A method and system for dynamic evaluation of three-dimensional airspace communication holes based on unmanned aerial vehicles

Abstract

This invention relates to a method and system for dynamic assessment of communication voids in three-dimensional airspace based on unmanned aerial vehicles (UAVs), belonging to the field of UAV intelligent control technology. It aims to solve the problem of accurate assessment of communication voids in three-dimensional airspace. The invention collects collaborative data from multiple UAVs to obtain raw communication signaling data from multiple networks; extracts multi-dimensional quality indicator features; filters the raw communication signaling data and multi-dimensional quality indicator features according to different communication modes and then performs normalization processing to obtain local area network (LAN) communication datasets, wide area network (WAN) communication datasets, and satellite communication datasets; uses a density-based spatiotemporal clustering method to identify communication voids and obtain the spatiotemporal distribution results of communication voids in the three-dimensional airspace; establishes a communication quality classification and assessment model using a fusion method of random forest and SVM to obtain the communication quality classification and assessment results; and generates a multi-dimensional spatial resolution heatmap of communication voids to complete the dynamic assessment of communication voids in three-dimensional airspace based on UAVs. This invention can accurately identify weak communication regions.

Description

Technical Field

[0001] This invention belongs to the field of unmanned aerial vehicle (UAV) intelligent control technology, specifically relating to a method and system for dynamic evaluation of three-dimensional airspace communication voids based on UAVs. Background Technology

[0002] Against the backdrop of strong economic development in low-altitude regions, the operation of drone swarms is an inevitable trend in the future. Currently, drone communication mainly employs the following methods: 1) local area network communication with the controller; 2) wide area internet communication based on the four major telecom operators (China Mobile, China Unicom, China Telecom, and China Broadcasting Network); and 3) satellite network communication. These three communication methods exhibit different coverage characteristics and communication quality performance in the three-dimensional airspace, necessitating the establishment of a unified spatial resolution evaluation system. However, existing technologies face the following key challenges in evaluating communication holes in the three-dimensional airspace: 1. Lack of spatial resolution standards: Existing methods are mostly two-dimensional planar evaluations, failing to reflect the true communication quality distribution in the three-dimensional airspace; 2. Insufficient collaborative analysis of multiple communication modes: It is difficult to simultaneously evaluate the performance differences of 2G / 3G / 4G / 5G and satellite communication under different airspace conditions; 3. Lack of real-time dynamic evaluation capabilities: Traditional static measurements cannot adapt to changes in the communication environment during drone flight; 4. Insufficient accuracy in identifying communication holes: There is a lack of intelligent hole detection mechanisms based on machine learning algorithms.

[0003] Currently, traditional communication signal assessment methods mainly rely on ground base station measurements, which have the following technical shortcomings: 1. Insufficient spatial coverage: Ground measurements cannot fully reflect the communication quality distribution in the three-dimensional airspace, especially the inability to accurately identify communication holes in the vertical direction; 2. Poor real-time performance: Static measurements cannot adapt to dynamically changing communication environments and cannot meet the real-time flight decision-making needs of UAVs; 3. Lack of multi-network coordination: There is a lack of comprehensive assessment of mobile, China Unicom, China Telecom networks, and satellite communications, making it impossible to provide a comprehensive communication assurance solution; 4. Weak scene adaptability: Existing methods are difficult to apply simultaneously to various complex environments such as urban high-rise buildings, mountainous terrain, and emergency communications. During flight missions, UAVs frequently encounter communication interruptions, signal attenuation, and other communication hole problems, which seriously affect flight safety and mission execution efficiency. Therefore, there is an urgent need for a technical method that can achieve accurate assessment of communication holes in the three-dimensional airspace. Summary of the Invention

[0004] The problem to be solved by this invention is to achieve accurate assessment of three-dimensional airspace communication voids, and to propose a dynamic assessment method and system for three-dimensional airspace communication voids based on UAVs.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A method for dynamic evaluation of three-dimensional airspace communication voids based on unmanned aerial vehicles (UAVs) includes the following steps:

[0007] S1. Collect collaborative data from multiple UAVs. Utilize the built-in communication unit and dedicated airborne signal acquisition unit of the UAVs to collect communication signaling data from 2G, 3G, 4G, 5G and satellite communications in real time to obtain raw communication signaling data from multiple networks.

[0008] S2. Signaling multidimensional feature extraction: Extract multidimensional quality indicator features from the original multi-network communication signaling data obtained in step S1;

[0009] S3. Spatiotemporal joint analysis and clustering: Based on the original multi-network communication signaling data obtained in step S1 and the multi-dimensional quality index features obtained in step S2, the data are filtered according to different communication modes and then normalized to obtain local area network communication datasets, wide area network communication datasets, and satellite communication datasets. Then, a density-based spatiotemporal clustering method is used to identify communication voids and obtain the spatiotemporal distribution results of communication voids in the three-dimensional spatial domain.

