Method and device for correcting simulation communication link data

Abstract

The invention provides a simulation communication link data correction method and device, and the method comprises the steps: obtaining first simulation communication link data in a target three-dimensional space from a communication simulation model, and obtaining first real communication link data on a flight path synchronously collected by an unmanned plane; performing space-time reference alignment processing on the first simulation communication link data and the first real communication link data to obtain second simulation communication link data and second real communication link data which are aligned in time and space; and correcting the second simulation communication link data according to the second real communication link data to obtain corrected communication link data. Therefore, the time-space reference alignment processing can ensure that the simulation data and the measured data have strict comparability under the unified time coordinate and space coordinate system, and the second simulation communication link data is subjected to targeted correction according to the aligned real communication link data, so that the accuracy of the second simulation communication link data is improved. The fidelity and prediction accuracy of a simulation result in a complex airspace scene can be improved.

Description

Technical Field

[0001] This application relates to the field of communication technology, and in particular to a method and apparatus for correcting simulated communication link data. Background Technology

[0002] In the field of low-altitude communication, related technologies typically rely primarily on test data, such as communication link data collected by UAVs (hereinafter referred to as flight test data), supplemented by prediction data from communication simulation models. These data are used independently or indirectly verified only through "simulation-actual comparison," failing to effectively solve the problem of integrating flight test data and simulation prediction data. Summary of the Invention

[0003] This application provides a method, apparatus, and electronic device for correcting simulated communication link data. The specific solution is as follows: The first aspect of this application proposes a method for correcting simulated communication link data, including: The first simulated communication link data within the target's three-dimensional space is obtained from the communication simulation model, and the first real communication link data on the flight trajectory synchronously collected by the UAV is also obtained; wherein, the flight trajectory is located within the target's three-dimensional space; The first simulated communication link data and the first real communication link data are aligned with a spatiotemporal reference to obtain the second simulated communication link data and the second real communication link data that are aligned in time and space. Based on the second real communication link data, the second simulated communication link data is corrected to obtain the corrected communication link data.

[0004] A second aspect of this application provides a device for correcting simulated communication link data, comprising: The acquisition module is used to acquire the first simulated communication link data in the target's three-dimensional space from the communication simulation model, and to acquire the first real communication link data on the flight trajectory synchronously acquired by the UAV; wherein, the flight trajectory is located in the target's three-dimensional space; The alignment module is used to perform spatiotemporal reference alignment processing on the first simulated communication link data and the first real communication link data to obtain the second simulated communication link data and the second real communication link data that are aligned in time and space. The correction module is used to correct the second simulated communication link data based on the second real communication link data to obtain the corrected communication link data.

[0005] A third aspect of this application provides an electronic device, including: a processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the method described in the first aspect embodiment above.

[0006] A fourth aspect of this application provides a computer-readable storage medium storing computer-executable instructions that, when executed by a processor, are used to implement the method described in the first aspect above.

[0007] A fifth aspect of this application provides a computer program product including a computer program that, when executed by a processor, implements the method described in the first aspect of the present application.

[0008] The method, apparatus, electronic device, and storage medium for correcting true communication link data provided in this application ensure the comparability of simulation and measured data in spatiotemporal reference by synchronously acquiring the first simulated communication link data output from the communication simulation model and the first real communication link data obtained by the UAV on the flight trajectory in the target three-dimensional space. Furthermore, by aligning the two data in spatiotemporal reference, the simulation data and measured data are ensured to be strictly comparable under a unified time coordinate and spatial coordinate system. Based on this, the second simulated communication link data is specifically corrected using the aligned second real communication link data as a calibration benchmark, which can improve the fidelity and prediction accuracy of the simulation results in complex airspace scenarios.

[0009] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0010] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 A flowchart illustrating a method for correcting simulated communication link data provided in an embodiment of this application; Figure 2 A flowchart illustrating another method for correcting simulated communication link data provided in this application embodiment; Figure 3 A flowchart illustrating another method for correcting simulated communication link data provided in this application embodiment; Figure 4 This is a schematic diagram illustrating the fusion process of simulation data and flight test data provided in the embodiments of this application. Figure 5 This is a schematic diagram of a device for correcting simulated communication link data provided in an embodiment of this application. Detailed Implementation

[0011] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0012] It should be noted that the acquisition, storage, use, and processing of data in this application comply with the relevant provisions of national laws and regulations and do not violate public order and good morals.

[0013] The following description, with reference to the accompanying drawings, outlines a method, apparatus, electronic device, and storage medium for correcting simulated communication link data according to embodiments of this application.

[0014] Figure 1 This is a flowchart illustrating a method for correcting simulated communication link data, provided in an embodiment of this application.

[0015] like Figure 1 As shown, the method for correcting simulated communication link data includes the following steps: Step 101: Obtain the first simulated communication link data in the target three-dimensional space from the communication simulation model, and obtain the first real communication link data on the flight trajectory synchronously collected by the UAV.

[0016] In this application, the communication simulation model is a virtual communication system constructed by software, used to simulate the transmission, propagation, reception, and performance of wireless signals in the real world in a digital environment.

[0017] For example, a communication simulation model can be started to simulate the transmission, propagation, and reception of wireless signals in the target's three-dimensional space. After the communication simulation model is running, its output interface can provide a standardized data stream or file format from which data within a specific time period can be extracted to obtain the first simulated communication link data.

[0018] For example, the first simulated communication link data may include spatiotemporal information of all sampling points in the target three-dimensional space, related communication performance indicators, etc.

[0019] For example, the first simulated communication link data may include, but is not limited to, spatial location coordinates [x] sim , y sim ,z sim (Unit: meters), where [x] sim , y sim ] represents a two-dimensional planar coordinate system, z sim Indicates altitude; signal strength S sim (e.g., dBm or modulus relative to a reference); timestamp; end-to-end delay T sim(Units include, for example, milliseconds or microseconds); RSRP (Reference Signal Received Power) can be expressed as RSRP. sim SINR (Signal to Interference plus Noise Ratio) can be expressed as SINR. sim Among them, the end-to-end delay T sim RSRP sim SINR sim These are all communication performance indicators.

[0020] For example, the time resolution requirement for the communication simulation model can be that the timestamp interval of the first simulated communication link data is no greater than [a certain value]. seconds (e.g.) The time resolution is set to a specific value (in seconds) to ensure comparability in both time and space with subsequent communication link data collected by the drone. Choosing this time resolution ensures that key state points in the dynamic process can be captured and that the matching accuracy required for subsequent data fusion is met.

[0021] Before using drones to collect communication link data, one or more drones equipped with high-precision radio frequency sensors can be prepared. These drones are used to fly in a real environment and one or more preset paths are planned in advance. The flight control system precisely controls the drones to perform flight tasks along the preset paths. During the flight of the drones, the radio frequency sensors collect the first real communication link data on the flight trajectory in real time.

[0022] For example, these preset paths can cover the area to be fused within the target's three-dimensional space. For instance, a drone can travel along the actual link trajectory of the communication simulation model that needs to be corrected within the target area, or at least pass through its key internal locations.

[0023] For example, the target 3D space can be a cubic region within the target area that meets the side length requirements. For instance, the region to be fused is a cubic region with a side length less than X meters and greater than Y meters (e.g., X=50km, Y=10km). Choosing this size can cover typical communication scenario requirements while ensuring computational efficiency.

[0024] For example, the drone's flight path lies within the target's three-dimensional space.

