An unmanned aerial vehicle cluster communication interaction method and system based on an ad hoc network

Abstract

The application discloses an unmanned aerial vehicle cluster communication interaction method and system based on an ad hoc network, relates to the wireless ad hoc network technical field, and comprises the following steps: acquiring three-dimensional positioning coordinates of all unmanned aerial vehicles, collecting residual power percentage data, electing cluster head nodes based on a dynamic weight formula, and constructing a cluster head responsibility airspace; constructing a frequency band value evaluation function based on electromagnetic wave propagation characteristics and spectrum availability data, establishing a comprehensive map integrating electromagnetic characteristics and terrain constraints, generating a main communication path and an alternative path through an A* algorithm combined with a double-time-scale dynamic optimization mechanism, and outputting a communication path set; and through dynamic election of cluster head nodes and construction of a responsibility airspace, efficient organization and management of the unmanned aerial vehicle cluster communication topology are realized, the stability and energy efficiency of the network structure are improved, the communication reliability and anti-interference capability are improved, and the link failure risk is reduced, so that solid protection is provided for efficient cooperative communication of the unmanned aerial vehicle cluster in a complex task scenario.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of wireless Ad Hoc network, and particularly relates to a method and system for communication interaction of unmanned aerial vehicle cluster based on Ad Hoc network. BACKGROUND

[0002] With the rapid development of wireless communication technology and unmanned aerial vehicles, unmanned aerial vehicle cluster shows a wide application prospect in many fields such as military reconnaissance, disaster relief and smart city. In order to realize efficient cooperative work, unmanned aerial vehicle cluster needs to have the ability of autonomous networking and dynamic communication interaction mechanism. Ad Hoc network becomes an important supporting technology for unmanned aerial vehicle cluster communication because of its characteristics of not needing to rely on fixed infrastructure and supporting fast deployment and movement of nodes. In recent years, the Ad Hoc network architecture based on clustering structure is widely researched and applied. Spectrum resource is the core element of wireless communication, and presents the characteristics of high dynamicity and spatial heterogeneity in the high-density unmanned aerial vehicle environment. Therefore, how to effectively perceive, predict and optimize the spectrum usage state has become one of the key research directions.

[0003] The prior art usually adopts a static or semi-dynamic cluster head election strategy, which is difficult to adapt to the frequent changes of topology caused by high-speed movement of unmanned aerial vehicles. At the same time, spectrum sensing is mostly dependent on independent measurement of local nodes, and lacks a global perspective of spectrum state modeling mechanism, which limits the resource allocation efficiency. Especially in the face of sudden interference or link quality deterioration, the traditional method lacks an effective real-time response mechanism, which is easy to cause communication interruption or resource waste. SUMMARY

[0004] In view of the above existing problems, the present application is proposed.

[0005] Therefore, the present application provides a method for communication interaction of unmanned aerial vehicle cluster based on Ad Hoc network to solve the problem of communication link failure response lag.

[0006] To solve the above technical problems, the present application provides the following technical solutions.

[0007] In a first aspect, the application provides a method for unmanned aerial vehicle cluster communication interaction based on an ad hoc network, which comprises: obtaining three-dimensional positioning coordinates of all unmanned aerial vehicles and collecting remaining battery percentage data; electing a cluster head node based on a dynamic weight formula and constructing a cluster head responsibility airspace; scanning electromagnetic wave signals in the cluster head responsibility airspace by using the cluster head node to obtain spectrum feature parameters, inputting the spectrum feature parameters into a spectrum prediction model to generate a weight sensitive state, performing gradient aggregation through gradient-temperature signal conversion and weighted average to generate a global spectrum heat map; constructing a frequency band value evaluation function based on electromagnetic wave propagation characteristics and spectrum availability data, and establishing a comprehensive map that integrates electromagnetic characteristics and terrain constraints, generating a main communication path and an alternative path through A* algorithm combined with a double-time-scale dynamic optimization mechanism, and outputting a communication path set; obtaining a member node sequence and interference frequency band data by analyzing fields in the communication path set, constructing a two-dimensional link-frequency band cost matrix, performing optimal matching of the two-dimensional link-frequency band cost matrix through the Hungarian algorithm, monitoring link quality for calculation, obtaining an instantaneous bit error rate, triggering path switching when the instantaneous bit error rate exceeds a link failure criterion threshold, and generating a channel allocation instruction set; switching a transceiving frequency band and adjusting a radiation beam direction according to working frequency points, time slot parameters, and power parameters of the channel allocation instruction set, embedding forward error correction codes for data transmission, collecting transmission status codes in real time through the cluster head node, and generating a network status communication report through fusion of three-dimensional frequency domain-time domain-space domain analysis.

[0008] As a preferred scheme of the method for unmanned aerial vehicle cluster communication interaction based on an ad hoc network, the method comprises the following steps of forming a clustering topology graph,

[0009] calculating a dynamic weight score of each unmanned aerial vehicle node, selecting a node with the highest score as a cluster head node, and collecting a historical maximum communication distance of the cluster head node, constructing a cluster head responsibility airspace based on three-dimensional positioning coordinates of the unmanned aerial vehicle and taking the cluster head node as a center and the historical maximum communication distance as a radius.

[0010] calculating a Euclidean distance between the candidate node and the cluster head node based on the remaining battery percentage data, transmitting a detection signal through the cluster head node, measuring a received signal strength of the candidate node, and taking the candidate node with a Euclidean distance smaller than the historical maximum communication distance and a received signal strength greater than the transmitted detection signal as a member node.

[0011] As a preferred scheme of the method for unmanned aerial vehicle cluster communication interaction based on an ad hoc network, the method comprises the following steps of generating a global spectrum heat map,

[0012] The electromagnetic wave signal of the cluster head node scanning the cluster head responsibility airspace is used to obtain the spectrum feature parameters by spectrum feature analysis of the electromagnetic wave signal, and the spectrum feature parameters are input into the spectrum prediction model, the weight sensitive state is calculated by back propagation, the weight sensitive state is coded into a temperature fluctuation signal, and weighted average is performed to obtain the aggregated weight sensitive state;

[0013] The spectrum prediction model is updated by the aggregated weight sensitive state, the updated spectrum prediction model is obtained, the three-dimensional positioning coordinates of all unmanned aerial vehicles are input into the updated spectrum prediction model, the frequency band availability probability matrix of the three-dimensional positioning coordinates is obtained, and the global spectrum heat map is converted.

