An unmanned aerial vehicle anti-jamming system based on spectrum sensing

Abstract

The present application relates to the technical field of unmanned aerial vehicle communication anti-jamming, and particularly relates to an unmanned aerial vehicle anti-jamming system based on spectrum sensing. A spectrum sensing data acquisition module acquires multi-band spectrum signal data to generate a spectrum interference characteristic spectrum; a jamming source positioning module calculates a jamming source azimuth and distance based on the spectrum interference characteristic spectrum to generate a spatial distribution data set; a three-dimensional interference field construction module combines real-time altitude and geographic coordinates of the unmanned aerial vehicle to construct a three-dimensional interference field model containing a jamming intensity gradient and a propagation direction; an anti-jamming strategy generation module analyzes communication link vulnerability to dynamically adjust band switching priority and transmission power allocation weight; and a beamforming control module calculates antenna array phase offset and beam pointing angle to generate an adaptive beamforming parameter set. The system can improve the anti-jamming capability and communication stability of the unmanned aerial vehicle in a complex electromagnetic environment.

Description

Technical Field

[0001] This invention relates to the field of anti-jamming technology for unmanned aerial vehicle (UAV) communication, and particularly to an anti-jamming system for UAVs based on spectrum sensing. Background Technology

[0002] With the rapid popularization of drone technology in civilian and industrial applications, the stability and security of its communication links have gradually become key issues restricting the expansion of applications. Currently, drones mainly rely on wireless communication in specific frequency bands to achieve data transmission and flight control. However, in complex electromagnetic environments, the presence of various unintentional and malicious interference signals can easily lead to communication interruptions, data loss, and even serious accidents such as drone crashes due to loss of control.

[0003] Existing anti-jamming technologies for drones mostly employ fixed frequency band switching, spread spectrum communication, or power enhancement, but these methods have significant limitations. Fixed frequency band switching relies on a preset frequency band list and cannot detect dynamic changes in the spectrum in the environment in real time. When all preset frequency bands are interfered with, the anti-jamming capability instantly fails. While spread spectrum communication can improve anti-jamming stability, it increases equipment complexity and energy consumption, and in environments with strong interference, communication speed and signal quality still drop significantly. Power enhancement methods may be limited by exceeding electromagnetic radiation standards and cannot solve the communication blockage problem caused by multiple interference sources.

[0004] Current technologies lack the ability to accurately locate interference sources and effectively model interference fields, making it impossible to determine the location, distance, and intensity distribution of interference sources. This results in a lack of targeted anti-interference strategies. For example, when multiple interference sources exist within the UAV's flight area, current technologies cannot distinguish the impact range of different sources, forcing the adoption of uniform anti-interference measures. This makes it difficult to accurately avoid key interference areas, leading to a waste of anti-interference resources and failing to guarantee the stable flight and communication needs of UAVs in complex interference environments.

[0005] As drone applications expand into complex electromagnetic environments such as densely populated urban areas and industrial plants, existing anti-interference technologies are no longer sufficient to meet practical application needs. There is an urgent need for an integrated solution that can sense spectrum dynamics in real time, accurately locate interference sources, construct interference field models, and dynamically generate anti-interference strategies to improve the survivability and communication stability of drones in complex electromagnetic environments. Summary of the Invention

[0006] The main objective of this invention is to provide a spectrum-aware anti-jamming system for unmanned aerial vehicles (UAVs), aiming to solve the technical problems in the prior art.

[0007] This invention proposes a spectrum-aware anti-jamming system for unmanned aerial vehicles (UAVs), comprising:

[0008] The spectrum sensing data acquisition module acquires spectrum signal data of multiple frequency bands in the UAV's flight area, and generates a spectrum interference characteristic spectrum by analyzing the matching degree between the spectrum signal intensity change rate and historical interference patterns.

[0009] Based on the aforementioned spectral interference feature spectrum, the interference source localization module extracts interference signal intensity distribution data, calculates the azimuth and distance parameters of the interference source relative to the UAV, and generates a spatial distribution dataset of the interference source.

[0010] The three-dimensional interference field construction module constructs a three-dimensional interference field model that includes the interference intensity gradient and propagation direction based on the spatial distribution dataset of the interference source and the real-time flight altitude and geographic coordinate data of the UAV.

[0011] The anti-interference strategy generation module analyzes the vulnerability index of the UAV communication link in the interference field based on the three-dimensional interference field model, dynamically adjusts the frequency band switching priority and the transmission power allocation weight, and generates an anti-interference strategy dataset.

[0012] The beamforming control module calculates the phase offset parameters and beam pointing angle of the UAV communication antenna array based on the anti-interference strategy dataset, and generates an adaptive beamforming parameter set.

[0013] Preferably, the spectral interference feature spectrum includes the interference signal intensity change rate, frequency band occupancy anomaly index, and historical interference mode matching coefficient; the interference source spatial distribution dataset includes interference source azimuth angle estimation, interference source distance parameter, and interference signal intensity spatial gradient; the three-dimensional interference field model includes interference intensity isosurface data, interference propagation direction vector, and spatial vulnerability distribution map; the anti-interference strategy dataset includes priority communication frequency band sequence, transmit power graded weight, and link redundancy backup scheme; and the adaptive beamforming parameter set includes beam main lobe pointing angle, beam null formation parameter, and array phase compensation value.

[0014] Preferably, the spectrum sensing data acquisition module includes:

[0015] The spectrum signal strength analysis submodule acquires real-time sampling data from multi-band spectrum monitoring equipment in the UAV flight area, calculates the rate of change of spectrum signal strength within each sampling time window, and obtains dynamic change data of spectrum signal by performing matching degree analysis with the historical interference pattern database.

[0016] The frequency band occupancy assessment submodule calculates the ratio of signal occupancy duration to idle duration for each frequency band based on the dynamic change data of the spectrum signal, and judges the abnormal state of frequency band occupancy by combining the threshold values, and obtains a set of abnormal frequency band occupancy indicators.

[0017] The interference feature extraction submodule integrates the spectrum signal strength change rate data with the historical interference pattern matching results based on the frequency band occupancy anomaly index set to generate a spectrum interference feature spectrum.

[0018] Preferably, the interference source localization module includes:

[0019] The interference signal intensity distribution analysis submodule receives the spectral interference feature spectrum, extracts the peak point data of the interference signal intensity in the feature spectrum and its corresponding frequency band information, calculates the signal intensity attenuation gradient at different monitoring points, and obtains the interference signal intensity distribution data.

[0020] Based on the interference signal intensity distribution data, the multi-source positioning calculation submodule uses a time difference of arrival and frequency difference of arrival fusion positioning algorithm to calculate the azimuth and distance parameters of the interference source relative to the UAV platform and obtain preliminary positioning results.