[0025] For example, the first real communication link data may include, but is not limited to, timestamps, spatial coordinates, signal strength, carrier frequency, RSRP, SINR, etc. Here, the spatial coordinates are the coordinates of the UAV's spatial position point on its flight path, and the signal strength, RSRP, SINR, etc. in the first real communication link data are communication link performance information collected by the UAV at the time indicated by the timestamp when it flies to that spatial coordinate.

[0026] For example, a drone can use a GPS positioning system to obtain latitude and longitude coordinates and convert them into a unified spatial coordinate system, or use an inertial navigation system to achieve centimeter-level accuracy in position and attitude information.

[0027] For example, the positioning accuracy of a drone can be no less than W. real Rice, for example, W real It can be set to 5m, and the time synchronization accuracy requirement is within Tst. real Within microseconds, such as Tst real =10μs or use GNSS (Global Navigation Satellite System) timestamps, with an accuracy better than 100 nanoseconds.

[0028] In this application, data collaborative acquisition from two sources (i.e., communication simulation model and UAV) can be realized, and the data dimensions of the first simulated communication link data and the first real communication link data are matched.

[0029] In some embodiments, simulated communication link data may also be referred to as predicted communication link data, and real communication link data may also be referred to as measured communication link data. This application does not limit the names of simulated communication link data, real communication link data, etc.

[0030] Step 102: Perform spatiotemporal reference alignment processing on the first simulated communication link data and the first real communication link data to obtain second simulated communication link data and second real communication link data that are aligned in time and space.

[0031] In this application, during the process of correcting the communication link data output by the communication simulation model, it is necessary to ensure that the simulation data and the actual measurement data of the UAV can correspond accurately.

[0032] In this application, spatial and temporal reference alignment can be performed on the first simulated communication link data and the first real communication link data to obtain second simulated communication link data and second real communication link data that are aligned in time and space.

[0033] For example, the spatial location information in the first simulated communication link data and the spatial location information in the first real communication link data can be mapped to the same spatial coordinate system to obtain the third simulated communication link data and the third real communication link data. The timestamps in the third simulated communication link data and the third real communication link data can be converted to the same time reference system to obtain the fourth simulated communication link data and the second real communication link data. Then, the fourth simulated communication link data can be interpolated to obtain the second simulated communication link data, so that the resolution of the second simulated communication link data matches the sampling frequency of the UAV.

[0034] For example, the sampling frequency f of the drone drone This indicates how many data points the drone collects per second.

[0035] As an example, the first simulated communication link data D is obtained. sim D sim It includes unconverted spatial location information [x, y, z], and also acquires measured data collected by the UAV (i.e., the first real communication link data) D real D real It already includes location information collected based on standard positioning systems (such as GPS) [X] UTM / Y WGS84 / Z altitude The Cartesian rectangular coordinate system is used as the target normalization reference. D sim The spatial coordinate information in the simulation is mapped to a unified target coordinate system. The specific standard used can be UTM or WGS84 to obtain the third simulation communication link data. ,Will After converting to the same time reference system, interpolation is performed to obtain the second simulated communication link data. .

[0036] For example, the target spatial coordinate system can be determined based on actual application requirements and data compatibility considerations. For instance, in applications requiring high-precision positioning and terrain matching, such as urban 3D modeling or accurate path planning correction, WGS84 can be selected as the geographic reference frame, and UTM can be used for planar projection processing. The accuracy requirements of the coordinate transformation should meet the needs of subsequent data fusion, error calculation, and model validation. For example, when correcting simulation models to improve prediction accuracy, the normalization error of spatial location typically needs to be controlled within a certain range. Within a meter range (e.g.) (m). This range was chosen to ensure positioning consistency and to provide a basis for subsequent precise comparisons. For example, if the original simulation coordinate system S... origIf the coordinates are private or different from the standard 3D spatial grid coordinates, then specific transformation functions (such as using affine transformation matrices, projection transformation libraries, etc.) can be written to accurately map them to the UTM / WGS84 coordinate system.

[0037] For example, the interpolation method can be selected based on the characteristics of the communication simulation model and the required correction accuracy. For example, the interpolation method can include, but is not limited to, linear interpolation, spline interpolation, etc. For instance, to obtain the first real communication link data D... real Each sampling point precisely corresponds to a specific instant t and position (X). t , Y t Z t If linear interpolation is selected, processing can be performed based on a specified field (such as signal strength) at instant t.

[0038] For example, the time resolution of the third simulated communication link data (i.e., the time interval between its output data points) is denoted as... (For example s), this interval may be much larger than the sampling frequency of the UAV, resulting in an inaccurate correspondence. The goal of interpolation is to align the behavioral characteristics of the communication simulation model at a finer time scale (such as the trend of signal strength change over time) with real, high-frequency measured data. The time resolution after interpolation can be set to s. (For example s), the interpolation process uses the third simulated communication link data Based on the data points, calculations are performed between two adjacent data points according to their internal time field information to generate a result that meets the required time resolution. The sampling frequency of the drone, fdrone, is used to create a dense sequence of simulated data points, thereby obtaining second simulated communication link data that has undergone spatiotemporal alignment and carries standard coordinate information.

[0039] For example, if the spatial location information of the first real communication link data is already in the target coordinate system, then only the spatial location information in the first simulated communication link data can be mapped to the target coordinate system to obtain the third simulated communication link data. Similarly, if the timestamp of the first real communication link data is in the target time reference system, then only the timestamp in the first real communication link data can be converted to the target time reference system.

[0040] In a three-dimensional spatial communication simulation correction scenario in a city, the first simulated communication link data D covering a certain urban area is obtained. sim (e.g., containing) The simulated communication link data for the region. The original simulated coordinate system x, y may be an internally defined spatial grid. Then, temporal and spatial alignment is performed: spatial alignment uses a transformation function to accurately map all (X,Y) coordinate points within this urban area to the WGS84 (World Geodetic System 1984 Coordinate System) standard geographic coordinates, ensuring they align with the first real communication link data D obtained by the UAV over this area. real The spatial location is completely aligned. Time alignment involves the UAV flying along a predetermined path at a specific altitude, recording data points in real time after GPS time calibration. The data output by the communication simulation model may be based on its calculation cycle, with large intervals (e.g., outputting once per second). In this case, an interpolation algorithm is needed to fill in time nodes with finer time steps that match the UAV's sampling frequency between each output of the communication simulation model.

[0041] In communication link correction scenarios under the influence of complex electromagnetic pulses, temporal-spatial alignment is used to perform precise temporal-spatial alignment analysis between the changes in communication parameters (such as signal strength decrease) within the pulse-affected area predicted by the simulation model and the changes actually measured by the UAV when it flies over these areas. This achieves synchronization of two sets of data from the communication simulation model and real-world UAV measurements. Temporal synchronization ensures that the simulation time and the timestamps collected by the UAV are in the same time system (e.g., UTC time or calibrated local time), enabling precise pairing of the two data sources based on time points. Spatial correlation ensures that each time sampling point corresponds to a specific three-dimensional position (x, y, z) within the target's three-dimensional space, whether it's the output of the communication simulation model or the data collected by the UAV.

[0042] In practical applications, such as 3D channel modeling in cities or link budget analysis in complex electromagnetic environments, a communication simulation model is first run to simulate the three-dimensional spatial communication situation of the target city. Then, the UAV, equipped with corresponding radio frequency sensors, performs flight missions in a real urban environment according to a preset flight plan (which may include path smoothing to adapt to the UAV's maneuverability) and collects actual measurement data.