[0014] As a preferred scheme of the unmanned aerial vehicle cluster communication interaction method based on the self-organizing network, wherein: the output communication path set has the following specific steps,

[0015] The three-dimensional positioning coordinates of each member node in the cluster head responsibility airspace are traversed, and the frequency band availability probability and background noise corresponding to the position of the member node are obtained by querying the global spectrum heat map, to obtain the spectrum availability data;

[0016] Based on the electromagnetic wave propagation characteristics, the channel transmission potential is obtained, the frequency band value evaluation function is constructed according to the spectrum availability data, the communication quality classification label is obtained by fusing the frequency band availability data and the channel transmission potential double factors, and the high-value communication area is labeled;

[0017] Based on the high-value communication area, the communication quality classification label is mapped to the three-dimensional geographic space, the comprehensive map integrating electromagnetic characteristics and terrain constraints is established, and the safe passage area, the restricted area and the absolute forbidden area are divided;

[0018] The main communication path is obtained by path search in the three-dimensional geographic space through the A* algorithm, a plurality of alternative paths are generated on the basis of the main communication path by using a double-time-scale dynamic optimization mechanism, the quality scores of the alternative paths are calculated and arranged according to the quality scores, and the communication path set is obtained.

[0019] As a preferred scheme of the unmanned aerial vehicle cluster communication interaction method based on the self-organizing network, wherein: the instantaneous bit error rate is obtained, and the specific steps are as follows,

[0020] The communication link sequence field and the interference frequency band data in the communication path set are analyzed, a two-dimensional link-frequency band cost matrix is constructed, the communication link is taken as the left node set, the available frequency band is taken as the right node set, the Hungarian algorithm is used to traverse all communication links and calculate the matching cost, and a channel allocation mapping table is output;

[0021] The radio frequency front end of the member node is reconfigured according to the channel allocation mapping table, the received signal strength, the number of symbol errors and the attenuation trend are recorded on the communication link, the signal-to-noise ratio is calculated through the instantaneous power ratio formula, the monitoring data are obtained, and the transient bit error rate is calculated through the bit error instantaneous quantization rate formula.

[0022] As a preferred scheme of the unmanned aerial vehicle cluster communication interaction method based on the self-organizing network, when the instantaneous bit error rate exceeds the link failure criterion threshold, path switching is triggered, and a channel allocation instruction set is generated, and the specific steps are as follows,

[0023] In each communication period, the transient bit error rate is compared with the preset link failure threshold, and when the transient bit error rate in the continuous communication period exceeds the link failure criterion threshold, the link failure is determined.

[0024] The cluster head node activates the communication path set, selects the highest quality score of the alternative path, scans the optimal frequency band of the new communication path, dynamically allocates a non-overlapping frequency band group, and generates a channel allocation instruction set.

[0025] As a preferred scheme of the unmanned aerial vehicle cluster communication interaction method based on the self-organizing network, the network state communication report is generated, and the specific steps are as follows,

[0026] Based on the channel allocation instruction set, the working frequency of the radio frequency front end is switched to the specified working frequency point, the power amplifier is dynamically calibrated, the phased array antenna is used to adjust the radiation beam direction, the error correction coding and interleaver parameters are embedded in the transmitted data stream, and the data transmission is performed, and the data is packaged as a structured state data packet.

[0027] All the structured state data packets are converged by the cluster head node, the frequency domain dimension, the time domain dimension and the space domain dimension are analyzed through three-dimensional situation, the spectrum load thermodynamic map data entity, the time slot quality rating matrix and the three-dimensional power distribution cloud diagram are obtained, and the network state communication report is integrated.

[0028] In a second aspect, the present application provides a UAV cluster communication interaction system based on an ad hoc network, comprising: a responsibility airspace module, configured to obtain three-dimensional positioning coordinates of all UAVs, collect remaining battery percentage data, elect a cluster head node based on a dynamic weight formula, and construct a cluster head responsibility airspace; a heat map module, configured to obtain spectrum feature parameters by scanning electromagnetic wave signals in the cluster head responsibility airspace using the cluster head node, input the spectrum feature parameters into a spectrum prediction model to generate a weight sensitive state, perform gradient aggregation through gradient-temperature signal conversion and weighted average, and generate a global spectrum heat map; a path module, configured to construct a frequency band value evaluation function based on electromagnetic wave propagation characteristics and spectrum availability data, establish a comprehensive map that fuses electromagnetic characteristics and terrain constraints, generate a main communication path and an alternative path through an A* algorithm combined with a double-time-scale dynamic optimization mechanism, and output a communication path set; a channel allocation module, configured to obtain member node sequences and interference frequency band data by analyzing fields in the communication path set, construct a two-dimensional link-frequency band cost matrix, perform optimal matching of a bipartite graph of the two-dimensional link-frequency band cost matrix using a Hungarian algorithm, monitor link quality for calculation, obtain an instantaneous bit error rate, trigger path switching when the instantaneous bit error rate exceeds a link failure criterion threshold, and generate a channel allocation instruction set; and a reporting module, configured to switch a transceiving frequency band and adjust a radiation beam direction according to working frequency points, time slot parameters, and power parameters of the channel allocation instruction set, embed a forward error correction code for data transmission, collect transmission status codes in real time through the cluster head node, and generate a network status communication report by fusing three-dimensional frequency domain-time domain-space domain analysis.

[0029] In a third aspect, the present application provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, any step of the UAV cluster communication interaction method based on an ad hoc network according to the first aspect of the present application is implemented.

[0030] In a fourth aspect, the present application provides a computer readable storage medium, which stores a computer program, and when the computer program is executed by a processor, any step of the UAV cluster communication interaction method based on an ad hoc network according to the first aspect of the present application is implemented.

[0031] The application has the beneficial effects that: through dynamic election of cluster head nodes and construction of responsibility airspace, efficient organization and management of the unmanned aerial vehicle cluster communication topology are realized, the stability and energy efficiency of the network structure are improved, the cluster head nodes are used for scanning electromagnetic signals and generating a global spectrum heat map, real-time sensing and modeling capability of complex electromagnetic environment is achieved, and the utilization efficiency of spectrum resources is improved significantly; the spectrum heat map and the double time scale mechanism are combined to output a communication path set, the flexibility and robustness of communication path planning are enhanced, high-quality communication links can be maintained in a dynamic environment, the communication reliability and anti-interference capability are effectively improved, and the link failure risk is reduced, thereby providing a solid guarantee for efficient cooperative communication of the unmanned aerial vehicle cluster in complex task scenarios. BRIEF DESCRIPTION OF DRAWINGS

[0032] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0033] Fig. 1 Flowchart of the unmanned aerial vehicle cluster communication interaction method based on the ad hoc network.