[0010] S4. Based on the original multi-network communication signaling data obtained in step S1, the multi-dimensional quality index features obtained in step S2, and the spatiotemporal distribution results of communication holes in the three-dimensional spatial domain obtained in step S3, a communication quality classification and evaluation model is established using the random forest and SVM fusion method to obtain the communication quality classification and evaluation results.

[0011] S5. Based on the communication quality classification and evaluation results obtained in step S4, generate a multi-dimensional spatial resolution communication hole heatmap to complete the dynamic evaluation of three-dimensional airspace communication holes based on UAVs.

[0012] Furthermore, in step S1, the key parameters for data sampling are set, including sampling frequency, spatial resolution, and time window. The sampling frequency is 1-10Hz, the spatial resolution is the standard spatial resolution of 50m×50m×50m, and the time window size is 5-30 minutes. The collected data includes longitude, latitude, altitude, timestamp, signal strength index, signal-to-noise ratio, bit error rate, delay jitter, packet loss rate, spectral efficiency, and link quality indicator.

[0013] Furthermore, the multi-dimensional quality indicator features mentioned in step S2 include signal strength indicators, signal-to-noise ratio, bit error rate, latency jitter, packet loss rate, spectral efficiency, and link quality indicators. The signaling multi-dimensional feature extraction process includes the following steps:

[0014] S2.1. Temporal feature extraction: Calculate the mean, standard deviation, skewness, and kurtosis of the signal intensity;

[0015] S2.2. Calculate the signal-to-noise ratio based on the logarithmic ratio of signal power and noise power;

[0016] S2.3. Frequency Domain Feature Extraction: Calculate the spectral centroid, spectral bandwidth, spectral flatness, spectral roll-off point, spectral entropy, spectral kurtosis, and spectral skewness;

[0017] S2.4. Statistical Feature Extraction: Calculate the skewness and kurtosis statistics of the signal distribution.

[0018] Furthermore, the specific implementation method of step S3 includes the following steps:

[0019] S3.1. Prepare clustering data: including longitude, latitude, altitude, timestamp, signal strength index, signal-to-noise ratio, bit error rate, latency jitter, packet loss rate, spectral efficiency, and link quality indicator;

[0020] S3.2. Filter the clustering data prepared in step S3.1 according to communication mode, and then use the Z-score normalization method to normalize it to obtain local area network communication dataset, wide area network communication dataset, and satellite communication dataset;

[0021] S3.3. Apply the density-based spatiotemporal clustering algorithm ST-DBSCAN to identify communication gaps;

[0022] The core distance metric formula of the ST-DBSCAN spatiotemporal clustering algorithm is:

[0023]

[0024] Where d(x_i, x_j) is the spatiotemporal distance between the i-th data point and the j-th data point, x_i, y_i, z_i are the longitude, latitude, and altitude coordinates of the i-th data point, respectively, x_j, y_j, z_j are the longitude, latitude, and altitude coordinates of the j-th data point, t_i is the timestamp of the i-th data point, and t_j is the timestamp of the j-th data point; Let d(x_i, x_j) ≤ ε and |t_i - t_j| ≤ δ_t, then the i-th data point and the j-th data point are considered to be spatially adjacent, where ε is the spatial distance threshold and δ_t is the time difference threshold.

[0025] S3.4. Analyze the clustering results obtained in step S3.3, and identify data points with a label of -1 as communication holes.

[0026] Furthermore, in step S4, a random forest classifier is created, with parameters including 100 decision trees and a maximum depth of 10 layers. The spatiotemporal distribution results of communication holes in the three-dimensional spatial domain obtained in step S3 are used as training data. The random forest classification algorithm is used to intelligently classify and evaluate communication quality, and a communication quality classification model adapted to different environments is established. Step S3 provides labeled data and feature enhancement for classification evaluation. The labeled data and feature enhancement constitute a collaboratively optimized evaluation system. Finally, the communication quality classification level is defined into five levels according to the random forest classifier, including excellent, good, average, poor, and very poor.