[0043] In this application, spatial reference alignment can eliminate measurement deviations and registration errors caused by different coordinate definitions, and temporal reference alignment can align the originally irregular, low-frequency or non-corresponding simulation data points with the precise timestamps of the UAV. Spatiotemporal reference alignment can significantly reduce the impact of positioning errors and response delays caused by inconsistent reference frames on the correction effect, and solve the problem of inconsistency between the output data of the communication simulation model (which may use an internal private or non-standard coordinate system) and the actual UAV measurement data in terms of reference frame and sampling frequency.

[0044] This application, through spatiotemporal reference alignment, can solve the problems of data asynchrony between communication simulation models and the real environment, and the inability to accurately compare and correct them, by ensuring D sim and D real They are comparable in time and space (i.e., they describe the communication status at the same point in time and space), providing reliable data support for subsequent parameter correction.

[0045] Step 103: Based on the second real communication link data, correct the second simulated communication link data to obtain the corrected communication link data.

[0046] In some embodiments, the second simulated communication link data can be compared with the second real communication link data to determine the difference information between the two, and the second simulated communication link data can be corrected based on the difference information. For details, please refer to the specific explanation of the following embodiments.

[0047] In some embodiments, the second simulated communication link data can be corrected based on the second real communication link data using the Gaussian process field fusion method, as detailed in the following embodiments.

[0048] It should be noted that the method for correcting the communication link data output by the communication simulation model based on the communication link data collected by the UAV in this application can also be understood as a fusion method between the communication link data collected by the UAV and the communication link data output by the communication simulation model, that is, a data fusion method.

[0049] In this embodiment, by synchronously collecting the first simulated communication link data output by the communication simulation model and the first real communication link data obtained by the UAV on the flight trajectory in the target three-dimensional space, it can be ensured that the simulation data and the measured data are comparable in spatiotemporal reference. Furthermore, spatiotemporal reference alignment processing is performed on the two to ensure that the simulation and measured data are strictly comparable under a unified time coordinate and spatial coordinate system. On this basis, the second simulated communication link data is specifically corrected using the aligned second real communication link data as a calibration benchmark, which can improve the fidelity and prediction accuracy of the simulation results in complex airspace scenarios.

[0050] Figure 2 This is a flowchart illustrating another method for correcting simulated communication link data provided in an embodiment of this application. Figure 2 As shown, the method for correcting simulated communication link data may include the following steps: Step 201: Obtain the first simulated communication link data in the target three-dimensional space from the communication simulation model, and obtain the first real communication link data on the flight trajectory synchronously collected by the UAV.

[0051] Step 202: Perform spatiotemporal reference alignment processing on the first simulated communication link data and the first real communication link data to obtain second simulated communication link data and second real communication link data that are aligned in time and space.

[0052] In this application, steps 201-202 can be implemented in any of the embodiments of this application, so they will not be described in detail here.

[0053] Step 203: Quantize the difference features of the second simulated communication link data and the second real communication link data to obtain difference information.

[0054] In this application, the difference information can be used to measure the deviation, similarity, etc. between the second simulated communication link data and the second real communication link data.

[0055] In other words, the difference features of the second simulated communication link data and the second real communication link data are quantified to accurately identify and measure the deviation and similarity between the simulated data and the real measured data, thereby providing a quantitative basis for subsequent correction strategies.

[0056] For example, the difference features between the second simulated communication link data and the second real communication link data can be quantified by methods such as generating local error fields and evaluating spatial distribution similarity.

[0057] For example, the local error field generation uses the mean absolute error (MAE) to quantify the prediction deviation within a three-dimensional spatial region, constructs a buffer for gridded error calculation, delineates a three-dimensional buffer with the UAV's flight trajectory as the center, calculates the difference between the simulation prediction value and the actual observation value within this region, and generates a gridded error field map.

[0058] For example, spatial distribution similarity assessment is used to analyze the degree to which the overall trend of change of the second simulated communication link data matches that of the second real communication link data.

[0059] Step 204: Based on the difference information and the second real communication link data, correct the second simulated communication link data to obtain the corrected communication link data.

[0060] In this application, the predicted link quality data at any spatial location point in the second simulated communication link data can be corrected based on the difference information and the real link quality data at any spatial location point in the second real link data, so as to obtain the corrected communication link data.

[0061] For example, predicted link quality data may include, but is not limited to, communication performance indicators such as signal strength, RSRP, and SINR, while real link quality data may include, but is not limited to, communication performance indicators such as signal strength, RSRP, and SINR. For example, based on the difference information, the confidence weight at the spatial location point to be corrected can be determined. Based on this confidence weight and the real link quality data at the spatial location point to be corrected in the second real link data, the predicted link quality data at the same spatial location point to be corrected in the second simulated communication link data is corrected to obtain the corrected communication link data.

[0062] In this embodiment of the application, by quantifying the difference features between the aligned second simulated communication link data and the second real communication link data, the deviation pattern between the communication simulation model and the real environment in terms of wireless communication performance can be accurately characterized. Using this difference information as the basis for correction, and combining it with real measurement data to make targeted corrections to the simulation data, the accuracy, fidelity and reliability of the simulation results can be improved.

[0063] Figure 3 This is a flowchart illustrating another method for correcting simulated communication link data provided in an embodiment of this application.

[0064] like Figure 3 As shown, the method for correcting simulated communication link data may include the following steps: Step 301: Obtain the first simulated communication link data in the target three-dimensional space from the communication simulation model, and obtain the first real communication link data on the flight trajectory synchronously collected by the UAV.

[0065] Step 302: Perform spatiotemporal reference alignment processing on the first simulated communication link data and the first real communication link data to obtain second simulated communication link data and second real communication link data that are aligned in time and space.

[0066] In this application, steps 301-302 can be implemented in any of the embodiments of this application, so they will not be described in detail here.

[0067] Step 303: Spatial registration is performed on the second simulated communication link data and the second real communication link data to determine the same spatial location points in the second simulated communication link data and the second real communication link data, which are used as evaluation points.

[0068] Since the UAV collects data along its flight path, while the communication simulation model outputs data within the entire three-dimensional space of the target, in order to facilitate subsequent corrections, this application can perform spatial registration between the second simulated communication link data and the second real communication link data to determine at which spatial locations in the second simulated communication link data and the second real communication link data both have link quality data, and these spatial locations can be used as evaluation points.

[0069] Step 304: Spatial expansion is performed with key sampling points on the flight trajectory as the center to obtain the target three-dimensional buffer.

[0070] In this application, each spatial location point (i.e., spatial sampling point) on the flight trajectory can be used as a key sampling point. Alternatively, to reduce the amount of computation, a spatial location point can be used as a key sampling point every certain number of spatial location points.

[0071] For example, a key sampling point is used as the center point, and a three-dimensional spatial region is defined around the center point as the target three-dimensional buffer.

[0072] For example, the target three-dimensional buffer zone can be a cylindrical space region, or a regular hexagonal prism, or a cuboid prism, etc.

[0073] As an example, a cylindrical spatial region can be selected as the target 3D buffer, with its radius set to R meters (e.g., R=5m or R=10m). The radius R can be determined based on scene complexity, UAV flight accuracy, and the desired level of detail correction in the model. The reason for choosing a cylinder rather than a sphere for the target 3D buffer shape is that in many communication flight tests, changes in spatial position are often symmetrical or approximate about the centerline. Depending on the approximation method, a regular hexagonal prism or a cuboid prism could also be chosen. In complex 3D urban channel modeling scenarios, to capture the effects of buildings, streets, etc., it is necessary to correct the simulation data at a certain height above the measured trajectory line. Therefore, the target 3D buffer can be elongated along the UAV flight path and have a certain thickness in the vertical direction.