[0034] Fig. 2 Schematic diagram of the unmanned aerial vehicle cluster communication interaction system based on the ad hoc network.

[0035] Fig. 3 Flowchart of generating a global spectrum heat map.

[0036] Fig. 4 Flowchart of outputting a communication path set. DETAILED DESCRIPTION

[0037] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the drawings.

[0038] In the following description, many specific details are set forth in order to provide a thorough understanding of the present application, but the present application can also be implemented in other ways different from those described herein, and those skilled in the art can make similar generalizations without departing from the connotation of the present application, therefore the present application is not limited to the specific embodiments disclosed below.

[0039] Secondly, the "one embodiment" or "embodiment" referred to herein means that the specific features, structures or characteristics can be included in at least one implementation of the present application. "In one embodiment" appearing in different places in the specification does not mean the same embodiment, nor is it an independent or selective embodiment that excludes other embodiments.

[0040] Reference Figs. 1-4 As one embodiment of the present invention, this embodiment provides a method for communication and interaction of unmanned aerial vehicle (UAV) swarms based on ad hoc networks, comprising the following steps:

[0041] S1. Obtain the three-dimensional positioning coordinates of all drones and collect the remaining battery percentage data. Elect the cluster head node based on the dynamic weight formula and construct the cluster head responsibility airspace.

[0042] Calculate the dynamic weight score of each UAV node, select the node with the highest score as the cluster head node, and the remaining UAV nodes as candidate nodes. Collect the historical maximum communication distance of the cluster head node, and construct the cluster head responsibility airspace based on the UAV's three-dimensional positioning coordinates, with the cluster head node as the center and the historical maximum communication distance as the radius.

[0043] Specifically, the system locates all drone nodes within the drone swarm, acquires the three-dimensional positioning coordinates of each drone node using an onboard global navigation satellite receiver, and collects the remaining battery percentage data of each drone node using a voltage sensor. It collects communication performance parameters such as historical signal-to-noise ratio, link stability, and packet loss rate to obtain communication quality. Based on a dynamic weighting formula, it calculates the dynamic weight score for each drone node, periodically broadcasts the dynamic weight scores of all drone nodes, and selects the drone node with the highest score as the cluster head node, with the remaining drone nodes serving as candidate nodes. It then retrieves the historical communication database of the cluster head node and extracts the historical maximum communication distance record value. Finally, it constructs a spherical cluster head responsibility airspace with the three-dimensional positioning coordinates of the cluster head node as the center and the historical maximum communication distance record value as the radius.

[0044] It should be noted that the formula for calculating the dynamic weight score of each drone node is as follows:

[0045]

[0046] in, Indicates the first The dynamic weight score of each drone node. Indicates the first The remaining battery percentage of each drone node. Indicates the first Communication quality of each drone node, This represents the energy weighting coefficient. Represents the communication weighting coefficient. This represents the position weighting coefficient. This represents the drone node index variable. Indicates the first The distance between each drone node and the cluster head node.

[0047] The Euclidean distance between candidate nodes and cluster head nodes is calculated based on the remaining power percentage data. The cluster head node transmits a probe signal, and the received signal strength of the candidate nodes is measured. Candidate nodes whose Euclidean distance is less than the historical maximum communication distance and whose received signal strength is greater than the transmitted probe signal are selected as member nodes.

[0048] Specifically, the Euclidean distance between each candidate node and the cluster head node is calculated using the distance formula between two points in three-dimensional space. The cluster head node transmits a probe signal with a fixed power value, and each candidate node measures the received signal strength value. Two parallel conditional verifications are performed on each candidate node: the first verifies whether the Euclidean distance of the current candidate node is less than the historical maximum communication distance record value, and the second verifies whether the received signal strength value of the current candidate node is greater than or equal to the probe signal. When a candidate node simultaneously satisfies the conditions of having an Euclidean distance value less than the historical maximum communication distance record value and a received signal strength value greater than or equal to the probe signal, the candidate node is certified as a member node.

[0049] It should be noted that the formula for calculating the Euclidean distance between a candidate node and the cluster head node is:

[0050]

[0051] in, Indicates the first The Euclidean distance between each candidate node and the cluster head node. Indicates the first Candidate nodes in three-dimensional coordinates The position of the axis Indicates the first Candidate nodes in three-dimensional coordinates The position of the axis Indicates the first Candidate nodes in three-dimensional coordinates The position of the axis Indicates the first Cluster head nodes in three-dimensional coordinates The position of the axis Indicates the first Cluster head nodes in three-dimensional coordinates The position of the axis Indicates the first Cluster head nodes in three-dimensional coordinates The position of the axis Indicates the first Candidate node index variables.

[0052] S2. Use cluster head nodes to scan electromagnetic wave signals in the cluster head responsibility space to obtain spectral characteristic parameters. Input the spectral characteristic parameters into the spectral prediction model to generate weighted sensitive states. Perform gradient aggregation through gradient-temperature signal conversion and weighted averaging to generate a global spectral heatmap.

[0053] The electromagnetic wave signals in the cluster head responsibility airspace are scanned, the frequency spectrum characteristic parameters are obtained by frequency spectrum characteristic analysis on the electromagnetic wave signals, and the frequency spectrum characteristic parameters are input into the frequency spectrum prediction model. The weight sensitive state is calculated through back propagation, the weight sensitive state is coded into a temperature fluctuation signal, the cluster head node performs weighted average, and the aggregated weight sensitive state is obtained.

