals, ground wireless relays, cloud platforms, etc. to build the system. Then, the cloud platform center issues commands to collect database information, obtaining the drone pose data (drone attitude data) transmitted by the drone based on satellite equipment and the terminal position data (ground test terminal data) collected by the ground test terminal interacting with the drone. Preprocess the drone pose data and the terminal position data respectively to obtain the preprocessed drone pose data and the preprocessed terminal position data. Based on the preprocessed drone pose data and the preprocessed terminal position data, establish multiple position relationship features: including multi-dimensional dynamic antenna gain (dynamic 3D gain), horizontal distance (2D distance), straight-line distance (3D distance), elevation angle (first elevation angle, second elevation angle), azimuth angle (first azimuth angle, second azimuth angle), etc. Then, select at least one sample target feature from the multiple position relationship features; use the at least one sample target feature to train the initial signal prediction model to obtain the target signal prediction model (prediction by Xgboost algorithm); use the target signal prediction model to predict the signal strength of the area to be detected (network signal prediction).

[0138] It can be understood that the ground propagation model system and method in the prior art are only applicable to ground and low-altitude scenarios below 300m. The method in the embodiments of the present application fills the gap in the air-to-ground medium and high-altitude propagation model correction system. The traditional propagation model correction system has a fixed and stationary emission source, which does not meet the high-dynamic operation scenario of the drone hovering flight during emergency rescue. It does not consider the influence of other parameters such as antenna dynamic gain, drone flight attitude, elevation angle, etc. on the propagation model construction in the high-dynamic scenario. In the correction system of the embodiments of the present application, by installing a transmitting base station on the high-altitude drone, the ground test end transmits the received data to the cloud platform center. The cloud platform center combines the received feature vectors considering features such as 3D dynamic antenna gain in the high-dynamic scenario during the flight process, and can comprehensively consider the influencing factors of the propagation model in each feature environment by using the XGboost machine learning algorithm, and the constructed model has higher prediction accuracy.

[0139] It can be understood that the signal strength detection device establishes multiple position relationship features between the drone and the ground test terminal according to the drone pose data and the terminal position data. The multiple position relationship features include relationship features of multiple types such as distance, angle, and channel; enabling the signal parameters of the air-to-ground wireless channel in the emergency state of the fixed-wing drone to be accurately described by using the relationship features of multiple types when the drone carrying the aerial base station continuously hovers and moves at high speed, that is, improving the accuracy during network signal prediction.

[0140] Based on the same inventive concept as the above signal strength detection method, the embodiments of the present application provide a signal strength detection device 10, corresponding to a signal strength detection method;Figure 7 Structural schematic diagram of a signal strength detection device provided by an embodiment of the present application Figure 1 The signal strength detection device 10 may include:

[0141] A receiving unit 101, configured to receive the drone pose data of the drone in the area to be detected, and the terminal position data collected by a ground test terminal interacting with the drone;

[0142] A establishing unit 102, configured to establish a plurality of position relationship features between the drone and the ground test terminal according to the drone pose data and the terminal position data;

[0143] An input unit 103, configured to input the plurality of position relationship features into a target signal prediction model to obtain the signal strength of the area to be detected.

[0144] In some embodiments of the present application, the plurality of position relationship features include:

[0145] The horizontal distance between the drone and the ground test terminal, the vertical distance between the drone and the ground test terminal, the straight-line distance between the drone and the ground test terminal, the first azimuth angle of the ground test terminal relative to the drone, the first elevation angle of the ground test terminal relative to the drone, the second azimuth angle of the ground test terminal relative to the antenna, the second elevation angle of the ground test terminal relative to the antenna, and the multi-dimensional dynamic antenna gain.

[0146] In some embodiments of the present application, the device further includes a screening unit and a determining unit;

[0147] The screening unit is configured to screen at least one target feature from the plurality of position relationship features;

[0148] The input unit 103 is configured to input the at least one target feature into the target signal prediction model to obtain the target path loss of the area to be detected;

[0149] The determining unit is configured to determine the signal strength according to the target path loss.

[0150] In some embodiments of the present application, the device further includes a sorting unit and an obtaining unit;

[0151] The determining unit is configured to determine a first path loss; respectively determine the correlation coefficients between the plurality of position relationship features and the first path loss to obtain a plurality of correlation coefficients;

[0152] The sorting unit is configured to sort the plurality of correlation coefficients in a numerical descending order to obtain a sorted sequence;

[0153] The screening unit is configured to start from the first sorting position of the sorting sequence, screen a preset number of correlation coefficients, and obtain at least one target correlation coefficient;

[0154] The obtaining unit is configured to obtain at least one target feature corresponding to the at least one target correlation coefficient among the multiple position relationship features.