[0074] It should be noted that the specific shape and size of the target 3D buffer can be determined according to the application scenario, and there are no restrictions on this.

[0075] Step 305: Based on the second simulated communication link data and the second real communication link data, determine the mean absolute error of the key sampling points using the link quality data of the evaluation points in the target three-dimensional buffer.

[0076] In this application, evaluation points in the target three-dimensional buffer can be determined from the evaluation points determined by spatial registration of the second simulated communication link data and the second real communication link data. The mean absolute error of the key sampling points is determined based on the difference between the predicted link quality data of the second simulated communication link data in the target three-dimensional buffer and the real link quality data of the second real communication link data in the same evaluation point in the target three-dimensional buffer.

[0077] For example, taking signal strength as an example, the mean absolute error (MAE) can be calculated using the following formula (1): (1) in, Indicates the number of evaluation points within the target buffer. Indicating the second simulated communication link data, the first... Signal strength at each evaluation point Indicating the second real communication link data, the first Signal strength at each evaluation point.

[0078] Alternatively, the target three-dimensional buffer can be determined first, the simulation communication link data of the target three-dimensional buffer can be extracted from the second simulation communication link data, and the simulation communication link data of the target three-dimensional buffer can be spatially registered with the second real communication link data to determine the evaluation points in the target three-dimensional buffer.

[0079] For example, a cylindrical spatial region with a radius of R = 5 meters can be used as a three-dimensional buffer zone for the target, and the height of the cylindrical spatial region can be set. Rice (e.g.) (meters), and the distance along the path is L. path Here, the centerline (x, y) can be used as a reference for three-dimensional spatial registration to ensure that there is corresponding simulation data for each UAV spatial sampling point.

[0080] Taking a simulated and corrected scenario of urban three-dimensional spatial communication as an example, a drone flies along a path through a dense area of ​​high-rise buildings, and the first real communication link data D collected during the flight... real (Assume D) realThe data (already in the target space coordinate system and time reference system) includes the signal strength along this path. The first simulated communication link data output by the communication simulation model is aligned with the spatiotemporal reference to ensure that the second simulated communication link data covers all three-dimensional spatial positions traversed by the UAV. Taking each spatial sampling point on the UAV's flight path as the center, the data is extended vertically to a certain range above and below that height (e.g., a buffer radius R = 5m). The mean absolute error between the simulated signal strength and the measured value at all evaluation points within this cylinder is calculated. For example, if building structures in a certain area cause a significant occlusion effect, the mean absolute error may be higher in that area.

[0081] In this application, an average absolute error can be calculated for a key sampling point on the flight trajectory. Then, by quantifying and gridding the deviation (i.e., average absolute error) of each key sampling point on the flight trajectory, a three-dimensional gridded error field map can be generated, thereby intuitively showing which three-dimensional spatial regions the communication simulation model has a large prediction deviation.

[0082] Step 306: Obtain the difference information based on the mean absolute error.

[0083] In some embodiments, the mean absolute error can be used as difference information.

[0084] Since simply calculating differences at local points cannot fully reflect the overall deviation and trend capture capability of a communication simulation model—for example, in complex terrain or urban environments, signal strength variations with altitude may exhibit specific nonlinear gradients—some embodiments can obtain spatial distribution similarity and, based on the mean absolute error and spatial distribution similarity, acquire difference information.

[0085] For example, the difference information may include mean absolute error and spatial distribution similarity.

[0086] For example, spatial distribution similarity can be obtained by extracting the predicted link quality data sequence of the communication simulation model on the flight trajectory from the second simulated communication link data, extracting the real link quality data sequence from the second real communication link data, and evaluating the spatial distribution similarity between the predicted link quality data sequence and the real link quality data sequence to obtain the spatial distribution similarity.

[0087] For example, spatial distribution similarity can reflect the degree to which the overall trend of the predicted link quality data sequence matches that of the actual link quality data sequence.

[0088] For example, a trend comparison method can be used to check whether the trends of two sequences are consistent, such as whether their overall change trends both show an initial rise followed by a decline and then stabilization, in order to obtain spatial distribution similarity.

[0089] As an example, spatial distribution similarity can be determined by comparing the positions of the maximum and minimum values ​​of two sequences (e.g., the distance from the maximum and minimum values ​​to the origin) and the positions of key turning points (e.g., the distance from the zeros of each derivative to the origin). This method is relatively simple and easy to implement.

[0090] For example, statistical matching can also be used to obtain spatial distribution similarity. For instance, statistical matching uses techniques such as statistical correlation analysis or fuzzy logic to quantify the degree of matching of the overall distribution patterns.

[0091] As an example, the correlation coefficient between the predicted link quality data sequence and the actual link quality data sequence in space (such as distance, altitude, etc.) can be calculated. ( ), and based on the correlation coefficient To determine spatial distribution similarity, for example, the correlation coefficient can be used. As a measure of spatial distribution similarity, this method yields relatively high accuracy in calculating spatial distribution similarity.

[0092] Taking a simulated correction scenario with multiple hills and valleys and significant terrain undulations as an example, a drone collects data along a pre-defined flight path (e.g., from the mountaintop to the valley floor), specifically measuring the signal strength variation trend along the path. It not only calculates the mean absolute error at each key sampling point but also compares the trends of the simulated output and measured values ​​along the entire path. For example, it checks whether the simulated predicted signal strength attenuates as rapidly as in reality after crossing a ridge, and generates a spatial distribution similarity.

[0093] Step 307: Based on the difference information and the second real communication link data, correct the second simulated communication link data to obtain the corrected communication link data.

[0094] In this application, the credibility weight of each evaluation point in the region to be fused can be determined based on the difference information. Then, based on the credibility weight and the second real communication link data, the second simulated communication link data is corrected to obtain the corrected communication link data. Here, the evaluation point is a spatial location point shared by the second simulated communication link data and the second real communication link data.

[0095] For example, a confidence weight is a scalar factor or probability value that can be used by a communication simulation model to determine the degree to which predicted link quality data should be accepted at an evaluation point.

[0096] For example, the difference information includes the mean absolute error of the evaluation points. The confidence weight can be determined based on the mean absolute error in the region to be fused in the following way: the confidence weight of each evaluation point in the region to be fused is determined based on the relationship between the mean absolute error of each evaluation point in the region to be fused and a first preset threshold.

[0097] For example, if the difference between the mean absolute error and the first preset threshold is greater than zero, the larger the difference, the greater the confidence weight.

[0098] For example, the method for calculating the mean absolute error of the evaluation points can be found in the method for calculating the mean absolute error of the key sampling points in the above embodiments, and therefore will not be repeated here.

[0099] For example, each evaluation point in the region to be merged can be a key sampling point in the region to be merged.

[0100] For example, if the mean absolute error at the evaluation points exceeds a first threshold If the confidence level of the communication simulation model for that spatial location is low, then the confidence level is low. For example, the first threshold... This can be determined based on the definition of acceptable error and risk assessment. For example, the mean absolute error can be used. The region designated as the high-weight correction region, i.e., the high-error region, can be used. For example, in scenarios requiring very high prediction accuracy (such as precise path planning or link budgeting), a smaller region can be selected. Value, such as dB can be set to a slightly larger value in scenarios where there is a greater need to correct the overall trend.

[0101] For example, the difference information may include spatial distribution similarity, and the credibility weight can be determined based on the spatial distribution similarity in the following way: the credibility weight of each evaluation point in the region to be fused is determined based on the relationship between the spatial distribution similarity and the second preset threshold.