[0054] Specifically, the electromagnetic wave signals of the specified frequency band in the cluster head responsibility airspace are scanned by the cluster head node, the captured electromagnetic wave signals are analyzed by using a spectrum analyzer to perform fast Fourier transform, after filtering out out-of-band noise by an anti-aliasing filter (example: elliptical filter passband ripple 0.1 dB), a discrete sequence is generated by an analog-to-digital converter at the Nyquist rate (example: 100 MHz bandwidth sampling 250 MS / s); the sampling sequence is subjected to a Hanning window to suppress spectral leakage; a base-2 time decimation FFT algorithm is performed to obtain four frequency spectrum characteristic parameters: frequency spectrum occupancy rate, noise floor, channel coherence bandwidth, and multipath time delay spread; the frequency spectrum characteristic parameter values are input into the frequency spectrum prediction model for forward propagation calculation, and the frequency band availability probability prediction value is output; the cross-entropy loss function value between the frequency band availability probability prediction value and the actual frequency spectrum measurement value is compared, and the weight sensitive state of each weight parameter in the frequency spectrum prediction model is obtained through the back propagation algorithm; the weight sensitive state is converted into the temperature change of the temperature control unit by using a semiconductor thermosensitive material, and a temperature fluctuation signal waveform carrying gradient information is generated; the cluster head node receives the temperature fluctuation signal waveforms sent by all member nodes, assigns a weight coefficient (example: the weight is increased by 0.1 for every 3 dB increase in signal-to-noise ratio) according to the signal-to-noise ratio values of each member node to the cluster head node, performs weighted arithmetic average calculation on the demodulated gradient values, and outputs the aggregated weight sensitive state.

[0055] The frequency spectrum prediction model is updated by the aggregated weight sensitive state, an updated frequency spectrum prediction model is obtained, the three-dimensional positioning coordinates of all unmanned aerial vehicles are input into the updated frequency spectrum prediction model, a frequency band availability probability matrix of the three-dimensional positioning coordinates is obtained, and the frequency band availability probability matrix is converted into a global frequency spectrum heat map.

[0056] Specifically, the frequency spectrum prediction model weight parameters are adjusted by the received and demodulated aggregated weight sensitive state, the frequency spectrum prediction model weight parameters are iteratively updated by using the gradient descent principle, the three-dimensional positioning coordinates of each unmanned aerial vehicle are input as input data of the updated frequency spectrum prediction model to perform a forward inference process, and the availability probability of each communication frequency band corresponding to the spatial position is output; all unmanned aerial vehicle node spatial positions and frequency band availability probabilities are organized into a two-dimensional data matrix, the row dimension of the two-dimensional data matrix corresponds to the unmanned aerial vehicle node number index, the column dimension corresponds to the frequency band index, and the two-dimensional data matrix elements store the frequency band availability probability; the frequency band availability probability matrix is converted into a visual heat map representation form according to the chroma mapping rule, the available probability range is mapped to a continuous color spectrum distribution, and a global frequency spectrum heat map is generated.

[0057] S3. Construct a frequency band value evaluation function based on electromagnetic wave propagation characteristics and spectrum availability data, and establish a comprehensive map integrating electromagnetic characteristics and terrain constraints, generate main communication paths and alternative paths through A* algorithm combined with double-time-scale dynamic optimization mechanism, and output a communication path set.

[0058] Traverse the three-dimensional positioning coordinates of each member node within the cluster head responsibility airspace, and query the global spectrum heat map to obtain the frequency band availability probability and background noise corresponding to the member node position, and obtain the spectrum availability data.

[0059] Specifically, the three-dimensional positioning coordinates of all member nodes within the cluster head responsibility airspace are traversed, and coordinate index matching operations are performed point by point. The three-dimensional positioning coordinates are aligned with the grid index system of the global spectrum heat map through a space mapping algorithm to obtain a grid unit. The frequency band availability probability and background noise values are derived from the global spectrum heat map, and the frequency band availability probability and background noise values of the corresponding grid unit are retrieved. The spectrum availability data of each member node is formed.

[0060] Based on the propagation characteristics of electromagnetic waves in real environment, the channel transmission potential is obtained, and the frequency band value evaluation function is constructed according to the spectrum availability data. The frequency band availability data and channel transmission potential are integrated to obtain communication quality classification labels and mark high-value communication areas.

[0061] Specifically, based on the propagation characteristics of electromagnetic waves in real environment, the path loss, multipath delay and terrain shielding effect are analyzed to generate channel transmission potential parameters. Combined with the frequency band availability probability and background noise values in the spectrum availability data, the frequency band value evaluation function is constructed. The frequency band availability probability is defined as the availability factor of the frequency band value evaluation function, and the background noise reciprocal is taken as the suppression factor. The path loss compensation coefficient in the channel transmission potential parameter is superimposed, the path loss compensation coefficient is input into the frequency band value evaluation function to obtain the comprehensive evaluation value. According to the interval division of the comprehensive evaluation value, the communication quality classification labels are divided: the coordinate points with comprehensive evaluation value greater than or equal to the optimal communication quality threshold are marked as first-level quality labels, the coordinate points with comprehensive evaluation value between the lower limit of the basic communication guarantee threshold and the optimal communication quality threshold are marked as second-level quality labels, and the coordinate points with comprehensive evaluation value lower than the lower limit of the basic communication guarantee threshold are marked as third-level quality labels. Extract the three-dimensional geographic coordinate points of all first-level quality labels and perform adjacent space clustering. The boundary coordinate set of the continuous three-dimensional space region formed is defined as the high-value communication area.

[0062] It should be noted that the optimal communication quality threshold is determined by the minimum service quality requirement of the target service, for example, high-definition video transmission requires a signal-to-noise ratio of ≥20 dB, and the minimum value of the comprehensive evaluation value in the historical communication data that meets the signal-to-noise ratio requirement (example value 80) is set as the optimal communication quality threshold. The basic communication guarantee threshold is determined by the maximum bit error rate, for example, when the bit error rate is ≤0.001, the lower quartile of the distribution of the comprehensive evaluation value (example value 50) is set as the basic communication guarantee threshold.

[0063] Based on the high-value communication area, the communication quality classification label is mapped to the three-dimensional geographic space, an integrated map combining electromagnetic characteristics and terrain constraints is established, and safe passage areas, restricted areas and absolute prohibited areas are divided.