[0027] Furthermore, the specific implementation method of step S5 includes the following steps:

[0028] S5.1. Spatial Mesh Generation: Create a three-dimensional mesh based on a spatial resolution of 50 meters × 50 meters × 50 meters;

[0029] S5.2. Data Point Mapping: Mapping communication data points to corresponding spatial resolution cells. The formula for dividing the spatial resolution cells is as follows:

[0030] x_bins = [x_min, x_min + Δx, x_min + 2Δx, ..., x_max]

[0031] y_bins = [y_min, y_min + Δy, y_min + 2Δy, ..., y_max]

[0032] z_bins = [z_min, z_min + Δz, z_min + 2Δz, ..., z_max]

[0033] Where x_bins is the sequence of dividing points in the longitude direction, x_min and x_max are the minimum and maximum values ​​in the longitude range, respectively, Δx is the spatial resolution in the longitude direction, y_bins is the sequence of dividing points in the latitude direction, y_min and y_max are the minimum and maximum values ​​in the latitude range, respectively, Δy is the spatial resolution in the latitude direction, z_bins is the sequence of dividing points in the altitude direction, z_min and z_max are the minimum and maximum values ​​in the altitude range, respectively, Δz is the spatial resolution in the altitude direction;

[0034] S5.3. Communication Quality Calculation: Calculate the average communication quality score within each spatial resolution cell. The calculation formula is as follows:

[0035] Q(i,j,k) = (1 / N_ijk) × ∑ w_n × q_n

[0036] Where Q(i,j,k) is the communication quality score of the (i,j,k)th spatial unit, N_ijk is the number of data points in the spatial unit, w_n is the weight coefficient of the nth data point, and q_n is the communication quality score of the nth data point;

[0037] Then, communication void identification is performed, using the following formula:

[0038] Void(i,j,k) = { 1, if Q(i,j,k) < Q_void or N_ijk = 0, otherwise}

[0039] Where Void(i,j,k) indicates whether the (i,j,k)-th spatial unit is a communication hole, 1 indicates a hole and 0 indicates no hole, and Q_void is the communication hole threshold, which is set to 20;

[0040] S5.4. Color Mapping: Use red, yellow, and green color gradients to represent the strength of communication quality;

[0041] S5.5. 3D Visualization: Generate a heatmap of communication voids with multi-dimensional spatial resolution;

[0042] S5.6. Multi-mode support: Generates independent multi-dimensional spatial resolution communication hole heatmaps for local area network communication, wide area network communication, and satellite communication, and completes dynamic assessment of three-dimensional airspace communication holes based on UAVs.

[0043] A dynamic assessment system for three-dimensional airspace communication voids based on unmanned aerial vehicles (UAVs) is implemented based on the aforementioned dynamic assessment method for three-dimensional airspace communication voids based on UAVs. The system includes a data acquisition module, a feature extraction module, a cluster analysis module, a classification assessment module, a heat map generation module, and a visualization module, which are connected sequentially.

[0044] The data acquisition module is responsible for collaborative data acquisition among multiple UAVs, including a built-in communication unit and an aerial signal acquisition unit; the feature extraction module is responsible for extracting multi-dimensional quality indicators from the raw communication signaling; the clustering analysis module is responsible for spatiotemporal joint clustering analysis using the ST-DBSCAN algorithm; the classification evaluation module is responsible for intelligent classification of communication quality based on the random forest algorithm; the heatmap generation module is responsible for generating a three-dimensional airspace communication hole heatmap; and the visualization display module is responsible for providing an interactive three-dimensional visualization interface.

[0045] The beneficial effects of this invention are:

[0046] This invention discloses a dynamic assessment method for three-dimensional airspace communication voids based on unmanned aerial vehicles (UAVs). Through real-time communication void assessment, it provides optimal flight paths for UAVs, avoiding communication blind spots. This invention can accurately identify areas of weak communication, providing data support for base station deployment and signal enhancement. Simultaneously, it assesses the communication quality of China Mobile, China Unicom, China Telecom, and satellite networks, achieving network collaborative optimization. This invention provides communication reliability assurance for low-altitude flight, promoting the development of the low-altitude economy. It fully considers the special characteristics of complex environments such as cities and mountainous areas, exhibiting good adaptability. Attached Figure Description

[0047] Figure 1 This is a flowchart of a three-dimensional airspace communication hole dynamic evaluation method based on UAVs, as described in this invention.

[0048] Figure 2 This is a schematic diagram of the three-dimensional spatial resolution division of the present invention;

[0049] Figure 3 This is a schematic diagram illustrating the training of the communication quality classification model of the present invention. Detailed Implementation

[0050] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only for explaining the invention and are not intended to limit the invention; that is, the described specific embodiments are merely a part of the embodiments of the invention, and not all of them. The components of the specific embodiments of the invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations, and the invention may also have other embodiments.

[0051] Therefore, the following detailed description of specific embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected specific embodiments of the invention. All other specific embodiments obtained by those skilled in the art based on these specific embodiments without inventive effort are within the scope of protection of this invention.