[0021] The spatial distribution integration submodule integrates multiple sets of azimuth and distance parameters from the preliminary positioning results, and generates a spatial distribution dataset of interference sources through spatial coordinate transformation and data fusion.

[0022] Preferably, the three-dimensional interference field construction module includes:

[0023] The spatial interpolation calculation submodule obtains the coordinates and intensity data of the interference sources in the spatial distribution dataset of the interference sources. Combined with the real-time flight altitude and geographic coordinate information of the UAV, it uses a three-dimensional spatial interpolation algorithm to generate a continuous interference intensity distribution field and obtain interference intensity isosurface data.

[0024] The propagation model integration submodule, based on the interference intensity isosurface data, imports the electromagnetic wave propagation attenuation model and multipath effect parameters, calculates the propagation direction and attenuation gradient of the interference signal in three-dimensional space, and generates a set of interference propagation direction vectors.

[0025] The vulnerability mapping submodule integrates the interference intensity isosurface data and the interference propagation direction vector set, analyzes the signal-to-interference ratio parameters of the UAV communication link at each spatial point, constructs a spatial vulnerability distribution map, and finally generates a three-dimensional interference field model.

[0026] Preferably, the anti-interference strategy generation module includes:

[0027] The link vulnerability assessment submodule receives the spatial vulnerability distribution map in the three-dimensional interference field model, extracts the data of the lowest signal-to-interference ratio point on the current flight trajectory of the UAV, calculates the communication link interruption probability and capacity degradation index, and obtains the link vulnerability assessment result.

[0028] Based on the link vulnerability assessment results, the frequency band switching decision submodule analyzes the interference intensity distribution of each candidate frequency band in the three-dimensional interference field, calculates the frequency band availability score, and generates a priority communication frequency band sequence.

[0029] The power allocation optimization submodule calculates the power allocation weight and link redundancy backup scheme for each frequency band based on the priority communication frequency band sequence and the UAV transmission power constraints, and generates an anti-interference strategy dataset.

[0030] Preferably, the beamforming control module includes:

[0031] The beam pointing calculation submodule obtains the priority communication frequency band sequence and power allocation weight in the anti-interference strategy dataset. Based on the azimuth information in the spatial distribution dataset of interference sources, it calculates the interference direction that the antenna array needs to avoid and the communication direction that needs to be enhanced, and obtains the beam main lobe pointing angle parameter.

[0032] The null formation control submodule calculates the antenna array weighting coefficients based on the beam main lobe pointing angle parameter and the interference propagation direction vector in the three-dimensional interference field model to form beam nulls in the interference direction, thereby obtaining the beam null formation parameter set.

[0033] The phase compensation adjustment submodule integrates the beam main lobe pointing angle parameter and the beam null formation parameter set, calculates the phase compensation value and amplitude weighting value of each element of the antenna array, and generates an adaptive beamforming parameter set.

[0034] Preferably, the system further includes:

[0035] The real-time spectrum monitoring module continuously collects spectrum signal data of the UAV's flight area, compares and analyzes it with the spectrum interference feature spectrum, detects newly emerging interference signal features, and generates real-time interference feature update data.

[0036] The dynamic strategy adjustment module updates the data based on the real-time interference characteristics, reassesses the link vulnerability in conjunction with the three-dimensional interference field model, and updates the frequency band switching priority and power allocation weight in the anti-interference strategy dataset.

[0037] Preferably, the system further includes:

[0038] The communication quality assessment module monitors the actual bit error rate and throughput parameters of the UAV communication link, compares them with the communication performance predicted based on the anti-interference strategy dataset, and generates communication quality deviation data.

[0039] The parameter optimization feedback module adjusts the interference pattern matching threshold in the spectrum interference feature spectrum and the positioning accuracy parameter in the interference source positioning module based on the communication quality deviation data, thereby optimizing the construction accuracy of the three-dimensional interference field model.

[0040] Preferably, the system further includes:

[0041] The multi-UAV collaborative module acquires the spectrum interference characteristic spectrum and spatial distribution dataset of interference sources of each UAV in the cluster, generates a cluster collaborative interference map through data fusion, and uniformly adjusts the anti-interference strategy dataset and adaptive beamforming parameter set of each UAV.

[0042] The beneficial effects of this invention are as follows:

[0043] This spectrum-aware anti-jamming system for unmanned aerial vehicles (UAVs) achieves comprehensive perception of spectrum dynamics and interference sources in the UAV flight environment through the collaborative work of multiple modules, significantly improving anti-jamming capabilities and communication stability.

[0044] The spectrum sensing data acquisition module can acquire spectrum signal data from multiple frequency bands in real time, and generate a spectrum interference feature spectrum by analyzing the matching degree between the spectrum signal intensity change rate and historical interference patterns. This breaks the limitation of existing technologies that rely on preset frequency bands, and can dynamically capture the changing trend of spectrum interference in the environment, identify potential interference risks in advance, and provide accurate spectrum information support for the formulation of subsequent anti-interference strategies, avoiding the problem of anti-interference failure due to the inability to perceive dynamic changes in the spectrum.

[0045] The interference source localization module extracts interference signal intensity distribution data based on the spectral interference characteristic spectrum, calculates the azimuth and distance parameters of the interference source relative to the UAV, and generates a spatial distribution dataset of the interference source, solving the problem of existing technologies being unable to accurately locate interference sources. By clearly defining the spatial location information of the interference source, the UAV can clearly understand the distribution of surrounding interference sources, no longer blindly adopting uniform anti-interference measures. It can adjust its flight path and communication parameters according to the azimuth and distance of the interference source, reducing ineffective anti-interference operations and improving anti-interference efficiency.

[0046] The 3D interference field construction module combines real-time UAV flight altitude and geographic coordinate data to construct a 3D interference field model that includes interference intensity gradients and propagation directions, presenting the impact of interference sources in a visualized and quantifiable manner. This model can intuitively reflect the differences in interference intensity and the propagation patterns of interference signals at different spatial locations, enabling UAVs to accurately assess the severity of their interference environment and the interference risk level in different areas. This provides comprehensive field information for the dynamic adjustment of subsequent anti-interference strategies, avoiding the problem of untargeted anti-interference strategies due to insufficient understanding of the interference field.

[0047] The anti-interference strategy generation module analyzes the vulnerability indicators of the UAV communication link based on a three-dimensional interference field model, dynamically adjusts the frequency band switching priority and transmit power allocation weight, and generates an anti-interference strategy dataset. Compared with the fixed anti-interference mode of existing technologies, this module can formulate differentiated anti-interference strategies according to the actual situation of the interference field. For example, in areas with low interference intensity, the transmit power is appropriately reduced to reduce energy consumption; when the interference intensity is high and there are multiple selectable frequency bands, the switching priority is adjusted according to the degree of interference of the frequency band to ensure that the communication link always selects the optimal frequency band, achieving a balance between anti-interference effect and equipment energy consumption, while avoiding the problem of excessive electromagnetic radiation caused by blindly increasing power.