[0155] In some embodiments of the present application, the determining unit is configured to determine a multi-dimensional dynamic antenna gain; and determine the first path loss according to the multi-dimensional dynamic antenna gain, the cell transmission power, the receiving antenna gain, and the received power;

[0156] The obtaining unit is configured to obtain the cell transmission power of the drone transmitting a signal to the ground test terminal, the receiving antenna gain of the ground test terminal, and the received power of the ground test terminal receiving the signal transmitted by the drone.

[0157] In some embodiments of the present application, the obtaining unit is configured to obtain the corresponding relationship between a preset second elevation angle, a preset second azimuth angle, and a preset gain;

[0158] The determining unit is configured to determine the multi-dimensional dynamic antenna gain according to the second elevation angle, the second azimuth angle, and the corresponding relationship.

[0159] In some embodiments of the present application, the device further includes a processing unit;

[0160] The processing unit is configured to preprocess the drone pose data to obtain preprocessed drone pose data; and preprocess the terminal position data to obtain preprocessed terminal position data;

[0161] The establishing unit 102 is configured to establish the multiple position relationship features according to the preprocessed drone pose data and the preprocessed terminal position data.

[0162] In some embodiments of the present application, the device further includes a training unit;

[0163] The obtaining unit is configured to obtain sample drone pose data transmitted by the drone, and sample terminal position data collected by the ground test terminal interacting with the drone;

[0164] The establishing unit 102 is configured to establish multiple sample position relationship features between the drone and the ground test terminal according to the sample drone pose data and the sample terminal position data;

[0165] The training unit is configured to train an initial signal prediction model by using the multiple sample position relationship features to obtain a target signal prediction model.

[0166] In some embodiments of the present application, the screening unit is configured to screen at least one sample target feature from the multiple sample position relationship features;

[0167] The training unit is configured to train an initial signal prediction model by using the at least one sample target feature to obtain a target signal prediction model.

[0168] In some embodiments of the present application, the input unit 103 is configured to input the at least one sample target feature into the initial signal prediction model to obtain an output path loss;

[0169] The determining unit is configured to determine a model loss of the initial signal prediction model according to the output path loss and the sample path loss; and determine the training model as the target signal prediction model when the training model loss of the training model is less than the preset loss threshold;

[0170] The training unit is configured to continue to train the initial signal prediction model by using the at least one sample target feature and the sample path loss to obtain a training model when the model loss is greater than or equal to the preset loss threshold.

[0171] In some embodiments of the present application, the determining unit is configured to determine a training loss according to the output path loss and the sample path loss; determine a regularization loss of the initial signal prediction model according to the gradient update parameter; and determine a model loss of the initial signal prediction model according to the training loss and the regularization loss;

[0172] The obtaining unit is configured to obtain a gradient update parameter of the initial signal prediction model.

[0173] It should be noted that in practical applications, the above receiving unit 101, establishing unit 102, and input unit 103 can be implemented by a processor 104 on the signal strength detection device 10, specifically implemented by a CPU (Central Processing Unit), MPU (Microprocessor Unit), DSP (Digital Signal Processing), or field programmable gate array (FPGA, Field Programmable Gate Array), etc.; the above data storage can be implemented by a memory 105 on the signal strength detection device 10.

[0174] An embodiment of the present application also provides a signal strength detection device 10, as Figure 8 shown. The signal strength detection device 10 includes: a processor 104, a memory 105, and a communication bus 106. The memory 105 communicates with the processor 104 through the communication bus 106. The memory 105 stores programs executable by the processor 104. When the programs are executed, the signal strength detection method described above is executed by the processor 104.

[0175] In practical applications, the above-mentioned memory 105 may be a volatile memory, such as a random access memory (RAM); or a non-volatile memory, such as a read-only memory (ROM), a flash memory, a hard disk drive (HDD), or a solid-state drive (SSD); or a combination of the above types of memories, and provides instructions and data to the processor 104.

[0176] An embodiment of the present application provides a computer-readable storage medium with a computer program thereon. When the program is executed by the processor 104, the signal strength detection method described above is implemented.

[0177] Exemplarily, an embodiment of the present application also provides a computer program product, including a computer program, which can be executed by the processor 104 of the signal strength detection device 10 to complete the steps described in any of the foregoing methods.