[0052] To further understand the invention's content, features, and effects, the following specific embodiments are provided, along with accompanying drawings. Figure 1 - Appendix Figure 3 Detailed explanation is as follows:

[0053] Example 1:

[0054] A method for dynamic evaluation of three-dimensional airspace communication voids based on unmanned aerial vehicles (UAVs) includes the following steps:

[0055] S1. Collect collaborative data from multiple UAVs. Utilize the built-in communication unit and dedicated airborne signal acquisition unit of the UAVs to collect communication signaling data from 2G, 3G, 4G, 5G and satellite communications in real time to obtain raw communication signaling data from multiple networks.

[0056] Furthermore, in step S1, the key parameters for data sampling are set, including sampling frequency, spatial resolution, and time window. The sampling frequency is 1-10Hz, the spatial resolution is the standard spatial resolution of 50m×50m×50m, and the time window size is 5-30 minutes. The collected data includes longitude, latitude, altitude, timestamp, signal strength index, signal-to-noise ratio, bit error rate, delay jitter, packet loss rate, spectral efficiency, and link quality indicator.

[0057] Furthermore, the drone cluster is initialized and the communication unit is activated. By connecting to the drone cluster, the unique identifier and communication status of each drone are verified.

[0058] The initialization process of an unmanned aerial vehicle (UAV) communication system includes the following steps:

[0059] 1. Create a drone cluster management object to maintain the drone list and communication unit configuration;

[0060] 2. Configure supported communication network types: 2G, 3G, 4G, 5G, satellite communication;

[0061] 3. Configure communication modes: LAN communication, WAN communication, satellite communication;

[0062] 4. Set the data sampling frequency to 1Hz, which can be dynamically adjusted according to flight speed and communication mode;

[0063] 5. Activate the built-in communication unit and airborne signal acquisition unit of each drone;

[0064] 6. Enable multi-network monitoring to monitor all configured network types simultaneously.

[0065] Furthermore, the communication modes simultaneously support LAN communication, WAN communication, and satellite communication.

[0066] S2. Signaling multidimensional feature extraction: Extract multidimensional quality indicator features from the original multi-network communication signaling data obtained in step S1;

[0067] Furthermore, the multi-dimensional quality indicator features mentioned in step S2 include signal strength indicators, signal-to-noise ratio, bit error rate, latency jitter, packet loss rate, spectral efficiency, and link quality indicators. The signaling multi-dimensional feature extraction process includes the following steps:

[0068] S2.1. Temporal feature extraction: Calculate the mean, standard deviation, skewness, and kurtosis of the signal intensity;

[0069] S2.2. Calculate the signal-to-noise ratio based on the logarithmic ratio of signal power and noise power;

[0070] S2.3. Frequency Domain Feature Extraction: Calculate the spectral centroid, spectral bandwidth, spectral flatness, spectral roll-off point, spectral entropy, spectral kurtosis, and spectral skewness;

[0071] S2.4. Statistical Feature Extraction: Calculate the skewness and kurtosis statistics of the signal distribution.

[0072] Furthermore, communication quality assessment involves several key algorithmic formulas:

[0073] 1. Signal-to-noise ratio calculation formula:

[0074]

[0075] in: For signal-to-noise ratio, For signal power, Noise power;

[0076] 2. Bit error rate calculation formula:

[0077] BER =

[0078] Where: BER is the bit error rate. Number of error bits This represents the total number of bits.

[0079] 3. Latency jitter calculation formula:

[0080] ;

[0081] in, For time delay jitter, Let be the transmission delay of the i-th data packet. For average delay, Total number of data packets It is the square root function;

[0082] 4. Mean signal strength:

[0083]

[0084] in, The average signal strength Let N be the signal strength value of the i-th sampling point, and N be the total number of sampling points.

[0085] 5. Signal strength standard deviation

[0086]

[0087] in, The standard deviation of signal strength The average signal strength;

[0088] 6. Signal skewness

[0089]

[0090] in, Signal skewness;

[0091] 7. Signal kurtosis

[0092]

[0093] in, For signal kurtosis;

[0094] S3. Spatiotemporal joint analysis and clustering: Based on the original multi-network communication signaling data obtained in step S1 and the multi-dimensional quality index features obtained in step S2, the data are filtered according to different communication modes and then normalized to obtain local area network communication datasets, wide area network communication datasets, and satellite communication datasets. Then, a density-based spatiotemporal clustering method is used to identify communication voids and obtain the spatiotemporal distribution results of communication voids in the three-dimensional spatial domain.