[0064] Specifically, based on the three-dimensional geographic coordinate points of the high-value communication area, the annotated communication quality classification labels are mapped to the three-dimensional geographic space. The first-level quality label coverage area is superimposed with low-noise high-usable frequency band electromagnetic characteristics, the second-level quality label area is superimposed with medium-noise medium-usable frequency band electromagnetic characteristics, and the third-level quality label area is superimposed with high-noise low-usable frequency band electromagnetic characteristics. Combined with the terrain elevation database (source example: airborne LiDAR historical scanning data) and the terrain constraint area with fixed obstacles; according to the coupling results of electromagnetic characteristics and terrain constraints, the area is divided. The high-value communication area that meets the first-level quality label and has no obstruction is classified as a safe passage area. The second-level quality label is classified as a restricted area. The third-level quality label or the area with obstacles is classified as an absolute prohibited area.

[0065] The A* algorithm is used for path search in the three-dimensional geographic space to obtain the main communication path. A double-time-scale dynamic optimization mechanism is used to generate multiple alternative paths based on the main communication path. The quality scores of the alternative paths are calculated and arranged in descending order of quality scores to obtain a set of communication paths.

[0066] Specifically, the coordinate set of the safe passage area and the restricted area of the integrated map is collected, and the A* algorithm path search is performed. The three-dimensional geographic coordinates of the starting node and the target node are input, and the Euclidean distance and the path loss compensation coefficient (electromagnetic wave propagation characteristic parameter library) are used as the movement cost function. The member nodes are traversed under the constraints of safe passage area first and restricted area second, and the continuous member node sequence with the maximum total comprehensive evaluation value is output as the main communication path.

[0067] A dual-timescale dynamic optimization mechanism is adopted. In the short timescale, path variants are generated through a high-frequency local perturbation mechanism, while in the long timescale, new obstacle-avoidance paths are generated through a low-frequency global replanning mechanism. The short-timescale dynamic optimization mechanism generates a spherical perturbation zone (e.g., radius 50 meters) at the position of each member node in the main communication path with a short period (e.g., period T1 is 1 second). In each perturbation zone, a new three-dimensional coordinate point located in the safe passage zone is randomly selected to replace the original member node position, and several path variants are formed through local node replacement operations.

[0068] The long-term dynamic optimization mechanism re-triggers the A* algorithm to search for paths based on the main communication path at a longer period (e.g., period T2 is 10 seconds). During the new path search process, approximately 80% of the original member node positions in the main communication path are fixed and retained. The path extension is recalculated at the remaining node positions to form obstacle bypass paths. The path variants generated based on the short-term and obstacle bypass paths generated based on the long-term together constitute multiple candidate paths. A quality score is calculated for all generated paths (including the main communication path and multiple candidate paths), and the communication path set is generated by sorting the quality scores from high to low.

[0069] It should be noted that the formula for calculating the quality score is:

[0070] ;

[0071] in, Indicates the first Quality score of each communication path Indicates the first The length of the communication path, Indicates the first The average signal-to-noise ratio of the communication paths, Indicates the first The average availability probability of each communication path Indicates the first The cumulative time taken for each communication path to traverse restricted or absolute no-go zones. Indicates path length weight. Indicates communication quality weights. Indicates spectrum availability weights. Indicates the weight of safe travel time. This represents the communication path index variable.

[0072] S4. Obtain the member node sequence and interference frequency band data by parsing the fields in the communication path set, construct a two-dimensional link-frequency band cost matrix, use the Hungarian algorithm to perform bipartite graph optimal matching of the two-dimensional link-frequency band cost matrix, monitor the link quality to calculate, and obtain the instantaneous bit error rate.

[0073] The communication link sequence field and interference frequency band data in the communication path set are parsed to construct a two-dimensional link-frequency band cost matrix. The communication links are used as the left node set and the available frequency bands are used as the right node set. The Hungarian algorithm is used to traverse all communication links and calculate the matching cost, and the channel allocation mapping table is output.

[0074] Specifically, the source node identifier and destination node identifier information are obtained by parsing the communication link sequence field in the communication path set. At the same time, the center frequency and bandwidth parameters of the disabled frequency band are extracted from the interference frequency band data entity. A two-dimensional link-frequency band cost matrix is ​​constructed, with the row dimension of the two-dimensional link-frequency band cost matrix corresponding to the communication link index and the column dimension corresponding to the available frequency band index. The Hungarian algorithm is used to traverse all rows and columns of the two-dimensional link-frequency band cost matrix, and a three-step loop is performed to determine the minimum value of row reduction, the minimum value of column reduction, and the minimum number of lines covering zero elements until the optimal matching solution is obtained. The channel allocation mapping table is then output.

[0075] The radio frequency front-end of member nodes is reconfigured according to the channel allocation mapping table. The received signal strength, number of symbol errors and tracking attenuation trend are recorded on the communication link. The signal-to-noise ratio is calculated by the instantaneous power ratio formula to obtain monitoring data. The transient bit error rate is calculated by the instantaneous quantization rate formula.

[0076] Specifically, the member node RF front-end is configured according to the frequency band allocation field and power compensation value field of the channel allocation mapping table. The center frequency value of the frequency band is locked (example value 2480 MHz), the bandwidth value is set (example value 20 MHz), and the power compensation value is loaded to the power amplifier control unit. Four-dimensional monitoring is performed on the active communication link to record the received signal strength value (in decibels and milliwatts), the number of symbol errors (comparing demodulated symbols with error correction codebook sequences), and the average received signal power value of three consecutive time slots to obtain the attenuation slope value (in decibels per millisecond). The signal-to-noise ratio value is obtained according to the instantaneous power ratio formula. The monitoring data is then summarized to obtain the monitoring data. The transient bit error rate is calculated based on the monitoring data according to the instantaneous quantization rate formula for bit errors.

[0077] It should be noted that the formula for calculating the transient bit error rate is:

[0078]

[0079] in, Indicates the first On the communication link at a certain time point Transient bit error rate, Indicates the first On the communication link at a certain time point Number of demodulation symbol errors Indicates the first On the communication link at a certain time point total number of received symbols, denotes the received signal power average value, denotes the received signal power average value, denotes the received signal power average value, denotes the time interval between two time slots, denotes the signal attenuation slope influence factor, denotes the received signal power average value, signal-to-noise ratio, signal-to-noise ratio correction coefficient, denotes the time point, denotes the index variable of the communication link.

[0080] S5. Trigger path switching when the instantaneous error rate exceeds the link failure criterion threshold value, and generate a channel allocation instruction set.

[0081] In each communication period, compare the instantaneous error rate with the preset link failure threshold value. When the instantaneous error rate in consecutive communication periods exceeds the link failure criterion threshold value, it is determined that the link is failed.