[0102] For example, low-similarity regions with spatial distribution similarity less than the second threshold can be designated as low-confidence regions, and evaluation points in low-confidence regions can be assigned higher confidence weights.

[0103] For example, if the difference information includes mean absolute error and spatial distribution similarity, the confidence weight of each evaluation point can be determined based on the mean absolute error and spatial distribution similarity of each evaluation point in the region to be fused. For example, based on the mean absolute error and a first threshold, high-error regions in the region to be fused can be identified, and based on spatial distribution similarity and a second threshold, low-similarity regions in the region to be fused can be identified, and evaluation points in these two regions can be assigned higher confidence weights.

[0104] Taking a three-dimensional spatial communication simulation correction scenario in a city as an example, a drone flies along a path that passes through dense commercial areas and gaps between tall buildings. During the flight, it collects communication link data and can calculate which points in the area (e.g., near a large billboard) have a higher mean absolute error. At the same time, the overall similarity score (i.e., spatial distribution similarity) is calculated over the entire business district. It identifies these high-error points and low-similarity regions and assigns them higher confidence weights.

[0105] In this application, a non-intrusive data fusion method can be used to correct the second simulated communication link data. "Non-intrusive" means that the basic distribution characteristics (such as the overall proportion of shape and amplitude) of the second simulated communication link data remain unchanged during the correction process.

[0106] In some embodiments, a linear fusion approach can be used to correct the second simulated communication link data.

[0107] As one possible implementation, the second simulated link communication data is projected onto a two-dimensional flight trajectory at each evaluation point in the region to be merged, resulting in two-dimensional data points on the flight trajectory corresponding to the evaluation points. Based on the link quality data of the second real link communication data at the evaluation points, a weighted average method is used to adjust the link quality data of the second simulated link communication data at each evaluation point in the region to be merged, i.e., to perform local correction.

[0108] For example, the following formula (2) can be used for local correction: (2) in, Spatial position coordinates are The link quality data after the evaluation points are corrected. Represents spatial location coordinates The corresponding two-dimensional data points are real link quality data. The spatial coordinates in the second simulated communication link data are: The predicted link quality data for the evaluation points, Spatial position coordinates are The credibility weight of the evaluation points.

[0109] As another possible approach, for trends that require overall correction, the regional weight fusion method can be used for correction.

[0110] For example, the correction using the regional weight fusion method includes: determining the offset based on the error between the second simulated communication link data and the second real communication link data at the evaluation points in the region to be fused, and adjusting the predicted link quality data of each evaluation point in the region to be fused in the second simulated communication link data by weighting the offset and the confidence weight, so as to obtain the corrected communication link data.

[0111] For example, the offset can be obtained by minimizing the error at all evaluation points along the flight path.

[0112] For example, for each evaluation point within the region to be integrated, the following formula (3) can be used for weighted translation adjustment: (3) in, This indicates the offset.

[0113] For example, in a scenario involving accurate terrain modeling (such as simulation correction of base station coverage in mountainous areas), the measured drone trajectory provides accurate ground feature information, and the overall fit score for areas with large terrain undulations is high. Lower (e.g., rating <40%), within this range for all Perform uniform offset adjustment For example, -2dB or 1.5dB, while in areas with flat terrain and high similarity, the original simulation value is maintained or fine-tuned with a lower weight. Here, the overall similarity score of the region can be the spatial distribution similarity within the region.

[0114] Therefore, this weighted translation adjustment method can more naturally correct the deviation of the communication simulation model, while keeping the simulation output of other regions unchanged.

[0115] In this embodiment, the credibility weight of each evaluation point in the region to be fused is determined based on the difference information. Based on the credibility weight of each evaluation point and the real communication link data, the number of simulated communication links is corrected, which can improve the accuracy and reliability of data fusion.

[0116] In some embodiments, a Gaussian process field fusion method can also be used for correction. For example, the actual link quality data of the evaluation points in the fusion region of the second real communication link data can be used as the mean function of the Gaussian process prior, and a composite kernel function can be determined. Based on the composite kernel function, the second simulated communication link data, and the second real communication link data, a total covariance matrix and a noise diagonal matrix are constructed. Then, based on the total covariance matrix and the noise diagonal matrix, the predicted link quality data of each evaluation point in the fusion region of the second simulated communication link data are corrected to obtain the corrected communication link data.

[0117] Among them, the composite kernel function can be used to describe the signal correlation between any two points in space.

[0118] For example, composite kernel functions It is composed of the following sub-kernels added together: the first radial basis function kernel, the second radial basis function kernel, and the white noise kernel.

[0119] For example, the length scale of the first radial basis function kernel The range can be set to 10-50 meters, based on the fact that simulation data usually reflects trends such as large-scale path loss and macro-terrain shading, so a larger correlation radius is needed to capture this global smoothing characteristic.

[0120] For example, the length scale of the first radial basis function kernel The range can be set to 1 to 5 meters. This is because UAV flight test data is used to capture fine features such as small-scale fading and local multipath effects, so a smaller correlation radius is needed to characterize such rapid local changes.

[0121] For example, white noise kernels can be used to characterize random noise during UAV measurement processes.

[0122] This composite kernel function can automatically distinguish and fuse global trends from simulation data with local details from measured data. Based on a Bayesian inference framework, the aforementioned prior and measured data are fused to calculate the posterior Gaussian process for the entire space. For any point (x, y, z) to be predicted in the space, its corrected signal strength... The mean of the posterior distribution .

[0123] For example, it can be based on a composite kernel function. The total covariance matrix was calculated over all data points (including simulation data points and UAV test data points). For example, the real link quality data and simulated link quality data of each evaluation point in the region to be merged can be combined to obtain a merged data sequence. For any two data points in the merged data sequence, the covariance of any two data points can be calculated according to the composite function, and the total covariance matrix can be constructed based on the covariance.

[0124] For example, each data point in the merged data sequence is traversed, and a noise error is assigned to the data points of the UAV and the data points of the communication simulation model. Based on the noise errors of these data points, a noise diagonal matrix is ​​constructed. Each diagonal element in the noise diagonal matrix represents the noise variance of the corresponding data point. The noise diagonal matrix has the same size as the total covariance matrix, for example, both being [missing information]. The matrix.

[0125] For example, a smaller noise variance is assigned to flight measurement data points (i.e., data points collected by drones). (For example The noise variance is on the order of the mean, representing confidence in its high accuracy; for simulation data points, a larger noise variance is assigned. (For example The mean magnitude represents a relatively low level of confidence in it, thus directly solving the problem that traditional methods, which place equal trust in all data, lead to fusion results biased towards unreliable sources. Afterwards, posterior inference and field reconstruction are performed using Bayesian updates. The specific calculation formula is shown in formula (4) below: (4) in, Indicates assessment points The covariance vector of all known data points (including simulation data points and flight test data points), Represents the noise diagonal matrix. This represents a vector composed of all flight measurement data points. This represents the prior mean of the corresponding points, i.e. .

[0126] In some embodiments of this application, after the initial dynamic weight correction is completed, in order to ensure that the correction result of the simulation model has generalizability, that is, to maintain accuracy for the real environment parts that did not participate in the initial training or whose path coverage is incomplete, and to avoid overfitting the UAV measured trajectory data, the corrected communication link data can be closed-loop verified.

[0127] For example, closed-loop verification can be performed as follows: the overall correction error between the corrected communication link data and the verification set can be determined, the original overall error between the second real communication link data and the verification set can be determined, and the overall error reduction rate can be determined based on the overall correction error and the original overall error. If the overall error reduction rate exceeds a preset threshold, the corrected three-dimensional communication data can be used as the final correction result. If the overall error reduction rate does not exceed the preset threshold, the second simulated communication link data can be corrected again until the overall error reduction rate exceeds the preset threshold, and the final correction result can be obtained.