[0095] Furthermore, the specific implementation method of step S3 includes the following steps:

[0096] S3.1. Prepare clustering data: including longitude, latitude, altitude, timestamp, signal strength index, signal-to-noise ratio, bit error rate, latency jitter, packet loss rate, spectral efficiency, and link quality indicator;

[0097] S3.2. Filter the clustering data prepared in step S3.1 according to communication mode, and then use the Z-score normalization method to normalize it to obtain local area network communication dataset, wide area network communication dataset, and satellite communication dataset;

[0098] Furthermore, the formula for data standardization is:

[0099]

[0100] in: These are the standardized eigenvalues. These are the original eigenvalues. The characteristic mean, The characteristic standard deviation;

[0101] S3.3. Apply the density-based spatiotemporal clustering algorithm ST-DBSCAN to identify communication gaps;

[0102] The core distance metric formula of the ST-DBSCAN spatiotemporal clustering algorithm is:

[0103]

[0104] Where d(x_i, x_j) is the spatiotemporal distance between the i-th data point and the j-th data point, x_i, y_i, z_i are the longitude, latitude, and altitude coordinates of the i-th data point, respectively, x_j, y_j, z_j are the longitude, latitude, and altitude coordinates of the j-th data point, t_i is the timestamp of the i-th data point, and t_j is the timestamp of the j-th data point; Let d(x_i, x_j) ≤ ε and |t_i - t_j| ≤ δ_t, then the i-th data point and the j-th data point are considered to be spatially adjacent, where ε is the spatial distance threshold and δ_t is the time difference threshold.

[0105] S3.4. Analyze the clustering results obtained in step S3.3, and identify data points with a label of -1 as communication holes.

[0106] S4. Based on the original multi-network communication signaling data obtained in step S1, the multi-dimensional quality index features obtained in step S2, and the spatiotemporal distribution results of communication holes in the three-dimensional spatial domain obtained in step S3, a communication quality classification and evaluation model is established using the random forest and SVM fusion method to obtain the communication quality classification and evaluation results.

[0107] Furthermore, in step S4, a random forest classifier is created, with parameters including 100 decision trees and a maximum depth of 10 layers. The spatiotemporal distribution results of communication holes in the three-dimensional spatial domain obtained in step S3 are used as training data. The random forest classification algorithm is used to intelligently classify and evaluate communication quality, and a communication quality classification model adapted to different environments is established. Step S3 provides labeled data and feature enhancement for classification evaluation. The labeled data and feature enhancement constitute a collaboratively optimized evaluation system. Finally, the communication quality classification level is defined into five levels according to the random forest classifier, including excellent, good, average, poor, and very poor.

[0108] Furthermore, the communication quality classification and assessment process includes the following steps:

[0109] Step 1. Model initialization: Create a random forest classifier, setting parameters including 100 decision trees and a maximum depth of 10 layers;

[0110] Step 2. Data partitioning: Divide the feature data into 80% training set and 20% test set;

[0111] Step 3. Model Training: Train a random forest classifier using the training set data;

[0112] Step 4. Model Evaluation: Evaluate the model performance on the test set and output a classification report;

[0113] Step 5. Quality Prediction: Use the trained model to predict the communication quality level of the new data;

[0114] Step 6. Quality Level Definition: Communication quality is divided into five levels: Excellent, Good, Average, Poor, and Very Poor.

[0115] The core formula of the random forest classification algorithm is as follows:

[0116] 1. The expression for constructing a decision tree based on Gini impurity is as follows:

[0117] Gini(D) = 1 - ∑ p_k²

[0118] Where: Gini(D) is the Gini impurity of dataset D, and p_k is the proportion of the k-th class of samples in dataset D;

[0119] 2. The expression for information gain (used for feature selection) is:

[0120] Gain(D, A) = Entropy(D) - ∑(|D_v| / |D|) × Entropy(D_v)

[0121] Where: Gain(D, A) is the information gain of feature A, Entropy(D) is the information entropy of dataset D, D_v is the data subset divided according to the value v of feature A, |D_v| is the number of samples in data subset D_v, and |D| is the total number of samples in dataset D;

[0122] Entropy(D) = -∑ p_k log2 p_k

[0123] Where p_k is the proportion of the k-th class of samples;

[0124] 3. The expression for random forest ensemble voting is:

[0125] f(x) = argmax_c ∑ I(h_t(x) = c)

[0126] Where f(x) is the final prediction result of the random forest for sample x, argmax_c is the class c that maximizes the expression, h_t(x) is the prediction result of the t-th decision tree for sample x, c is the classification class, and I() is the indicator function.

[0127] 4. The expression for evaluating classification accuracy is:

[0128] Accuracy = (TP + TN) / (TP + TN + FP + FN)

[0129] Where Accuracy is the classification accuracy, TP is the number of correctly predicted positive examples, TN is the number of correctly predicted negative examples, FP is the number of incorrectly predicted positive examples, and FN is the number of incorrectly predicted negative examples.