[0082] Specifically, in each communication period (an example period length of fifty milliseconds), the instantaneous error rate generated by the receiving end physical layer is read and compared with the preset link failure criterion threshold value (an example value of ten to the power of minus three); if the current period instantaneous error rate is greater than the link failure criterion threshold value, the link abnormality counter is activated to perform an increment operation; if the current period instantaneous error rate is less than or equal to the link failure criterion threshold value, the link abnormality counter is reset to zero; when the cumulative value of the link abnormality counter reaches three, a link failure flag state is immediately generated; the determination process needs to meet the continuity condition, if the third period is not exceeded, the counting is interrupted, and only when the instantaneous error rate of three consecutive periods exceeds the link failure criterion threshold value, the link failure is output.

[0083] It should be noted that the theoretical error rate curve of the analysis target modulation and demodulation scheme (for example, QPSK) under the static Gaussian white noise channel is analyzed, and the signal-to-noise ratio inflection point region (an example signal-to-noise ratio of 10 dB corresponding to an error rate of 2.3x0.001) is located; the actual service tolerance test data (for example, the maximum acceptable error rate of video transmission is 0.001) is collected, combined with the historical environmental interference statistical value (for example, the average error rate is raised by 0.8 times due to urban interference), and the fixed link failure criterion threshold value is set through the triple redundancy calibration mechanism.

[0084] The cluster head node activates the communication path set, selects the path with the highest quality score, scans the optimal frequency band of the new communication path, dynamically allocates a non-overlapping frequency band group, and generates a channel allocation instruction set.

[0085] Specifically, when the cluster head node detects the link failure flag state activation, it immediately loads the stored communication path set; parses the path quality score field in the communication path set, selects the path identifier with the highest quality score value; based on the selected alternative path, traverses the spatial coordinate positions of each member node of the new communication path, queries the frequency band availability probability of the corresponding coordinate position in the global frequency spectrum thermal map; for each member node, filter the candidate frequency bands with an availability probability greater than the availability probability threshold (example value 0.9) and a background noise value less than the noise threshold (example value -95 dBm); perform non-overlapping frequency band group allocation: prohibit the allocation of frequency bands with the same center frequency to adjacent member nodes of the path, and force the allocation of orthogonal frequency band groups to member nodes with a distance less than the non-overlapping safety distance (example value 1 km); generate a channel allocation instruction set.

[0086] S6. Switch the transceiver frequency band and adjust the radiation beam direction according to the working frequency, time slot parameter, and power parameter of the channel allocation instruction set, and embed forward error correction code for data transmission. The cluster head node collects transmission status codes in real time and generates a network status communication report by fusing three-dimensional frequency domain-time domain-space domain analysis.

[0087] Based on the channel allocation instruction set, switch the working frequency of the radio frequency front end to the specified working frequency, dynamically calibrate the power amplifier, and adjust the radiation beam direction using a phased array antenna.

[0088] Specifically, according to the working frequency field, power parameter field, and beam pointing angle field of the channel allocation instruction set, perform a three-order hardware operation to lock the phase-locked loop frequency synthesizer of the radio frequency front end to the specified value of the working frequency, and configure the bandwidth to the bandwidth field value (example value 20 MHz) of the channel allocation instruction set; set the reference transmission power level of the power amplifier to the reference power value of the power parameter field, and obtain the distance-related compensation value according to the compensation coefficient value of the power parameter field, output it to the power amplifier bias voltage control end through the digital-to-analog converter to form a closed-loop power calibration; calculate the phase gradient according to the azimuth value and elevation angle value of the beam pointing angle field through the phase shifter array of the phased array antenna, generate a directional radiation beam, and achieve a radiation characteristic with the main lobe pointing to converge within an example value of ±15 degrees of error and a side lobe suppression ratio not less than -25 dB.

[0089] Embed error correction coding, interleaver parameters, and cyclic redundancy check fields in the transmitted data stream and perform data transmission, and package it as a structured state data packet.

[0090] Specifically, the check bit is generated according to the error correction coding scheme (for example, Turbo code rate 2 / 3) selected by the channel allocation instruction set, and is inserted into the redundant bit position of the data stream to be transmitted; the interleaver matrix structure is configured by loading the value specified in the interleaving parameter field of the channel allocation instruction set (for example, block size 128 symbols), and the symbol position replacement operation is performed on the coded data; the cyclic redundancy check field is calculated, and the thirty-two-bit checksum value is formed according to the generating polynomial (for example, CRC-32) for the physical layer frame payload, and is attached to the frame tail; the error correction coded data stream, the interleaver parameter identifier and the cyclic redundancy check field are encapsulated into the physical layer transmission frame data entity according to the frame structure specified by the communication protocol; and the transmission waveform modulation and radio frequency transmission are completed in the time division multiple access time slot window specified by the channel allocation instruction set.

[0091] The cluster head node converges all structured state data packets, analyzes the frequency domain dimension, the time domain dimension and the space domain dimension through three-dimensional situation analysis, obtains the spectrum load heat map data entity, the time slot quality rating matrix and the three-dimensional power distribution cloud map, and integrates the network state communication report.

[0092] Specifically, the cluster head node receives all the structured state data packet entities uploaded by the member nodes (including the timestamp field, the received signal strength value field, the symbol error number field, the attenuation slope value field, the signal-to-noise ratio value field and the instantaneous bit error rate value field); performs three-dimensional situation analysis, processes the frequency domain dimension, scans the working frequency range (for example, 2.4-2.483 GHz), counts the number of active nodes and the total number of nodes at each frequency point, calculates the frequency point occupation rate value, and generates the spectrum load heat map data entity;

[0093] The time domain dimension is processed, the symbol error number value and the signal-to-noise ratio value of all time slots of the time division multiple access frame are extracted, the single-time-slot average instantaneous bit error rate value is calculated, the maximum transmission delay value of the time slot is measured by time stamp difference, and the time slot quality rating matrix is constructed; the space domain dimension is processed, the node power data is aggregated according to the three-dimensional geographic grid unit (for example, 100m cube), the actual transmission power mean value and the variance value in the grid are calculated to generate the three-dimensional power distribution cloud map data entity, and the spectrum load heat map data entity, the time slot quality rating matrix data entity and the three-dimensional power distribution cloud map data entity are integrated to compress and encapsulate the network state communication report.