[0048] The beamforming control module calculates the phase offset parameters and beam pointing angle of the antenna array based on the anti-interference strategy dataset, generating an adaptive beamforming parameter set. This enables the signal energy of the UAV communication antenna to be concentrated towards the target receiver while suppressing signal interference from the direction of interference sources. This directional communication method significantly improves the anti-interference capability of the communication link. Even in environments with multiple interference sources, it can effectively isolate interference signals and ensure stable transmission of communication data. It solves the problem of communication quality degradation in existing spread spectrum communication under strong interference environments, further enhancing the communication stability and flight safety of UAVs in complex electromagnetic environments. Attached Figure Description

[0049] Figure 1 This is a timing diagram of a spectrum-aware anti-jamming system for unmanned aerial vehicles (UAVs) according to the present invention.

[0050] Figure 2 Create a relational flowchart for each core dataset;

[0051] Figure 3 This is a graph showing the characteristics of spectral interference.

[0052] Figure 4 This is a spatial distribution and gradient map of the interference sources.

[0053] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0054] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0055] like Figure 1As shown, this application provides a spectrum-sensing-based UAV anti-jamming system, including a spectrum sensing data acquisition module, an interference source localization module, a three-dimensional interference field construction module, an anti-jamming strategy generation module, and a beamforming control module. The spectrum sensing data acquisition module acquires real-time spectrum signal data through multi-band spectrum monitoring equipment deployed in the UAV's flight area. This module performs quantitative analysis on the rate of change of spectrum signal intensity and calculates the matching degree with a historical interference pattern database, generating a spectrum interference feature spectrum containing the rate of change of interference signal intensity, frequency band occupancy anomaly index, and historical interference pattern matching coefficients. The interference source localization module extracts interference signal intensity distribution data based on the spectrum interference feature spectrum and uses a time difference of arrival (TDOA) and frequency difference of arrival (FDI) fusion localization algorithm to calculate the azimuth and distance parameters of the interference source relative to the UAV, forming a spatial distribution dataset of the interference source containing the interference source azimuth estimate, interference source distance parameters, and spatial gradient of interference signal intensity. The three-dimensional interference field construction module combines the UAV's real-time flight altitude and geographic coordinate data, processes the spatial distribution dataset of the interference source using a three-dimensional spatial interpolation algorithm and an electromagnetic wave propagation attenuation model, and constructs a three-dimensional interference field model containing interference intensity isosurface data, interference propagation direction vector, and spatial vulnerability distribution map. The anti-jamming strategy generation module analyzes the signal-to-interference ratio (SIR) parameters of the UAV communication link in the 3D interference field model, dynamically calculates the frequency band availability score and power allocation constraints, and generates an anti-jamming strategy dataset containing priority communication frequency band sequences, transmit power grading weights, and link redundancy backup schemes. The beamforming control module calculates the phase offset and beam pointing parameters of the antenna array based on the anti-jamming strategy dataset, generating an adaptive beamforming parameter set containing the beam main lobe pointing angle, beam null formation parameters, and array phase compensation values. Finally, adaptive beamforming is implemented through the antenna array controller.

[0056] In one embodiment, Example 1: See Figure 2The generation of the spectrum interference characteristic spectrum is a multi-step data processing and analysis process. This system utilizes multiple fixed and airborne spectrum monitoring nodes deployed in the UAV flight area, working collaboratively. These nodes cover civilian and industrial frequency bands from 400MHz to 6GHz. Each node is equipped with a software-defined radio architecture receiver, capable of capturing broadband spectrum data at a high sampling rate. The spectrum signal strength analysis submodule continuously samples the received signal strength indication value for each frequency band using a 10-millisecond basic time window. It quantifies the instantaneous change in signal strength using a differential calculation method, specifically calculating the difference between the arithmetic mean of the signal strength within the current time window and the average value of the previous time window, then dividing by the time interval to obtain the rate of change. The historical interference pattern database pre-stores intensity change templates for various typical interference signals, including the linear change characteristics of swept-frequency interference, the steady-state characteristics of fixed-frequency interference, and the burst peak characteristics of pulse interference. The matching degree analysis uses an algorithm based on the Pearson correlation coefficient, calculating the similarity between the real-time acquired intensity change sequence and the template sequence in the database to generate a dynamically changing data sequence.

[0057] The frequency band occupancy assessment submodule performs binary classification of the signal status within each frequency band using a 1-second statistical period. A signal strength exceeding the noise threshold set by the International Telecommunication Union (ITU) is considered occupied; otherwise, it is considered idle. This module cumulatively calculates the ratio of occupied duration to idle duration for each frequency band per unit time, while also referencing the legally mandated maximum occupancy standard for that band. A frequency band occupancy anomaly flag is triggered when the measured ratio exceeds 20% of the standard value. For abnormal frequency bands, the module records the anomaly start time, duration, and maximum exceedance, forming a structured set of anomaly indicators. The interference feature extraction submodule employs data fusion technology to integrate spectral signal strength change rate data, historical interference pattern matching coefficients, and frequency band occupancy anomaly indicators. The strength change rate data provides the dynamic characteristics of interference, the matching coefficients reflect the similarity of interference types, and the anomaly indicators characterize the availability status of the frequency band. All data is encapsulated into a unified JSON format data packet, containing fields such as timestamp, geographic coordinates, frequency band identifier, strength change rate value, matching coefficient, anomaly status flag, and confidence level, constituting a complete spectral interference feature spectrum. The feature spectrum is transmitted to the subsequent processing module via an encrypted data link.

[0058] At the hardware implementation level, the spectrum monitoring nodes employ a multi-channel receiver design, with each channel independently monitoring a specific frequency band. Time-division multiplexing is used between channels to avoid mutual interference. The node's built-in preprocessing unit handles signal detection, strength calculation, and preliminary filtering, reducing the computational load on the central processing unit. Data time synchronization is achieved through a GPS timing module, with sampling time deviations across all monitoring nodes controlled within microseconds to ensure data comparability across nodes. The historical interference pattern database adopts a cloud-edge collaborative architecture. The cloud stores a large number of historical interference cases and feature templates, while edge nodes cache commonly used templates and update them regularly, ensuring both real-time matching analysis and template completeness.