[0130] S5. Based on the communication quality classification and evaluation results obtained in step S4, generate a multi-dimensional spatial resolution communication hole heatmap to complete the dynamic evaluation of three-dimensional airspace communication holes based on UAVs.

[0131] Furthermore, the specific implementation method of step S5 includes the following steps:

[0132] S5.1. Spatial Mesh Generation: Create a three-dimensional mesh based on a spatial resolution of 50 meters × 50 meters × 50 meters;

[0133] S5.2. Data Point Mapping: Mapping communication data points to corresponding spatial resolution cells. The formula for dividing the spatial resolution cells is as follows:

[0134] x_bins = [x_min, x_min + Δx, x_min + 2Δx, ..., x_max]

[0135] y_bins = [y_min, y_min + Δy, y_min + 2Δy, ..., y_max]

[0136] z_bins = [z_min, z_min + Δz, z_min + 2Δz, ..., z_max]

[0137] Where x_bins is the sequence of dividing points in the longitude direction, x_min and x_max are the minimum and maximum values ​​in the longitude range, respectively, Δx is the spatial resolution in the longitude direction, y_bins is the sequence of dividing points in the latitude direction, y_min and y_max are the minimum and maximum values ​​in the latitude range, respectively, Δy is the spatial resolution in the latitude direction, z_bins is the sequence of dividing points in the altitude direction, z_min and z_max are the minimum and maximum values ​​in the altitude range, respectively, Δz is the spatial resolution in the altitude direction;

[0138] S5.3. Communication Quality Calculation: Calculate the average communication quality score within each spatial resolution cell. The calculation formula is as follows:

[0139] Q(i,j,k) = (1 / N_ijk) × ∑ w_n × q_n

[0140] Where Q(i,j,k) is the communication quality score of the (i,j,k)th spatial unit, N_ijk is the number of data points in the spatial unit, w_n is the weight coefficient of the nth data point, and q_n is the communication quality score of the nth data point;

[0141] Then, communication void identification is performed, using the following formula:

[0142] Void(i,j,k) = { 1, if Q(i,j,k) < Q_void or N_ijk = 0, otherwise}

[0143] Where Void(i,j,k) indicates whether the (i,j,k)-th spatial unit is a communication hole, 1 indicates a hole and 0 indicates no hole, and Q_void is the communication hole threshold, which is set to 20;

[0144] S5.4. Color Mapping: Red, yellow, and green color gradients are used to represent the strength of communication quality. The logical expression is as follows:

[0145] Color(i,j,k) = { Red, if Q(i,j,k) < Q_threshold1; Yellow, if Q_threshold1 ≤ Q(i,j,k) < Q_threshold2; Green, if Q(i,j,k) ≥ Q_threshold2}

[0146] Where Color(i,j,k) is the color of the (i,j,k)th spatial unit, Q(i,j,k) is the communication quality score of the (i,j,k)th spatial unit, Q_threshold1 is the communication quality threshold 1, set to 30; Q_threshold2 is the communication quality threshold 2, set to 70;

[0147] S5.5. 3D Visualization: Generate a heatmap of communication voids with multi-dimensional spatial resolution;

[0148] S5.6. Multi-mode support: Generates independent multi-dimensional spatial resolution communication hole heatmaps for local area network communication, wide area network communication, and satellite communication, and completes dynamic assessment of three-dimensional airspace communication holes based on UAVs.

[0149] Example 2:

[0150] A dynamic assessment system for three-dimensional airspace communication voids based on unmanned aerial vehicles (UAVs) is implemented based on the dynamic assessment method for three-dimensional airspace communication voids based on UAVs described in Example 1. The system includes a data acquisition module, a feature extraction module, a cluster analysis module, a classification assessment module, a heat map generation module, and a visualization module, which are connected sequentially.

[0151] The data acquisition module is responsible for collaborative data acquisition among multiple UAVs, including a built-in communication unit and an aerial signal acquisition unit; the feature extraction module is responsible for extracting multi-dimensional quality indicators from the raw communication signaling; the clustering analysis module is responsible for spatiotemporal joint clustering analysis using the ST-DBSCAN algorithm; the classification evaluation module is responsible for intelligent classification of communication quality based on the random forest algorithm; the heatmap generation module is responsible for generating a three-dimensional airspace communication hole heatmap; and the visualization display module is responsible for providing an interactive three-dimensional visualization interface.

[0152] The UAV-based three-dimensional airspace communication hole dynamic assessment system described in this embodiment is based on the UAV-based three-dimensional airspace communication hole dynamic assessment method described in Embodiment 1. It includes using a 50m×50m×50m spatial resolution, and through the collaborative work of the UAV communication unit and the airborne signal acquisition unit, it can accurately assess the distribution of three-dimensional airspace communication holes under different communication modes, generate a communication hole spatial resolution heat map, and provide data support for UAV flight path optimization and communication network blind spot filling.