[0094] The embodiment also provides an unmanned aerial vehicle cluster communication interaction system based on an ad hoc network, comprising: a responsibility airspace module, configured to acquire three-dimensional positioning coordinates of all unmanned aerial vehicles, collect remaining battery percentage data, elect a cluster head node based on a dynamic weight formula, and construct a cluster head responsibility airspace; a heat map module, configured to obtain spectrum feature parameters by scanning electromagnetic wave signals in the cluster head responsibility airspace by using the cluster head node, input the spectrum feature parameters into a spectrum prediction model to generate a weight sensitive state, perform gradient aggregation through gradient-temperature signal conversion and weighted average, and generate a global spectrum heat map; a path module, configured to construct a frequency band value evaluation function based on electromagnetic wave propagation characteristics and spectrum availability data, establish a comprehensive map that fuses electromagnetic characteristics and terrain constraints, generate a main communication path and an alternative path through an A* algorithm combined with a double-time-scale dynamic optimization mechanism, and output a communication path set; a channel allocation module, configured to obtain member node sequences and interference frequency band data by analyzing fields in the communication path set, construct a two-dimensional link-frequency band cost matrix, perform optimal matching of a bipartite graph of the two-dimensional link-frequency band cost matrix by using a Hungarian algorithm, monitor link quality for calculation, obtain an instantaneous bit error rate, trigger path switching when the instantaneous bit error rate exceeds a link failure criterion threshold, and generate a channel allocation instruction set; and a reporting module, configured to switch a transmission frequency band and adjust a radiation beam direction according to working frequency points, time slot parameters and power parameters of the channel allocation instruction set, embed a forward error correction code, and perform data transmission, collect transmission status codes in real time by using the cluster head node, and generate a network status communication report by fusing three-dimensional frequency domain-time domain-space domain analysis.

[0095] The embodiment also provides a computer device suitable for the unmanned aerial vehicle cluster communication interaction method based on the ad hoc network, comprising: a memory and a processor; the memory is configured to store computer executable instructions, and the processor is configured to execute the computer executable instructions to implement the unmanned aerial vehicle cluster communication interaction method based on the ad hoc network proposed in the above embodiment.

[0096] The computer device can be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected by a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner. The wireless manner can be achieved by WIFI, an operator network, NFC (Near Field Communication) or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer overlaid on the display screen, or a key, trackball or touchpad arranged on the shell of the computer device, or an external keyboard, touchpad or mouse, etc.

[0097] The embodiment also provides a storage medium having a computer program stored thereon, the program being executed by a processor to implement the method for realizing unmanned aerial vehicle cluster communication interaction based on an ad hoc network as described above. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as a static random access memory (SRAM), an electrically erasable programmable read-only memory (EEPROM), an erasable programmable read-only memory (EPROM), a programmable read-only memory (PROM), a read-only memory (ROM), a magnetic storage, a flash memory, a magnetic disk or an optical disk.

[0098] To sum up, the present application realizes efficient organization and management of the unmanned aerial vehicle cluster communication topology by dynamically electing cluster head nodes and constructing responsibility airspace, improves the stability and energy efficiency of the network structure, uses the cluster head nodes to scan electromagnetic signals and generate a global spectrum heat map, has real-time sensing and modeling capabilities for complex electromagnetic environments, and significantly improves the utilization efficiency of spectrum resources; in combination with the spectrum heat map and the double-time-scale mechanism to output a communication path set, the flexibility and robustness of communication path planning are enhanced, high-quality communication links can still be maintained in a dynamic environment, the communication reliability and anti-interference capability are effectively improved, and the link failure risk is reduced, thereby providing a solid guarantee for efficient cooperative communication of the unmanned aerial vehicle cluster in complex task scenarios.

[0099] It should be noted that the above examples are only used to illustrate the technical solutions of the present application but not to limit the present application. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or equivalently replaced without departing from the spirit and scope of the present application, and all these modifications and equivalents should be included in the scope of the claims of the present application.