[0059] The data processing algorithm employs a sliding window mechanism, where newly arriving sampled data continuously replaces the oldest data, maintaining a rolling update of the time window. The rate of change calculation incorporates smoothing filtering and uses a moving average algorithm to eliminate spikes in random noise. Matching degree analysis sets dynamic thresholds, automatically adjusting the tolerance range of the correlation coefficient based on the signal-to-noise ratio (SNR) level, relaxing matching requirements in low SNR environments to avoid missed detections. The anomaly detection algorithm uses multi-feature joint judgment, considering not only the time occupied but also auxiliary indicators such as signal bandwidth and modulation identification features to improve false alarm suppression capabilities. The entire feature extraction process adopts a pipelined architecture, with data acquisition, intensity calculation, rate of change analysis, pattern matching, and anomaly assessment executed in parallel. Data transfer is achieved through a circular buffer, ensuring processing latency is controlled within 50 milliseconds. The data structure of the spectral interference feature spectrum is carefully designed, using a layered nested JSON format, with the top layer containing a metadata layer and a measurement data layer. The metadata layer records system information such as data packet version, data source identifier, and time reference system. The measurement data layer contains multiple frequency band data objects, each of which records in detail the center frequency, bandwidth, intensity statistics, rate of change sequence, matching coefficient list, and abnormal status code for that frequency band. Data encoding uses a binary compression format to reduce transmission overhead, and cyclic redundancy check codes are added to important fields to ensure data integrity. The feature spectrum update frequency is synchronized with the acquisition window, generating a new data packet every 10 milliseconds, but full data transmission is only triggered when a significant change is detected; otherwise, only difference identifiers are sent to save bandwidth.

[0060] See Figure 3 The diagram showcases the key outputs of the spectrum sensing data acquisition module. The left figure displays the change in signal strength rate over time across different frequency bands, revealing distinct interference patterns in certain bands. For example, swept-frequency interference exhibits a linearly increasing rate of change, while impulse interference displays sudden spikes. The right figure shows the occupancy anomaly index and historical interference pattern matching coefficients for each frequency band. High anomaly indices indicate potential interference in that band, while high matching coefficients indicate a high similarity between the detected signal and historical interference patterns.

[0061] In one embodiment, Example 2: The implementation of the interference source localization module is based on the input of the spectral interference feature spectrum. This module achieves accurate spatial localization of the interference source through multi-level signal processing and spatial calculation. The interference signal strength distribution analysis submodule first parses the input spectral interference feature spectrum data packet, extracts the spectral peak points marked as interference in each frequency band, and uses an adaptive threshold double sliding window algorithm for peak detection. The main window detects the signal strength while the reference window evaluates the background noise level. When the average strength of the main window exceeds the average strength of the reference window by 20 dB, it is confirmed as a valid peak. The signal strength attenuation gradient calculation between different monitoring points adopts a logarithmic distance path loss model, considering the known geographical coordinates of the monitoring nodes, and calculates the attenuation slope based on the signal strength difference to establish a mapping relationship between signal strength and propagation distance.

[0062] The multi-source positioning calculation submodule employs a positioning mechanism that fuses time difference of arrival (TDOA) and frequency difference of arrival (FDR). TDOA measurement relies on a high-precision time synchronization system, with all monitoring nodes equipped with GPS timing modules, achieving a time synchronization accuracy of 10 nanoseconds. By comparing the timestamp differences of interference signals arriving at different monitoring nodes, a hyperbolic position equation with time difference as the variable is constructed. FDR measurement is based on the Doppler frequency shift principle, utilizing the frequency shift generated by the relative motion between the monitoring node and the interference source, combined with the node's own motion state data to solve the frequency difference equation. The positioning solution process uses a weighted least squares optimization algorithm, transforming the time difference and frequency difference measurements into a nonlinear system of equations about the interference source's location. Iterative calculations yield the three-dimensional coordinate solution of the interference source relative to the UAV platform, considering the influence of environmental factors such as atmospheric refraction and terrain obstruction on the measured values. The spatial distribution integration submodule processes multiple sets of preliminary positioning results. First, it transforms the positioning data from the relative coordinate system to the geodetic coordinate system using a UTM projection algorithm, combined with elevation correction parameters to ensure coordinate accuracy. Data fusion employs an improved Kalman filter algorithm. The filtered state variables include the 3D position coordinates and velocity of the interference source. Multiple sets of positioning data are fused during the measurement update phase, and the process model considers possible motion patterns of the interference source. A constant velocity model is used for stationary interference sources, while an adaptively adjusted process noise covariance is employed for moving interference sources. The final generated spatial distribution dataset of the interference source includes its latitude and longitude coordinates, elevation information, signal strength spatial gradient, and location reliability index. The data is stored in a hierarchical structure to support rapid retrieval and access by subsequent modules.

[0063] In terms of hardware architecture, the monitoring network adopts a heterogeneous node deployment strategy. Fixed monitoring nodes are placed at known coordinate points to provide a reference, while airborne mobile nodes extend the spatial sampling range. Each node is equipped with a multi-band directional antenna array, and the antenna beams can be electronically scanned to enhance signal reception in specific directions. The time synchronization system adopts a master-slave architecture. The master node obtains an absolute time reference via GPS, while slave nodes maintain microsecond-level time alignment through a wireless clock synchronization protocol. Data transmission uses frequency hopping spread spectrum technology to resist interference, and forward error correction coding is added to important data to ensure transmission reliability.

[0064] The signal processing algorithm employs a multi-modal fusion strategy. For strong signal interference sources, it prioritizes time-of-arrival (TOA) positioning to provide high-precision location estimation. For weak signals or transient interference, it relies on frequency-of-arrival (FOA) measurements to improve detection sensitivity. A peak correlation algorithm addresses signal aliasing in multi-interference-source scenarios by classifying peak values ​​detected by different monitoring nodes into the same interference source through time-frequency feature matching. The localization solution incorporates the RANSAC algorithm to eliminate outlier measurements, improving the robustness of the localization results. Computational complexity management utilizes a partitioned processing mechanism, dividing the monitored airspace into several grids and prioritizing the processing of areas with the highest signal strength. The data management system implements a dynamic update mechanism, storing the spatial distribution dataset of interference sources in a time-series database, recording the trajectory history of each interference source. Data indexing is based on a geospatial database, supporting rapid querying of interference source distribution by region. The confidence management module calculates the reliability of each interference source location based on the positioning geometric accuracy factor, signal strength, and measurement consistency; low-confidence targets trigger a re-measurement process.

[0065] See Figure 4 The diagram illustrates the output of the interference source localization module. The left image shows the spatial distribution of interference sources on a two-dimensional plane, with contour lines representing the isopleths of interference intensity, clearly showing the location and influence range of different interference sources. The UAV is located at the center of the coordinate system, surrounded by multiple interference sources, their size representing the interference intensity. The right image uses polar coordinates to display the spatial gradient distribution of the interference signal intensity; the height of the bars in different directions represents the interference intensity gradient in that direction, helping to determine the main direction of interference propagation.