[0128] For example, the validation set can be a subset of data obtained from the same UAV flight mission or data from another time period, ensuring its independence from the data used in the correction process. For instance, a portion of UAV test data from areas with complex variations such as non-overlapping spatial and temporal dimensions and no obstructions can be reserved as the validation set. The selection of specific spatiotemporal locations or regions needs to be designed based on the objectives of the simulation model correction. For example, data from valleys or other locations with typical terrain features, different time states, or new path points that are not yet covered can be selected.

[0129] For example, the overall error reduction rate can be used to measure the extent of improvement in a communication simulation model.

[0130] A preset threshold is defined, which specifies the maximum allowable error for a single correction point and serves as the criterion for judging local areas.

[0131] For example, a quadratic difference analysis can be used to determine the overall correction error between the corrected communication link data and the validation set. For example, the overall correction error can be the mean absolute error or a variant thereof (such as the root mean square error).

[0132] For example, the specific process for determining the overall correction error is as follows: The corrected communication link data and the validation set are registered in three-dimensional spatial coordinates. For each point corresponding to the evaluation point after registration, a new error value is calculated. For example, the root mean square error or a weighted mean absolute error can be recalculated. This analysis can then yield a new overall error index or distribution chart. This more objectively reflects the prediction accuracy of the modified simulation model in areas not directly affected by its local offset strategy.

[0133] For example, if the overall error reduction rate does not exceed the preset threshold, the second simulation communication link data is corrected again until the overall error reduction rate exceeds the preset threshold. This ensures that the trajectory of a specific UAV is effectively corrected while also improving the overall performance of the simulation.

[0134] As an example, calculation and The overall mean absolute error or root mean square error is denoted as . Calculate the simulation data D before correction. sim Validation set with drones The original overall error between them is denoted as Next, calculate the overall error reduction rate: If the overall error reduction rate exceeds the preset threshold If the correction strategy is successful, the communication simulation model is considered to have achieved acceptable accuracy and met the generalization requirement. For example, This means that at least 30% of the initial overall error has been reduced. At this point, the final corrected result is output. .

[0135] If the overall error reduction rate Then, the confidence weight allocation strategy is adjusted based on the new verification information. For example, the upper limit of the correction weight for low similarity regions can be lowered, or the upper limit for high error regions can be increased. The degree of correction.

[0136] Therefore, through the closed-loop verification and feedback mechanism, it can be ensured that the correction of the communication simulation model by the UAV test data is effective and robust, and finally output a high-precision correction result verified in the real environment, thereby improving the accuracy of the correction result.

[0137] In some embodiments of this application, after outputting the modified communication link data or the final correction result, in order to intuitively demonstrate the correction process and its effects, and to facilitate users' understanding of model improvement and evaluation of subsequent applications (such as path planning, link budget, etc.), a visual output can be provided.

[0138] In some embodiments, a holistic view of the corrected communication link data can be constructed, and the three-dimensional spatial distribution relationship between the original communication simulation model, the UAV measured trajectory, and the corrected simulation model can be calculated and visualized. For example, a holistic view of the corrected communication link data can be constructed by generating a three-dimensional communication parameter distribution map. The specific process is as follows: the final correction result can be... The communication parameter data points are converted into graphic elements in three-dimensional space, such as signal strength. Each (x,y,z) point can be represented as a small sphere or cube grid with color or transparency attributes. Then, all these graphic elements are arranged and rendered according to their corresponding spatiotemporal positions to form a point cloud or contour map in three-dimensional space, thus obtaining a three-dimensional communication parameter distribution map.

[0139] For example, by overlaying difference heatmaps, the three-dimensional spatial distribution relationship between the original communication simulation model, the measured trajectory of the UAV, and the corrected simulation model can be calculated and visualized. This allows for a direct visual representation of the differences before and after correction, facilitating user understanding and analysis of the improved simulation model's effectiveness.

[0140] Heatmap overlay refers to comparing and displaying multiple datasets using different visual symbols within a unified three-dimensional coordinate system. Specifically, it involves overlaying communication link data collected by drones (D...). real Considered as a special type of point sequence data, it mainly contains spatial location ( x,y,z The original simulation dataset D contains the original simulation dataset D and the corresponding communication parameter values, such as signal strength. sim Visualization, that is, displaying each (using light colors, dashed lines, or low transparency). x,y,z The initial simulation parameter values ​​at the point, such as RSRP or SINR. This provides the state baseline before correction and the final correction result. Visualization, that is, displaying in a dark, solid line or high-contrast manner.

[0141] For example, the parameter value differences between the second simulated communication link data and the corrected communication link data at each evaluation point can be determined, and a difference heatmap can be generated based on the parameter value differences at each evaluation point.

[0142] For example, the specific process of overlaying difference heatmaps can be as follows: For each spatial location point with coordinates (x, y, z), calculate... D sim and Parameter value difference at this point .here S These differences could be signal strength, latency, etc. Map the differences to color codes for visualization.

[0143] For example, the following heatmap coloring principles can be defined:

[0144] This embodiment employs pseudo-color encoding, such as a gradient scheme based on MATLAB jet colormap or a custom scheme that transitions from blue (representing positive differences) to red (representing negative differences) and then to yellow / green (representing minor corrections or no significant changes). Pseudo-color encoding effectively maps continuous changes to an intuitive color space, facilitating understanding and analysis. For example, in a complex urban environment, a heatmap can clearly show which areas have stronger or weaker signal strengths after correction.

[0145] To facilitate understanding of the scheme in this application, the following is combined with... Figure 4 Please provide an explanation. Figure 4 This is a schematic diagram illustrating the fusion process of simulation data and flight test data provided in the embodiments of this application.

[0146] like Figure 4 As shown, the process includes: (1) Simultaneous acquisition of dual-source data Simultaneous dual-source data acquisition includes simulating link data from an authorized communication simulation model and real-time acquiring of actual link data in a UAV measured data module by flying along a planned path using high-precision sensors. Implementation may include setting the target area cube size, simulation data temporal resolution, and UAV positioning accuracy. This simultaneous dual-source data acquisition ensures the comparability of simulation and measured data in terms of spatiotemporal reference.

[0147] (2) Spatiotemporal reference alignment Spatiotemporal reference alignment includes spatial coordinate system normalization and time axis synchronization compensation. Spatial coordinate system normalization converts the simulation output into a unified coordinate system (such as UTM or WGS84) and achieves error control. Time axis synchronization compensation uses interpolation algorithms to improve the resolution of the simulation data to match the sampling frequency of the UAV. This can solve the problem of asynchrony between the communication simulation model and the real environment data.

[0148] (3) Quantification of differential characteristics The method for quantifying difference features can include generating local error fields and assessing spatial distribution similarity. The local error field can be quantified using mean absolute error (MAE) within a three-dimensional spatial region. A buffer zone is constructed for gridded error calculation. A three-dimensional buffer zone is defined centered on the UAV's measured trajectory line. Within this region, the difference between the simulated predicted values ​​and the actual observed values ​​is calculated, generating a gridded error field map. Spatial distribution similarity assessment analyzes the degree of agreement between the overall trend of the simulated data and the measured data using trend comparison or statistical matching methods.