Claims

1. An unmanned aerial vehicle cluster communication interaction method based on an ad hoc network, characterized in that: comprise, acquire three-dimensional positioning coordinates of all unmanned aerial vehicles and percentage of remaining power data, elect cluster head nodes based on a dynamic weight formula, and construct a cluster head responsibility airspace; scan electromagnetic wave signals in the cluster head responsibility airspace using the cluster head nodes to obtain spectral feature parameters, input the spectral feature parameters into a spectral prediction model to generate a weight sensitive state, perform gradient aggregation through gradient-temperature signal conversion and weighted averaging, and generate a global spectral heat map, the specific steps being as follows, scan electromagnetic wave signals in the cluster head responsibility airspace using the cluster head nodes, perform spectral feature analysis on the electromagnetic wave signals to obtain spectral feature parameters, and input the spectral feature parameters into a spectral prediction model, calculate a weight sensitive state through back propagation, encode the weight sensitive state into a temperature fluctuation signal, and perform weighted averaging to obtain an aggregated weight sensitive state; update the spectral prediction model through the aggregated weight sensitive state to obtain an updated spectral prediction model, input three-dimensional positioning coordinates of all unmanned aerial vehicles into the updated spectral prediction model to obtain a frequency band availability probability matrix of the three-dimensional positioning coordinates and convert the frequency band availability probability matrix into a global spectral heat map; construct a frequency band value evaluation function based on electromagnetic wave propagation characteristics and spectral availability data, establish a comprehensive map that integrates electromagnetic characteristics and terrain constraints, generate a main communication path and an alternative path through an A* algorithm combined with a double-time-scale dynamic optimization mechanism, and output a communication path set, the specific steps being as follows, traverse three-dimensional positioning coordinates of each member node in the cluster head responsibility airspace, query the global spectral heat map to obtain frequency band availability probability and background noise corresponding to the position of each member node, and obtain spectral availability data; obtain channel transmission potential based on electromagnetic wave propagation characteristics, construct a frequency band value evaluation function according to spectral availability data, integrate frequency band availability data and channel transmission potential as double factors, obtain a communication quality classification label, and label a high-value communication area; map the communication quality classification label to a three-dimensional geographic space based on the high-value communication area, establish a comprehensive map that integrates electromagnetic characteristics and terrain constraints, and divide the three-dimensional geographic space into a safe passage area, a restricted area, and an absolute forbidden area; perform path search in the three-dimensional geographic space through an A* algorithm to obtain a main communication path, generate multiple alternative paths based on the main communication path using a double-time-scale dynamic optimization mechanism, calculate quality scores of the alternative paths and arrange the alternative paths in descending order of the quality scores to obtain a communication path set; obtain member node sequences and interference frequency band data by analyzing fields in the communication path set, construct a two-dimensional link-frequency band cost matrix, perform optimal matching of a bipartite graph of the two-dimensional link-frequency band cost matrix using a Hungarian algorithm, and calculate link quality to obtain an instantaneous bit error rate, trigger path switching when the instantaneous bit error rate exceeds a link failure criterion threshold, and generate a channel allocation instruction set; switch a transceiving frequency band and adjust a radiation beam direction according to working frequency points, time slot parameters, and power parameters of the channel allocation instruction set, embed a forward error correction code, and perform data transmission, collect transmission status codes in real time through the cluster head nodes, and generate a network status communication report by integrating three-dimensional frequency domain-time domain-space domain analysis. 2.The self-organizing network based UAV swarm communication interaction method of claim 1, wherein: The cluster head responsibility airspace is constructed, and the specific steps are as follows, The dynamic weight score of each unmanned aerial vehicle node is calculated, the node with the highest score is selected as the cluster head node, the remaining unmanned aerial vehicle nodes are selected as candidate nodes, the historical maximum communication distance of the cluster head node is collected, a cluster head responsibility airspace is constructed based on the three-dimensional positioning coordinates of the unmanned aerial vehicle and taking the cluster head node as the center and the historical maximum communication distance as the radius; The Euclidean distance between the candidate node and the cluster head node is calculated based on the remaining percentage of power data, the cluster head node transmits a detection signal, and the received signal strength of the candidate node is measured, and the candidate node with the Euclidean distance less than the historical maximum communication distance and the received signal strength greater than the transmitted detection signal is selected as a member node. 3.The self-organizing network based UAV swarm communication interaction method of claim 1, wherein: The specific steps of obtaining the instantaneous bit error rate are as follows, The communication link sequence field and interference frequency band data in the communication path set are parsed, a two-dimensional link-frequency band cost matrix is constructed, the communication link is taken as the left node set, the available frequency band is taken as the right node set, the Hungarian algorithm is used to traverse all communication links and calculate the matching cost, and a channel allocation mapping table is output; According to the channel allocation mapping table, the member node radio frequency front end is re-distributed, the received signal strength, the number of symbol errors and the tracking attenuation trend on the communication link are recorded, the signal-to-noise ratio is calculated through the instantaneous power ratio formula, the monitoring data is obtained, and the instantaneous bit error rate is calculated through the bit error instantaneous quantization rate formula. 4.The self-organizing network based UAV swarm communication interaction method of claim 1, wherein: The specific steps of triggering path switching when the instantaneous bit error rate exceeds the link failure criterion threshold are as follows, In each communication period, the instantaneous bit error rate is compared with the link failure criterion threshold, and when the instantaneous bit error rate in the continuous communication period exceeds the link failure criterion threshold, it is determined that the link is failed; The cluster head node activates the communication path set, selects the candidate path with the highest quality score, scans the optimal frequency band of the new communication path, dynamically allocates a non-overlapping frequency band group, and generates a channel allocation instruction set. 5.The self-organizing network based UAV swarm communication interaction method of claim 1, wherein: The specific steps of generating a network state communication report are as follows, Based on the channel allocation instruction set, the working frequency of the radio frequency front end is switched to the specified working frequency point, the power amplifier is dynamically calibrated, the phased array antenna adjusts the radiation beam direction, the error correction coding and interleaver parameters are embedded in the transmitted data stream, and the data transmission is performed, and the structured state data packet is packaged; All structured state data packets are converged by the cluster head node, the frequency domain dimension, the time domain dimension and the space dimension are analyzed through three-dimensional situation, the spectrum load thermodynamic diagram data entity, the time slot quality rating matrix and the three-dimensional power distribution cloud diagram are obtained, and the network state communication report is integrated.

6. An unmanned aerial vehicle cluster communication interaction system based on an ad hoc network, based on the unmanned aerial vehicle cluster communication interaction method based on an ad hoc network in any one of claims 1-5, characterized in that: It comprises, The responsibility airspace module is used for acquiring the three-dimensional positioning coordinates of all unmanned aerial vehicles, collecting the remaining percentage of power data, electing the cluster head node based on the dynamic weight formula, and constructing the cluster head responsibility airspace; The thermodynamic diagram module is used for scanning the electromagnetic wave signal in the cluster head responsibility airspace by the cluster head node to obtain the spectrum feature parameters, inputting the spectrum feature parameters into a spectrum prediction model to generate a weight sensitive state, performing gradient aggregation through gradient-temperature signal conversion and weighted average, and generating a global spectrum thermodynamic diagram. A path module is configured to construct a frequency band value evaluation function based on electromagnetic wave propagation characteristics and spectrum availability data, and to establish a comprehensive map integrating electromagnetic characteristics and terrain constraints, to generate a main communication path and an alternative path by an A* algorithm combined with a double-time-scale dynamic optimization mechanism, and to output a communication path set; A channel allocation module is configured to obtain a member node sequence and interference frequency band data by analyzing fields in the communication path set, to construct a two-dimensional link-frequency band cost matrix, to perform optimal matching of a bipartite graph of the two-dimensional link-frequency band cost matrix by using a Hungarian algorithm, to monitor link quality for calculation, to obtain an instantaneous bit error rate, to trigger path switching when the instantaneous bit error rate exceeds a link failure criterion threshold, and to generate a channel allocation instruction set; A report module is configured to switch a transceiving frequency band and adjust a radiation beam direction according to operating frequency points, time slot parameters, and power parameters of the channel allocation instruction set, to embed a forward error correction code for data transmission, to collect transmission status codes in real time by a cluster head node, and to generate a network status communication report by integrating three-dimensional frequency domain-time domain-space domain analysis. 7.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is characterized in that: The processor executes the computer program to implement the steps of the self-organizing network-based unmanned aerial vehicle cluster communication interaction method of any one of claims 1-5.

8. A computer readable storage medium having stored thereon a computer program, characterized in that: The computer program is executed by the processor to implement the steps of the self-organizing network-based unmanned aerial vehicle cluster communication interaction method of any one of claims 1-5.

Images