[0066] In one embodiment, Example 3: The implementation of the 3D interference field construction module is based on a dataset of spatial distribution of interference sources and real-time telemetry data from UAVs. This module constructs an accurate 3D representation of the interference environment through multi-stage spatial modeling and electromagnetic analysis. The spatial interpolation calculation submodule first receives discrete interference source data from the positioning module. This data includes the latitude and longitude coordinates, elevation information, and signal strength values ​​of five main interference sources, ranging from -67dBm to -72dBm. The module uses the Kriging spatial interpolation algorithm to establish a spatial correlation model based on a semi-variogram, predicting the interference intensity in unsampled areas by calculating the spatial autocorrelation characteristics between known monitoring points. The interpolation grid is set to 15 meters resolution in the east-west direction, 15 meters resolution in the south-north direction, and 3 meters resolution in the vertical direction, covering a 500m × 500m × 150m 3D airspace. The terrain data integrates a digital elevation model and a 3D building model, with elevation data accuracy reaching 0.5 meters. The building outline data comes from a geographic information system database. The disturbance intensity isosurface is generated using the moving cube algorithm, which tracks isosurfaces with intensity values ​​of -70dBm, -75dBm, and -80dBm in a three-dimensional voxel mesh. Each isosurface is represented by a triangular mesh containing vertex coordinates and normal vector information.

[0067] The propagation model integration submodule imports a calibrated electromagnetic wave propagation model that incorporates free-space path loss, terrain diffraction loss, and building penetration loss. Free-space loss is calculated using the standard Friis formula; terrain diffraction loss is calculated based on a blade diffraction model; and building penetration loss is set with different attenuation values ​​according to the material database: 12 dB / m for concrete structures, 6 dB / m for glass curtain walls, and 3 dB / m for wood structures. Multipath effect modeling employs ray tracing, considering direct, reflected, and diffracted paths, with a maximum reflection order of 3. Diffraction edge identification is based on building geometry. Interference propagation direction vector calculation uses a three-dimensional gradient operator to calculate the spatial derivative of the intensity field on a regular grid. The vector direction points to the direction of the fastest intensity increase, and the vector magnitude represents the rate of intensity change. The calculation results are stored in vector field form.

[0068] The vulnerability mapping submodule calculates the communication quality index for each point in space. The signal-to-interference ratio (SIR) is calculated based on the following model:

[0069]

[0070] in: Point The letter from the place is more than that. Indicates the drone's transmission power. and These represent the transmit and receive antenna gains, respectively. Indicates the path loss factor. This indicates the interference intensity at that point. This represents thermal noise power. Useful signal power is calculated using models of UAV transmit power, antenna gain, and path loss, while interference signal power is derived from interference intensity isosurface data. The vulnerability index mapping employs a sigmoid function, converting the signal-to-interference ratio (SIR) to a normalized value between 0 and 1. When the SIR is below -3dB, the vulnerability index is greater than 0.8; when the SIR is above 20dB, the vulnerability index is less than 0.2. In the data processing flow, the module first performs coordinate system normalization, transforming all data to a local coordinate system with the mission area center as the origin. Spatial interpolation calculations employ a multi-resolution strategy, using a 5-meter fine grid for the near-field region and a 20-meter coarse grid for the far-field region. The propagation model calculation introduces a parallel processing mechanism, dividing the three-dimensional spatial domain into multiple sub-regions for simultaneous calculation. Vulnerability mapping utilizes real-time rendering technology, leveraging a graphics processor to accelerate full-field calculations. Data storage adopts a hierarchical structure; raw data is stored in NetCDF format, and processed results are stored in HDF5 format. Model updates employ an incremental update strategy, updating the affected area only when the interference environment changes. The entire construction process took approximately 2.5 seconds, and the generated 3D interference field model contained complete data of 5 million grid points.

[0071] Taking a drone swarm collaborative operation scenario in an urban industrial area as an example, three drones are performing equipment inspection tasks at an altitude of 80-120 meters. When the drone swarm flies to the area of ​​31.2457°N, 121.5012°E, the system detects abnormal signals in multiple frequency bands. The spectrum sensing data acquisition module reports interference signals with strengths exceeding -70dBm in the 1.2GHz, 1.8GHz, and 2.4GHz frequency bands, with the signal strength in the 2.4GHz band reaching -62dBm. The interference source localization module calculates, using a multi-point localization algorithm, that the main interference source is located on the roof of a steel structure factory building, with coordinates of 31.2462°N, 121.5018°E, and an altitude of 15 meters. The three-dimensional interference field construction module immediately initiates the processing flow, and the spatial interpolation calculation submodule first obtains the spatial distribution data of the four interference sources. These interference sources are distributed within a 200-meter radius centered on the factory building, with signal strengths ranging from -62dBm to -78dBm. The module employs the Kriging space interpolation algorithm to construct a 3D mesh with a resolution of 10m × 10m × 2m, covering the UAV operating area (500m × 500m × 150m). Digital elevation model data shows the presence of buildings ranging from 5m to 25m in height in the area, and this data is incorporated into the interpolation calculation. The calculated interference intensity field contains 125,000 grid points. The intensity isosurface shows a clear intensity gradient in the vicinity of the interference source, with the highest intensity of -62dBm appearing on the factory roof, gradually decreasing in intensity with increasing distance.

[0072] The propagation model integration submodule imports an electromagnetic wave propagation attenuation model, which comprehensively considers free-space loss and building obstruction effects. For steel structure factory areas, the signal attenuation coefficient is set to 3.2; for open areas, the attenuation coefficient uses a standard free-space model. Multipath effect parameters are set based on the building material database: the reflection coefficient of concrete structures is 0.7, and the reflection coefficient of glass curtain walls is 0.4. The calculated interference propagation direction vector shows that the main interference energy propagates radially along the factory building, with significant reflected waves generated in the southeast direction due to the influence of tall buildings. The propagation vector data is stored in the form of a vector field, containing the propagation direction and attenuation rate for each grid point. The vulnerability mapping submodule begins calculating the signal-to-interference ratio (SIR) parameters for each point in space. The UAV communication system operates in the 2.4 GHz band with a transmit power of 27 dBm and an antenna gain of 3 dBi. The system noise floor is set to -105 dBm. Calculation results show that within 50 meters of the interference source, the signal-to-interference ratio (SIR) is below -5 dB; within 100-200 meters, the SIR increases to 5-15 dB; and beyond 300 meters, the SIR reaches over 20 dB. The vulnerability index mapping uses a piecewise function: areas with an SIR below 0 dB are marked as high-vulnerability areas (index > 0.7), areas with an SIR of 5-15 dB are medium-vulnerability areas (index 0.3-0.7), and areas with an SIR above 20 dB are low-vulnerability areas (index < 0.3). The final generated spatial vulnerability distribution map shows a red high-vulnerability area within an 80-meter radius centered on the interference source, which completely covers the UAV's planned flight path.