[0178] It can be understood that the signal strength detection device establishes multiple position relationship features between the unmanned aerial vehicle and the ground test terminal according to the unmanned aerial vehicle pose data and the terminal position data. The multiple position relationship features include multiple types of relationship features such as distance, angle, and channel. This enables the accurate description of the signal parameters of the air-ground wireless channel in the emergency state of a fixed-wing unmanned aerial vehicle when the unmanned aerial vehicle carrying an aerial base station continuously circles and moves at high speed, that is, the accuracy of network signal prediction is improved.

[0179] In another embodiment of the present application, refer to Figure 9 , which shows the composition structure diagram of the signal strength detection system 1 provided by the embodiment of the present application. As Figure 9 shown, the signal strength detection system includes the aforementioned signal strength detection device 10, an unmanned aerial vehicle 11, and a ground test terminal 12; wherein,

[0180] The drone 11 is used to collect drone pose data;

[0181] The ground test terminal 12 is used to collect terminal position data;

[0182] The signal strength detection device 10 is used to establish multiple position relationship features between the drone and the ground test terminal according to the drone pose data and the terminal position data; input the multiple position relationship features into a target signal prediction model to obtain the signal strength of the area to be detected.

[0183] In some embodiments of the present application, the signal strength detection system further includes a satellite device;

[0184] The satellite device is used to obtain the drone pose data and forward the drone pose data to the signal strength detection device.

[0185] It can be understood that the signal strength detection system sets up a transmitting base station on the drone, uses the satellite device to forward the drone pose data to the signal strength detection device, and uses the ground test terminal to collect the terminal position data to the signal strength detection device, so as to establish multiple position relationship features between the drone and the ground test terminal according to the drone pose data and the terminal position data. The multiple position relationship features include multiple types of relationship features such as distance, angle, and channel; it enables the signal parameters of the air-ground wireless channel in the emergency state of the fixed-wing drone to be accurately described by using multiple types of relationship features when the drone carrying the aerial base station continuously hovers and moves at high speed, that is, the accuracy of network signal prediction is improved.

[0186] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) containing computer-usable program code.

[0187] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram can be realized by computer program instructions, and the combination of the processes and / or blocks in the flowchart and / or block diagram can also be realized by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate for realizing the process Figure 1One or more processes and / or blocks Figure 1 Apparatus for the functions specified in one or more blocks

[0188] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce a manufacture including an instruction apparatus that implements the functions in the process Figure 1 One or more processes and / or blocks Figure 1 The functions specified in one or more blocks

[0189] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus, such that a series of operational steps are performed on the computer or other programmable apparatus to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions in the process Figure 1 One or more processes and / or blocks Figure 1 The steps of the functions specified in one or more blocks

[0190] As described above, only the preferred embodiments of this application are given and are not intended to limit the protection scope of this application

Claims

1. A signal strength detection method, characterized in that, The method includes: Receiving the UAV pose data of the UAV in the area to be detected, and the terminal position data collected by a ground test terminal interacting with the UAV; Establishing a plurality of position relationship features between the UAV and the ground test terminal according to the UAV pose data and the terminal position data; Inputting the plurality of position relationship features into a target signal prediction model to obtain the signal strength of the area to be detected.

2. The method according to claim 1, wherein The plurality of position relationship features include: The horizontal distance between the UAV and the ground test terminal, the vertical distance between the UAV and the ground test terminal, the straight-line distance between the UAV and the ground test terminal, the first azimuth angle of the ground test terminal relative to the UAV, the first elevation angle of the ground test terminal relative to the UAV, the second azimuth angle of the ground test terminal relative to the antenna, the second elevation angle of the ground test terminal relative to the antenna, and the multi-dimensional dynamic antenna gain.

3. The method according to claim 1, wherein The inputting the plurality of position relationship features into a target signal prediction model to obtain the signal strength of the area to be detected includes: Screening at least one target feature from the plurality of position relationship features; Inputting the at least one target feature into the target signal prediction model to obtain the target path loss of the area to be detected; Determining the signal strength according to the target path loss.

4. The method according to claim 3, wherein The screening at least one target feature from the plurality of position relationship features includes: Determining a first path loss; Respectively determining the correlation coefficients between the plurality of position relationship features and the first path loss to obtain a plurality of correlation coefficients; Sorting the plurality of correlation coefficients in a numerical descending order to obtain a sorting sequence; Starting from the first sorting position of the sorting sequence, screening a preset number of correlation coefficients to obtain at least one target correlation coefficient; Obtaining at least one target feature corresponding to the at least one target correlation coefficient among the plurality of position relationship features.