[0153] It should be noted that relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0154] Although this application has been described above with reference to specific embodiments, various modifications can be made and components can be replaced with equivalents without departing from the scope of this application. In particular, as long as there is no structural conflict, the features in the specific embodiments disclosed in this application can be combined with each other in any way. The lack of an exhaustive description of these combinations in this specification is merely for the sake of brevity and resource conservation. Therefore, this application is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.

Claims

1. A method for dynamic evaluation of three-dimensional airspace communication voids based on unmanned aerial vehicles (UAVs), characterized in that, Includes the following steps: S1. Collect collaborative data from multiple UAVs. Utilize the built-in communication unit and dedicated airborne signal acquisition unit of the UAVs to collect communication signaling data from 2G, 3G, 4G, 5G and satellite communications in real time to obtain raw communication signaling data from multiple networks. S2. Signaling multidimensional feature extraction: Extract multidimensional quality indicator features from the original multi-network communication signaling data obtained in step S1; S3. Spatiotemporal joint analysis and clustering: Based on the original multi-network communication signaling data obtained in step S1 and the multi-dimensional quality index features obtained in step S2, the data are filtered according to different communication modes and then normalized to obtain local area network communication datasets, wide area network communication datasets, and satellite communication datasets. Then, a density-based spatiotemporal clustering method is used to identify communication voids and obtain the spatiotemporal distribution results of communication voids in the three-dimensional spatial domain. S4. Based on the original multi-network communication signaling data obtained in step S1, the multi-dimensional quality index features obtained in step S2, and the spatiotemporal distribution results of communication holes in the three-dimensional spatial domain obtained in step S3, a communication quality classification and evaluation model is established using the random forest and SVM fusion method to obtain the communication quality classification and evaluation results. S5. Based on the communication quality classification and evaluation results obtained in step S4, generate a multi-dimensional spatial resolution communication hole heatmap to complete the dynamic evaluation of three-dimensional airspace communication holes based on UAVs.

2. The method for dynamic evaluation of three-dimensional airspace communication voids based on unmanned aerial vehicles (UAVs) according to claim 1, characterized in that, In step S1, the key parameters for data sampling are set, including sampling frequency, spatial resolution, and time window. The sampling frequency is 1-10Hz, the spatial resolution is the standard spatial resolution of 50m×50m×50m, and the time window size is 5-30 minutes. The collected data includes longitude, latitude, altitude, timestamp, signal strength index, signal-to-noise ratio, bit error rate, delay jitter, packet loss rate, spectral efficiency, and link quality indicator.

3. A method for dynamic evaluation of three-dimensional airspace communication voids based on unmanned aerial vehicles (UAVs) according to claim 1 or 2, characterized in that, The multi-dimensional quality indicator features mentioned in step S2 include signal strength indicators, signal-to-noise ratio, bit error rate, latency jitter, packet loss rate, spectral efficiency, and link quality indicators. The signaling multi-dimensional feature extraction process includes the following steps: S2.

1. Temporal feature extraction: Calculate the mean, standard deviation, skewness, and kurtosis of the signal intensity; S2.

2. Calculate the signal-to-noise ratio based on the logarithmic ratio of signal power and noise power; S2.

3. Frequency Domain Feature Extraction: Calculate the spectral centroid, spectral bandwidth, spectral flatness, spectral roll-off point, spectral entropy, spectral kurtosis, and spectral skewness; S2.

4. Statistical Feature Extraction: Calculate the skewness and kurtosis statistics of the signal distribution.

4. The method for dynamic evaluation of three-dimensional airspace communication voids based on unmanned aerial vehicles (UAVs) according to claim 3, characterized in that, The specific implementation method of step S3 includes the following steps: S3.

1. Prepare clustering data: including longitude, latitude, altitude, timestamp, signal strength index, signal-to-noise ratio, bit error rate, latency jitter, packet loss rate, spectral efficiency, and link quality indicator; S3.

2. Filter the clustering data prepared in step S3.1 according to communication mode, and then use the Z-score normalization method to normalize it to obtain local area network communication dataset, wide area network communication dataset, and satellite communication dataset; S3.

3. Apply the density-based spatiotemporal clustering algorithm ST-DBSCAN to identify communication gaps; The core distance metric formula of the ST-DBSCAN spatiotemporal clustering algorithm is: ； Where d(x_i, x_j) is the spatiotemporal distance between the i-th data point and the j-th data point, x_i, y_i, z_i are the longitude, latitude, and altitude coordinates of the i-th data point, respectively, x_j, y_j, z_j are the longitude, latitude, and altitude coordinates of the j-th data point, t_i is the timestamp of the i-th data point, and t_j is the timestamp of the j-th data point; Let d(x_i, x_j) ≤ ε and |t_i - t_j| ≤ δ_t, then the i-th data point and the j-th data point are considered to be spatially adjacent, where ε is the spatial distance threshold and δ_t is the time difference threshold. S3.