[0073] The anti-interference strategy generation module formulates countermeasures based on a 3D interference field model. The link vulnerability assessment submodule identifies three highly vulnerable areas along the UAV's current flight path, with a total length of approximately 120 meters. Communication link interruption probability assessment shows an interruption probability exceeding 60% in these areas, and capacity degradation indicators suggest a potential data transmission rate reduction of over 80%. The frequency band switching decision submodule analyzes the performance of each candidate frequency band. While the 1.2GHz band has relatively weak interference, its bandwidth is limited; the 1.8GHz band is significantly affected by building attenuation; and the 5.8GHz band has sufficient available bandwidth but limited propagation distance. The calculated frequency band availability scores are: 5.8GHz band 85 points, 1.2GHz band 75 points, and 2.4GHz band 30 points. The power allocation optimization submodule formulates a power allocation scheme based on the frequency band scores: 60% of the transmit power is allocated to the 5.8GHz band, 25% to the 1.2GHz band, and 15% is reserved as an emergency backup. The link redundancy backup scheme adopts a dual-band simultaneous transmission mechanism, using the 5.8GHz band as the primary band to transmit critical data and the 1.2GHz band as a backup to transmit auxiliary data. The entire strategy generation process takes 320 milliseconds. The generated anti-interference strategy dataset contains detailed frequency band switching timing, power adjustment schemes, and redundant transmission protocols, which are distributed to each UAV in real time via data link for execution. The UAV swarm adjusts its communication parameters according to the new strategy.

[0074] In one embodiment, Example 4: The beamforming control module operates based on the anti-interference strategy dataset and the spatial distribution data of interference sources. This module achieves directional communication and interference suppression functions through multi-stage beam control calculations. The beam pointing calculation submodule first parses the priority communication frequency band sequence in the anti-interference strategy dataset. This sequence is arranged in descending order of frequency band availability score, which is calculated based on a comprehensive assessment of frequency band interference intensity, bandwidth capacity, and historical reliability. The interference direction avoidance calculation is based on the azimuth data in the spatial distribution dataset of interference sources. A spherical coordinate transformation is used to convert the ground coordinates of the interference source into the azimuth and elevation angles relative to the UAV antenna array. The influence of UAV attitude data on coordinate transformation is considered during the calculation. The beam main lobe pointing angle calculation adopts the maximum signal-to-interference ratio criterion. The optimal pointing angle is obtained by solving the eigenvectors of the array response matrix. The main lobe width is dynamically adjusted according to the communication distance. For long-distance communication, a narrow beam is used to improve gain, and for short-distance communication, a wide beam is used to enhance coverage.

[0075] The null formation control submodule processes the interference propagation direction vector in the 3D interference field model and uses a linearly constrained minimum variance algorithm to calculate the antenna array weighting coefficients. This algorithm sets two types of constraints: maintaining unity gain in the main lobe direction to ensure communication quality, and forming deep nulls in the interference direction to suppress interference, with a null depth requirement of below -30dB. The weighting coefficient calculation employs an adaptive iterative algorithm, with initial values ​​based on an ideal beam pattern, followed by gradient descent optimization based on real-time interference measurements. The beam null formation parameter set contains the null width, depth, and stability indices for each interference direction, used for subsequent phase compensation calculations. The phase compensation adjustment submodule calculates the phase compensation value for each element of the antenna array based on the main lobe pointing angle parameter and the beam null formation parameter set. The phase calculation uses a path difference model, considering the relationship between array element spacing and signal wavelength. For a uniform linear array, the phase difference between elements has a linear relationship with the direction of arrival. The amplitude weighting value is generated using a Chebyshev window function, achieving sidelobe suppression while controlling the main lobe width variation. The generated adaptive beamforming parameter set is loaded into the antenna array controller via a digital signal processor. The parameter update period is 100 milliseconds, which can be shortened to 10 milliseconds in emergency situations.

[0076] In practical scenarios, it is assumed that the UAV communication system operates in the C-band (4-8 GHz), and the antenna array adopts an 8×8 planar phased array structure. The anti-jamming strategy dataset provides the following priority communication frequency band sequence and power allocation scheme:

[0077] Table 1: Anti-interference strategy dataset

[0078]

[0079] The beam pointing calculation submodule selects the F12 band as the primary frequency band based on this strategy. Simultaneously, based on interference source location data, it identifies two main interference sources: interference source A (azimuth 35°, elevation 10°) and interference source B (azimuth 120°, elevation -5°). The main lobe pointing angle is calculated using the ground control station direction as a reference (azimuth 180°, elevation -15°). The optimal main lobe pointing angle, determined by the maximum signal-to-interference ratio algorithm, is azimuth 182°, elevation -14°. The null formation control submodule sets nulls in the direction of the interference sources, creating a null region of azimuth 35°±2° and elevation 10°±1° for interference source A, and a null region of azimuth 120°±3° and elevation -5°±1° for interference source B. The phase compensation adjustment submodule calculates the phase compensation values ​​for 64 array elements, with a compensation range of 0-360 degrees. The amplitude weighting uses Chebyshev weighting at the -35dB sidelobe level.

[0080] The hardware implementation employs a software-defined radio architecture, with the antenna array controller integrating an FPGA chip for real-time beamforming calculations. Each antenna element is connected to an independent RF link consisting of a digital attenuator and a phase shifter. The phase shifter achieves a phase resolution of 5.625 degrees, and the attenuator has a dynamic range of 30dB in 0.5dB steps. The control interface uses a high-speed serial bus, with parameter update latency controlled in the microsecond range. The system calibration module periodically performs channel amplitude and phase calibration by injecting a reference signal through a built-in test signal source to measure channel differences and storing calibration coefficients for real-time compensation. The algorithm implementation adopts a hierarchical optimization strategy: the first layer performs coarse-tuning calculations to quickly establish beam pointing, and the second layer performs fine-tuning to optimize null depth. The beamforming calculation uses a recursive least squares algorithm based on QR decomposition, improving computational efficiency while ensuring numerical stability. The real-time monitoring module continuously monitors the beamforming effect, dynamically adjusting algorithm parameters based on feedback received signal strength indicators and bit error rate metrics. The anomaly handling mechanism automatically switches to omnidirectional mode when beam misalignment is detected, recalculates the beam, and then resumes directional transmission.