5. The method according to claim 4, characterized in that, The determining a first path loss includes: Determining the multi-dimensional dynamic antenna gain; Obtaining the cell transmission power of the UAV transmitting a signal to the ground test terminal, the receiving antenna gain of the ground test terminal, and the receiving power of the ground test terminal receiving the signal transmitted by the UAV; Determining the first path loss according to the multi-dimensional dynamic antenna gain, the cell transmission power, the receiving antenna gain, and the receiving power.

6. The method according to claim 5, characterized in that, The determining the multi-dimensional dynamic antenna gain includes: Obtaining the corresponding relationship between a preset second elevation angle, a preset second azimuth angle, and a preset gain; Determining the multi-dimensional dynamic antenna gain according to the second elevation angle, the second azimuth angle, and the corresponding relationship.

7. The method according to claim 1, characterized in that, The establishing a plurality of position relationship features between the UAV and the ground test terminal according to the UAV pose data and the terminal position data includes: Performing preprocessing on the UAV pose data to obtain preprocessed UAV pose data; Performing preprocessing on the terminal position data to obtain preprocessed terminal position data; Based on the preprocessed UAV pose data and the preprocessed terminal position data, establish the multiple position relationship features.

8. The method according to claim 1, wherein Before inputting the multiple position relationship features into the target signal prediction model to obtain the signal strength of the area to be detected, the method further includes: Obtain the sample UAV pose data transmitted by the UAV, and the sample terminal position data collected by the ground test terminal interacting with the UAV; Based on the sample UAV pose data and the sample terminal position data, establish multiple sample position relationship features between the UAV and the ground test terminal; Use the multiple sample position relationship features to train the initial signal prediction model to obtain the target signal prediction model.

9. The method according to claim 8, wherein The using the multiple sample position relationship features to train the initial signal prediction model to obtain the target signal prediction model includes: Screen out at least one sample target feature from the multiple sample position relationship features; Use the at least one sample target feature to train the initial signal prediction model to obtain the target signal prediction model.

10. The method according to claim 9, wherein The using the at least one sample target feature to train the initial signal prediction model to obtain the target signal prediction model includes: Input the at least one sample target feature into the initial signal prediction model to obtain the output path loss; Determine the model loss of the initial signal prediction model according to the output path loss and the sample path loss; When the model loss is greater than or equal to the preset loss threshold, use the at least one sample target feature and the sample path loss to continue training the initial signal prediction model to obtain a training model; When the training model loss of the training model is less than the preset loss threshold, determine the training model as the target signal prediction model.

11. The method according to claim 10, characterized in that, The determining the model loss of the initial signal prediction model according to the output path loss and the sample path loss includes: Determine the training loss according to the output path loss and the sample path loss; Obtain the gradient update parameter of the initial signal prediction model; Determine the regularization loss of the initial signal prediction model according to the gradient update parameter; Determine the model loss of the initial signal prediction model according to the training loss and the regularization loss.

12. A signal strength detection system, characterized in that, The system includes a UAV, a ground test terminal, and a signal strength detection device: The UAV is used to collect UAV pose data; The ground test terminal is used to collect terminal position data; The signal strength detection device is used to establish multiple position relationship features between the UAV and the ground test terminal according to the UAV pose data and the terminal position data; Input the multiple position relationship features into the target signal prediction model to obtain the signal strength of the area to be detected.

13. A signal strength detection device, characterized in that, The device includes: A receiving unit, configured to receive the UAV pose data of the UAV in the area to be detected, and the terminal position data collected by the ground test terminal interacting with the UAV; A establishing unit, configured to establish multiple position relationship features between the UAV and the ground test terminal according to the UAV pose data and the terminal position data; An input unit for inputting the multiple positional relationship features into a target signal prediction model to obtain the signal strength of a region to be detected.

14. A signal strength detection device, characterized in that, The device includes: A memory, a processor, and a communication bus. The memory communicates with the processor through the communication bus. The memory stores a program for signal strength detection executable by the processor. When the program for signal strength detection is executed, the method according to any one of claims 1 to 11 is executed by the processor.

15. A storage medium having a computer program stored thereon, which is applied to a signal strength detection device, characterized in that, When the computer program is executed by the processor, the method according to any one of claims 1 to 11 is implemented.

16. A computer program product comprising a computer program, characterized in that, When the computer program is executed by the processor, the method according to any one of claims 1 to 11 is implemented.

Images