[0149] (4) Credibility weight allocation and non-intrusive data fusion This method incorporates linear fusion and Gaussian process field fusion, employing a non-intrusive approach to correct the simulation model. Confidence weights are determined based on MAE and similarity scores, with higher correction weights assigned to high-error regions to enhance the impact of measured data. The non-intrusive fusion methods include local weighted average shift and Gaussian process field fusion. The former adjusts the simulation output through weighted averaging, while the latter utilizes a composite kernel function to construct a prior mean and combines it with Bayesian inference to fuse global trends with local details.

[0150] (5) Closed-loop verification and feedback of the correction effect A closed-loop verification and feedback mechanism is established, including quadratic difference analysis and iterative termination decision. The correction effect is evaluated using an independent validation set. The generalization performance of the corrected simulation model in regions not included in the initial training is assessed, and the error reduction rate is calculated to determine if the expected accuracy has been achieved. If the condition is met, the final correction result is output; otherwise, the confidence weight allocation strategy is adjusted, and the correction process is iterated again. This ensures the generalization of the communication simulation model and avoids overfitting. The closed-loop verification and feedback mechanism for the correction effect is specifically for linear fusion methods.

[0151] (6) 3D visualization output This system includes the generation of 3D communication parameter distribution maps and the overlay of difference heatmaps, providing a visual representation of the changes in simulation data before and after correction. Heatmap technology can be used to overlay and compare the original simulation data, UAV measured data, and the corrected results. The corrected simulation data can be presented in voxel or contour format and compared with the original simulation data using color coding. Pseudo-color coding (such as a blue-to-red gradient) visually demonstrates the differences before and after correction, helping users understand the model improvement effect. The heatmap overlay shows the distribution and differences of key parameters such as signal strength and latency in 3D space, facilitating intuitive evaluation and analysis for subsequent applications (such as path planning and link budgeting).

[0152] In this embodiment, by employing techniques such as simultaneous acquisition of dual-source data, spatiotemporal benchmark alignment, differential feature quantification, credibility weight allocation and data fusion, and three-dimensional visualization output, the problems of lack of precise temporal and spatial correspondence between simulation data and real UAV measurement data, leading to difficulties in comparison and correction, the inability of a single data source to fully reflect the changing trends of communication parameters in complex three-dimensional space, and the potential for overall structural deviation and impact on prediction capabilities due to directly replacing the output values ​​of the simulation model are addressed.

[0153] The solution in this application achieves comprehensive fusion of flight test data and simulation data through simultaneous acquisition of dual-source data, overcoming the limitations of traditional solutions that rely on independent data use or indirect verification. Furthermore, spatiotemporal reference alignment technology ensures precise temporal and spatial correspondence of the data, successfully resolving the issues of data asynchrony and inaccurate comparison. Differential feature quantification and credibility weight allocation technologies enhance the intelligence of data fusion, dynamically adjusting weights to improve data accuracy and reliability, avoiding the impact of single data source bias on the overall results. Three-dimensional visualization output provides users with intuitive data display and interactive analysis capabilities, significantly enhancing their understanding and application of complex three-dimensional spatial communication parameters. Moreover, the solution in this application meets the urgent needs of the communication technology field for high precision, real-time performance, and reliability, demonstrating significant performance advantages in critical missions.

[0154] Figure 5 This is a schematic diagram of a device for correcting simulated communication link data provided in an embodiment of this application.

[0155] like Figure 5 As shown, the simulation communication link data correction device 500 includes: The acquisition module 510 is used to acquire the first simulated communication link data in the three-dimensional space of the target from the communication simulation model, and to acquire the first real communication link data on the flight trajectory synchronously acquired by the UAV; wherein the flight trajectory is located in the three-dimensional space of the target. Alignment module 520 is used to perform spatiotemporal reference alignment processing on the first simulated communication link data and the first real communication link data to obtain second simulated communication link data and second real communication link data that are aligned in time and space. The correction module 530 is used to correct the second simulated communication link data based on the second real communication link data to obtain the corrected communication link data.

[0156] Optionally, the correction module 530 is used for: The difference features of the second simulated communication link data and the second real communication link data are quantified to obtain the difference information. Based on the difference information and the second real communication link data, the second simulated communication link data is corrected to obtain the corrected communication link data.

[0157] Optionally, the correction module 530 is used for: Spatial registration is performed on the second simulated communication link data and the second real communication link data to determine the same spatial location points in the second simulated communication link data and the second real communication link data, which are used as evaluation points; By expanding the space around key sampling points on the flight trajectory, a three-dimensional buffer zone for the target is obtained. Based on the second simulated communication link data and the second real communication link data, the link quality data of the evaluation points in the target three-dimensional buffer are used to determine the mean absolute error of the key sampling points. Based on the mean absolute error, obtain the difference information.

[0158] Optionally, the correction module 530 is used for: Extract the predicted link quality data sequence of the communication simulation model on the flight trajectory from the second simulated communication link data, and extract the real link quality data sequence from the second real communication link data; Spatial distribution similarity is evaluated between the predicted link quality data sequence and the actual link quality data sequence to obtain spatial distribution similarity; whereby spatial distribution similarity reflects the degree to which the overall trend of the predicted link quality data sequence matches that of the actual link quality data sequence. Difference information is obtained based on mean absolute error and spatial distribution similarity.

[0159] Optionally, the correction module 530 is used for: Based on the difference information, the credibility weight of each evaluation point in the region to be merged is determined; where the evaluation point is the spatial location point shared by the second simulated communication link data and the second real communication link data; Based on the error between the second simulated communication link data and the second real communication link data at the evaluation points in the area to be fused, the offset is determined. Based on the offset and confidence weight, the predicted link quality data of each evaluation point in the fusion area of ​​the second simulated communication link number is weighted and offset adjusted to obtain the corrected communication link data.

[0160] Optionally, the correction module 530 is used for: Determine the overall correction error between the corrected communication link data and the verification set; Determine the original overall error between the second real communication link data and the verification set; The overall error reduction rate is determined based on the overall correction error and the original overall error. If the overall error reduction rate does not exceed the preset threshold, the second simulated communication link data is corrected again until the overall error reduction rate exceeds the preset threshold.

[0161] Optionally, the correction module 530 is used for: Determine the composite kernel function; whereby the composite kernel function is used to describe the signal correlation between any two points in space; Based on the composite kernel function, the second simulated communication link data, and the second real communication link data, construct the total covariance matrix and the noise diagonal matrix; Based on the total covariance matrix and the noise diagonal matrix, the second simulated communication link data is corrected to obtain the corrected communication link data.

[0162] Optionally, the alignment module 520 is used for: The spatial location information in the first simulated communication link data and the spatial location information in the first real communication link data are mapped to the same spatial coordinate system to obtain the third simulated communication link data and the third real communication link data. The timestamps in the third simulated communication link data and the third real communication link data are converted to the same time reference system to obtain the fourth simulated communication link data and the second real communication link data. The fourth simulated communication link data is interpolated to obtain the second simulated communication link data; wherein the resolution of the second simulated communication link data is matched with the sampling frequency of the UAV.

[0163] Optionally, the device may further include: The determination module is used to determine the parameter value differences between the second simulated communication link data and the corrected communication link data at each evaluation point; wherein, the evaluation point is a spatial location point shared by the second simulated communication link data and the second real communication link data; The generation and display module is used to generate and display a difference heatmap based on the differences in parameter values ​​at each evaluation point.

[0164] It should be noted that the explanation of the aforementioned method embodiment for correcting simulated communication link data also applies to the device for correcting simulated communication link data in this embodiment, and will not be repeated here.