[0081] In terms of data management, the beamforming parameter set is stored in binary encoding format, including array configuration parameters, weighting coefficient matrix, and calibration data. A historical parameter database records the optimal parameter combinations for different scenarios; when a similar interference pattern is detected, historical parameters are prioritized to accelerate convergence. The parameter verification mechanism verifies the validity of parameters through simulation before loading them into the actual system, avoiding communication interruptions caused by erroneous parameters. The system supports rapid switching between multiple parameter contingency plans, enabling beam reconfiguration to be completed in milliseconds when dealing with sudden interference scenarios.

[0082] In one embodiment, Example 5: The real-time spectrum monitoring module continuously collects spectrum data at a period of 50 milliseconds, and uses a sliding window mechanism to maintain the spectrum history of the most recent 10 seconds. The comparison between the new data and the spectrum interference feature spectrum is completed through a multi-feature joint detection algorithm, which simultaneously analyzes the signal strength change trend, spectrum shape characteristics, and modulation identification results. When a new interference signal feature is detected, the module generates real-time interference feature update data containing timestamps, frequency band identifiers, feature vectors, and confidence scores. This data is passed to the dynamic policy adjustment module through a shared memory interface. After receiving the real-time interference feature update data, the dynamic policy adjustment module immediately starts the reconstruction process of the three-dimensional interference field model. The reconstruction process uses an incremental update method to modify only the grid of the region affected by the new interference. The link vulnerability reassessment is based on the updated interference field data, calculates the signal-to-interference ratio (SIR) parameters of each point on the UAV's planned flight path, and identifies the region segments with vulnerability exceeding the threshold of 0.7. The anti-interference policy dataset is updated using a gradient descent optimization algorithm. The frequency band switching priority is reordered according to the performance of each frequency band in the new interference field, and the power allocation weight is adjusted according to the principle of maximizing channel capacity. The updated policy dataset is kept compatible with historical policies through a version management mechanism.

[0083] The communication quality assessment module acquires actual communication performance data through the physical layer and link layer monitoring interfaces of the UAV data link. Bit error rate (BER) measurement uses a statistical method based on CRC checksum, with each statistical unit consisting of 1000 data packets. Throughput monitoring is calculated based on the amount of successfully transmitted data within a 1-second time window. Communication performance is predicted based on the signal-to-interference ratio (SIR) parameter in the anti-interference strategy dataset, obtaining the theoretical value by querying a pre-stored SIR-BER mapping table. Communication quality deviation data calculation uses a relative error algorithm, recording the degree and duration of the difference between actual and predicted values. An optimization process is triggered when the deviation exceeds 15% for three consecutive cycles. The parameter optimization feedback module adjusts system parameters based on the communication quality deviation data. The adjustment of the historical interference pattern matching threshold uses a PID control algorithm, dynamically adjusting the matching coefficient tolerance range according to the deviation magnitude. Optimization of interference source positioning accuracy parameters is achieved by increasing the monitoring point density and the number of iterations in the positioning algorithm. Supplementary measurements are automatically triggered for areas with large positioning errors. The accuracy optimization of the three-dimensional interference field model is achieved by using multi-model fusion technology, which combines the outputs of the free space propagation model, ray tracing model and empirical propagation model, and weighted averages the outputs of each model according to the actual environmental conditions.

[0084] A multi-UAV collaborative module establishes a cluster data sharing mechanism based on time-division multiple access (TDMA). Each UAV broadcasts its own spectral interference characteristic spectrum and spatial distribution dataset of interference sources in a designated time slot. The data fusion center uses a distributed Kalman filter algorithm to process the data from each node, generating a cluster collaborative interference map. This map contains information on the location, intensity, and effective range of interference sources detected by all UAVs. Unified strategy adjustment is performed by the master UAV, which calculates the globally optimal anti-interference strategy based on the collaborative interference map. The strategy calculation considers the task priority, spatial location, and communication requirements of each UAV. Cooperative optimization of beamforming parameters is achieved through interference alignment technology. The beam pointing of each UAV is coordinated to avoid mutual interference, while forming a cooperative suppression beam targeting common interference sources.

[0085] At the hardware implementation level, system expansion capabilities rely on additional processing units and communication interfaces. The real-time spectrum monitoring module uses a dedicated signal processing DSP chip, possessing high-speed FFT computing capabilities and a large-capacity cache. The data sharing communication link employs anti-interference frequency hopping spread spectrum technology, separating the operating frequency band from the main communication link to avoid mutual interference. The cluster collaborative processing adopts a distributed computing architecture, with each UAV equipped with a collaborative processing unit responsible for local data preprocessing and fusion computing. The performance monitoring system integrates hardware performance counters to record the computational load and processing latency of each module in real time, providing a basis for resource scheduling.

[0086] The data processing workflow employs a combination of pipelined and parallel processing, with real-time data streams progressively refined through multiple processing nodes. A data quality verification mechanism includes data validity checks at each processing stage; invalid data is discarded, triggering a retransmission mechanism. A historical data management system maintains complete operational data for the past 24 hours, supporting data backtracking and analysis. An anomaly handling mechanism automatically switches to degrade mode when system performance degradation is detected, prioritizing basic communication functions while gradually restoring advanced features. System maintenance functions include automatic calibration and self-diagnosis modules, periodically performing system performance tests and generating health status reports. Software updates support over-the-air (OTA) download technology; new algorithms and parameter configurations can be remotely deployed via a secure channel. Operation logs record detailed system events and decision-making processes; log data is encrypted and stored, supporting post-event analysis. The entire extended system design meets high reliability and real-time requirements, continuously optimizing anti-interference performance in complex electromagnetic environments.

[0087] The above description is merely a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A spectrum-sensing-based anti-jamming system for unmanned aerial vehicles (UAVs), characterized in that, The system includes: The spectrum sensing data acquisition module acquires spectrum signal data of multiple frequency bands in the UAV's flight area, and generates a spectrum interference characteristic spectrum by analyzing the matching degree between the spectrum signal intensity change rate and historical interference patterns. Based on the aforementioned spectral interference feature spectrum, the interference source localization module extracts interference signal intensity distribution data, calculates the azimuth and distance parameters of the interference source relative to the UAV, and generates a spatial distribution dataset of the interference source. The three-dimensional interference field construction module constructs a three-dimensional interference field model that includes the interference intensity gradient and propagation direction based on the spatial distribution dataset of the interference source and the real-time flight altitude and geographic coordinate data of the UAV. The anti-interference strategy generation module analyzes the vulnerability index of the UAV communication link in the interference field based on the three-dimensional interference field model, dynamically adjusts the frequency band switching priority and the transmission power allocation weight, and generates an anti-interference strategy dataset. The beamforming control module calculates the phase offset parameters and beam pointing angle of the UAV communication antenna array based on the anti-interference strategy dataset, and generates an adaptive beamforming parameter set.