4. Analyze the clustering results obtained in step S3.3, and identify data points with a label of -1 as communication holes.

5. The method for dynamic evaluation of three-dimensional airspace communication voids based on unmanned aerial vehicles (UAVs) according to claim 4, characterized in that, In step S4, a random forest classifier is created, with parameters including 100 decision trees and a maximum depth of 10 layers. The spatiotemporal distribution results of communication holes in the three-dimensional spatial domain obtained in step S3 are used as training data. The random forest classification algorithm is used to intelligently classify and evaluate communication quality, and a communication quality classification model adapted to different environments is established. Step S3 provides labeled data and feature enhancement for classification evaluation. The labeled data and feature enhancement constitute a collaboratively optimized evaluation system. Finally, the communication quality classification level is defined into five levels according to the random forest classifier, including excellent, good, average, poor, and very poor.

6. The method for dynamic evaluation of three-dimensional airspace communication voids based on unmanned aerial vehicles (UAVs) according to claim 5, characterized in that, The specific implementation method of step S5 includes the following steps: S5.

1. Spatial Mesh Generation: Create a three-dimensional mesh based on a spatial resolution of 50 meters × 50 meters × 50 meters; S5.

2. Data Point Mapping: Mapping communication data points to corresponding spatial resolution cells. The formula for dividing the spatial resolution cells is as follows: x_bins = [x_min, x_min + Δx, x_min + 2Δx, ..., x_max]; y_bins = [y_min, y_min + Δy, y_min + 2Δy, ..., y_max]; z_bins = [z_min, z_min + Δz, z_min + 2Δz, ..., z_max]; Where x_bins is the sequence of dividing points in the longitude direction, x_min and x_max are the minimum and maximum values ​​in the longitude range, respectively, Δx is the spatial resolution in the longitude direction, y_bins is the sequence of dividing points in the latitude direction, y_min and y_max are the minimum and maximum values ​​in the latitude range, respectively, Δy is the spatial resolution in the latitude direction, z_bins is the sequence of dividing points in the altitude direction, z_min and z_max are the minimum and maximum values ​​in the altitude range, respectively, Δz is the spatial resolution in the altitude direction; S5.

3. Communication Quality Calculation: Calculate the average communication quality score within each spatial resolution cell. The calculation formula is as follows: Q(i,j,k) = (1 / N_ijk) × ∑ w_n × q_n; Where Q(i,j,k) is the communication quality score of the (i,j,k)th spatial unit, N_ijk is the number of data points in the spatial unit, w_n is the weight coefficient of the nth data point, and q_n is the communication quality score of the nth data point; Then, communication void identification is performed, using the following formula: Void(i,j,k) = { 1, if Q(i,j,k) < Q_void or N_ijk = 0, otherwise}; Where Void(i,j,k) indicates whether the (i,j,k)-th spatial unit is a communication hole, 1 indicates a hole and 0 indicates no hole, and Q_void is the communication hole threshold, which is set to 20; S5.

4. Color Mapping: Use red, yellow, and green color gradients to represent the strength of communication quality; S5.

5. 3D Visualization: Generate a heatmap of communication voids with multi-dimensional spatial resolution; S5.

6. Multi-mode support: Generates independent multi-dimensional spatial resolution communication hole heatmaps for local area network communication, wide area network communication, and satellite communication, and completes dynamic assessment of three-dimensional airspace communication holes based on UAVs.

7. A dynamic evaluation system for three-dimensional airspace communication voids based on unmanned aerial vehicles (UAVs), implemented using the dynamic evaluation method for three-dimensional airspace communication voids based on UAVs as described in any one of claims 1-6, characterized in that... It includes a data acquisition module, a feature extraction module, a cluster analysis module, a classification evaluation module, a heatmap generation module, and a visualization display module, which are connected in sequence. The data acquisition module is responsible for collaborative data acquisition among multiple UAVs, including a built-in communication unit and an aerial signal acquisition unit; the feature extraction module is responsible for extracting multi-dimensional quality indicators from the raw communication signaling; the clustering analysis module is responsible for spatiotemporal joint clustering analysis using the ST-DBSCAN algorithm; the classification evaluation module is responsible for intelligent classification of communication quality based on the random forest algorithm; the heatmap generation module is responsible for generating a three-dimensional airspace communication hole heatmap; and the visualization display module is responsible for providing an interactive three-dimensional visualization interface.

Images