[0165] In this embodiment, by synchronously collecting the first simulated communication link data output by the communication simulation model and the first real communication link data obtained by the UAV on the flight trajectory in the target three-dimensional space, it can be ensured that the simulation and measured data are comparable in spatiotemporal reference. Furthermore, by aligning the two data in spatiotemporal reference, it is ensured that the simulation and measured data are strictly comparable in a unified time coordinate and spatial coordinate system. Based on this, the second simulated communication link data is specifically corrected using the aligned second real communication link data as a calibration benchmark, which can improve the fidelity and prediction accuracy of the simulation results in complex airspace scenarios.

[0166] To implement the above embodiments, this application also proposes an electronic device, including: a processor and a memory communicatively connected to the processor; the memory stores computer execution instructions; the processor executes the computer execution instructions stored in the memory to implement the method provided in the foregoing embodiments. To implement the above embodiments, this application also proposes a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the methods provided in the foregoing embodiments.

[0167] To implement the above embodiments, this application also proposes a computer program product, including a computer program that, when executed by a processor, implements the methods provided in the foregoing embodiments.

[0168] The collection, storage, use, processing, transmission, provision, and disclosure of user personal information involved in this application all comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0169] It should be noted that personal information collected from users should be used for legitimate and reasonable purposes and should not be shared or sold outside of these legitimate uses. Furthermore, such collection / sharing should only be conducted after receiving the user's informed consent, including but not limited to notifying the user to read the user agreement / user notice and sign an agreement / authorization that includes authorization of relevant user information before the user uses the function. In addition, any necessary steps must be taken to protect and safeguard access to such personal information data and ensure that others with access to personal information data comply with their privacy policies and procedures.

[0170] This application is intended to provide an implementation scheme for users to selectively prevent the use or access to their personal information data. Specifically, this disclosure is intended to provide hardware and / or software to prevent or block access to such personal information data. Once personal information data is no longer needed, risks can be minimized by restricting data collection and deleting data. Furthermore, where applicable, such personal information is de-identified to protect user privacy.

[0171] In the foregoing descriptions of the embodiments, the terms "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0172] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0173] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0174] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0175] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0176] Those skilled in the art will understand that all or part of the steps of the methods described in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it includes one or a combination of the steps of the method embodiments.

[0177] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0178] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of this application.

Claims

1. A method for correcting simulated communication link data, characterized in that, include: The first simulated communication link data within the target's three-dimensional space is obtained from the communication simulation model, and the first real communication link data on the flight trajectory synchronously collected by the UAV is also obtained; wherein, the flight trajectory is located within the target's three-dimensional space; The first simulated communication link data and the first real communication link data are aligned with a spatiotemporal reference to obtain second simulated communication link data and second real communication link data that are aligned in time and space. Based on the second real communication link data, the second simulated communication link data is corrected to obtain the corrected communication link data.

2. The method as described in claim 1, characterized in that, The step of correcting the second simulated communication link data based on the second real communication link data to obtain corrected communication link data includes: The difference features of the second simulated communication link data and the second real communication link data are quantized to obtain difference information; Based on the difference information and the second real communication link data, the second simulated communication link data is corrected to obtain the corrected communication link data.

3. The method as described in claim 2, characterized in that, The step of quantifying the difference features of the second simulated communication link data and the second real communication link data to obtain difference information includes: Spatial registration is performed on the second simulated communication link data and the second real communication link data to determine the same spatial location points in the second simulated communication link data and the second real communication link data, which are used as evaluation points; By expanding the space around the key sampling points on the flight trajectory, a three-dimensional buffer zone of the target is obtained. Based on the second simulated communication link data and the second real communication link data, the mean absolute error of the key sampling points is determined using the link quality data of the evaluation points in the target three-dimensional buffer. The difference information is obtained based on the mean absolute error.

4. The method as described in claim 3, characterized in that, The step of quantifying the difference features of the second simulated communication link data and the second real communication link data to obtain difference information further includes: Extract the predicted link quality data sequence of the communication simulation model on the flight trajectory from the second simulated communication link data, and extract the real link quality data sequence from the second real communication link data; Spatial distribution similarity is evaluated between the predicted link quality data sequence and the actual link quality data sequence to obtain spatial distribution similarity; wherein, the spatial distribution similarity reflects the degree to which the overall trend of the predicted link quality data sequence matches that of the actual link quality data sequence. The difference information is obtained based on the mean absolute error and the spatial distribution similarity.

5. The method as described in claim 2, characterized in that, The step of correcting the second simulated communication link data based on the difference information and the second real communication link data to obtain the corrected communication link data includes: Based on the difference information, the credibility weight of each evaluation point in the region to be fused is determined; wherein, the evaluation point is a spatial location point shared by the second simulated communication link data and the second real communication link data; Based on the error between the second simulated communication link data and the second real communication link data at the evaluation point in the area to be fused, the offset is determined; Based on the offset and the confidence weight, the predicted link quality data of each evaluation point in the fusion region of the second simulated communication link number is weighted and offset adjusted to obtain the corrected communication link data.

6. The method as described in claim 2, characterized in that, The step of correcting the second simulated communication link data based on the difference information and the second real communication link data to obtain the corrected communication link data includes: Determine the overall correction error between the corrected communication link data and the verification set; Determine the original overall error between the second real communication link data and the verification set; The overall error reduction rate is determined based on the overall correction error and the original overall error. In response to the overall error reduction rate not exceeding a preset threshold, the second simulated communication link data is re-corrected until the overall error reduction rate exceeds the preset threshold.

7. The method as described in claim 1, characterized in that, The step of correcting the second simulated communication link data based on the second real communication link data to obtain the corrected communication link data includes: Determine the composite kernel function; wherein the composite kernel function is used to describe the signal correlation between any two points in space; Based on the composite kernel function, the second simulated communication link data, and the second real communication link data, a total covariance matrix is ​​constructed, and a noise diagonal matrix is ​​also constructed. Based on the total covariance matrix and the noise diagonal matrix, the second simulated communication link data is corrected to obtain the corrected communication link data.

8. The method as described in claim 1, characterized in that, The step of performing spatiotemporal reference alignment processing on the first simulated communication link data and the first real communication link data to obtain the second simulated communication link data and the second real communication link data includes: The spatial location information in the first simulated communication link data and the spatial location information in the first real communication link data are mapped to the same spatial coordinate system to obtain the third simulated communication link data and the third real communication link data. The timestamps in the third simulated communication link data and the third real communication link data are converted to the same time reference system to obtain the fourth simulated communication link data and the second real communication link data; The fourth simulated communication link data is interpolated to obtain the second simulated communication link data; wherein the resolution of the second simulated communication link data matches the sampling frequency of the UAV.

9. The method according to any one of claims 1-8, characterized in that, The method further includes: Determine the parameter value differences between the second simulated communication link data and the corrected communication link data at each evaluation point; wherein, the evaluation point is a spatial location point common to the second simulated communication link data and the second real communication link data; Based on the differences in parameter values ​​at each evaluation point, a difference heatmap is generated and displayed.

10. A device for correcting simulated communication link data, characterized in that, include: The acquisition module is used to acquire first simulated communication link data in the target three-dimensional space from the communication simulation model, and to acquire first real communication link data on the flight trajectory synchronously acquired by the UAV; wherein the flight trajectory is located in the target three-dimensional space; The alignment module is used to perform spatiotemporal reference alignment processing on the first simulated communication link data and the first real communication link data to obtain second simulated communication link data and second real communication link data that are aligned in time and space. The correction module is used to correct the second simulated communication link data based on the second real communication link data to obtain the corrected communication link data.

Images