2. The anti-jamming system for unmanned aerial vehicles based on spectrum sensing according to claim 1, characterized in that, The spectral interference feature spectrum includes the rate of change of interference signal intensity, frequency band occupancy anomaly index, and historical interference pattern matching coefficient. The interference source spatial distribution dataset includes interference source azimuth angle estimation, interference source distance parameter, and interference signal intensity spatial gradient. The three-dimensional interference field model includes interference intensity isosurface data, interference propagation direction vector, and spatial vulnerability distribution map. The anti-interference strategy dataset includes priority communication frequency band sequence, transmit power graded weight, and link redundancy backup scheme. The adaptive beamforming parameter set includes beam main lobe pointing angle, beam null formation parameter, and array phase compensation value.

3. The anti-jamming system for unmanned aerial vehicles based on spectrum sensing according to claim 2, characterized in that, The spectrum sensing data acquisition module includes: The spectrum signal strength analysis submodule acquires real-time sampling data from multi-band spectrum monitoring equipment in the UAV flight area, calculates the rate of change of spectrum signal strength within each sampling time window, and obtains dynamic change data of spectrum signal by performing matching degree analysis with the historical interference pattern database. The frequency band occupancy assessment submodule calculates the ratio of signal occupancy duration to idle duration for each frequency band based on the dynamic change data of the spectrum signal, and judges the abnormal state of frequency band occupancy by combining the threshold values, and obtains a set of abnormal frequency band occupancy indicators. The interference feature extraction submodule integrates the spectrum signal strength change rate data with the historical interference pattern matching results based on the frequency band occupancy anomaly index set to generate a spectrum interference feature spectrum.

4. The anti-jamming system for unmanned aerial vehicles based on spectrum sensing according to claim 3, characterized in that, The interference source localization module includes: The interference signal intensity distribution analysis submodule receives the spectral interference feature spectrum, extracts the peak point data of the interference signal intensity in the feature spectrum and its corresponding frequency band information, calculates the signal intensity attenuation gradient at different monitoring points, and obtains the interference signal intensity distribution data. Based on the interference signal intensity distribution data, the multi-source positioning calculation submodule uses a time difference of arrival and frequency difference of arrival fusion positioning algorithm to calculate the azimuth and distance parameters of the interference source relative to the UAV platform and obtain preliminary positioning results. The spatial distribution integration submodule integrates multiple sets of azimuth and distance parameters from the preliminary positioning results, and generates a spatial distribution dataset of interference sources through spatial coordinate transformation and data fusion.

5. The anti-jamming system for unmanned aerial vehicles based on spectrum sensing according to claim 4, characterized in that, The three-dimensional interference field construction module includes: The spatial interpolation calculation submodule obtains the coordinates and intensity data of the interference sources in the spatial distribution dataset of the interference sources. Combined with the real-time flight altitude and geographic coordinate information of the UAV, it uses a three-dimensional spatial interpolation algorithm to generate a continuous interference intensity distribution field and obtain interference intensity isosurface data. The propagation model integration submodule, based on the interference intensity isosurface data, imports the electromagnetic wave propagation attenuation model and multipath effect parameters, calculates the propagation direction and attenuation gradient of the interference signal in three-dimensional space, and generates a set of interference propagation direction vectors. The vulnerability mapping submodule integrates the interference intensity isosurface data and the interference propagation direction vector set, analyzes the signal-to-interference ratio parameters of the UAV communication link at each spatial point, constructs a spatial vulnerability distribution map, and finally generates a three-dimensional interference field model.

6. The anti-jamming system for unmanned aerial vehicles based on spectrum sensing according to claim 5, characterized in that, The anti-interference strategy generation module includes: The link vulnerability assessment submodule receives the spatial vulnerability distribution map in the three-dimensional interference field model, extracts the data of the lowest signal-to-interference ratio point on the current flight trajectory of the UAV, calculates the communication link interruption probability and capacity degradation index, and obtains the link vulnerability assessment result. Based on the link vulnerability assessment results, the frequency band switching decision submodule analyzes the interference intensity distribution of each candidate frequency band in the three-dimensional interference field, calculates the frequency band availability score, and generates a priority communication frequency band sequence. The power allocation optimization submodule calculates the power allocation weight and link redundancy backup scheme for each frequency band based on the priority communication frequency band sequence and the UAV transmission power constraints, and generates an anti-interference strategy dataset.

7. The anti-jamming system for unmanned aerial vehicles based on spectrum sensing according to claim 6, characterized in that, The beamforming control module includes: The beam pointing calculation submodule obtains the priority communication frequency band sequence and power allocation weight in the anti-interference strategy dataset. Based on the azimuth information in the spatial distribution dataset of interference sources, it calculates the interference direction that the antenna array needs to avoid and the communication direction that needs to be enhanced, and obtains the beam main lobe pointing angle parameter. The null formation control submodule calculates the antenna array weighting coefficients based on the beam main lobe pointing angle parameter and the interference propagation direction vector in the three-dimensional interference field model to form beam nulls in the interference direction, thereby obtaining the beam null formation parameter set. The phase compensation adjustment submodule integrates the beam main lobe pointing angle parameter and the beam null formation parameter set, calculates the phase compensation value and amplitude weighting value of each element of the antenna array, and generates an adaptive beamforming parameter set.

8. The anti-jamming system for unmanned aerial vehicles based on spectrum sensing according to claim 7, characterized in that, The system also includes: The real-time spectrum monitoring module continuously collects spectrum signal data of the UAV's flight area, compares and analyzes it with the spectrum interference feature spectrum, detects newly emerging interference signal features, and generates real-time interference feature update data. The dynamic strategy adjustment module updates the data based on the real-time interference characteristics, reassesses the link vulnerability in conjunction with the three-dimensional interference field model, and updates the frequency band switching priority and power allocation weight in the anti-interference strategy dataset.

9. A spectrum-aware anti-jamming system for unmanned aerial vehicles according to claim 8, characterized in that, The system also includes: The communication quality assessment module monitors the actual bit error rate and throughput parameters of the UAV communication link, compares them with the communication performance predicted based on the anti-interference strategy dataset, and generates communication quality deviation data. The parameter optimization feedback module adjusts the interference pattern matching threshold in the spectrum interference feature spectrum and the positioning accuracy parameter in the interference source positioning module based on the communication quality deviation data, thereby optimizing the construction accuracy of the three-dimensional interference field model.

10. A spectrum-aware anti-jamming system for unmanned aerial vehicles according to claim 9, characterized in that, The system also includes: The multi-UAV collaborative module acquires the spectrum interference characteristic spectrum and spatial distribution dataset of interference sources of each UAV in the cluster, generates a cluster collaborative interference map through data fusion, and uniformly adjusts the anti-interference strategy dataset and adaptive beamforming parameter set of each UAV.

Images