Unmanned aerial vehicle signal identification method and system in complex electromagnetic environment

Abstract

The invention discloses an unmanned aerial vehicle signal identification method and system in a complex electromagnetic environment, and belongs to the field of unmanned aerial vehicles. According to the invention, the form of the liquid metal channel of the liquid metal antenna is adjusted to adapt to the target frequency band, and regional electromagnetic signals are collected in all directions. And generating spatial distribution data including the unmanned aerial vehicle and the interference direction based on the phase difference and preset features. The deformation of the liquid metal droplets is regulated and controlled through the electrowetting voltage, so that the antenna main lobe is aligned with the unmanned aerial vehicle, and the side lobe suppression area covers an interference source. Then, a target direction signal set is collected, and time domain fluctuation, frequency domain distribution and spatial domain direction multi-dimensional features are extracted; and inputting the multi-dimensional features into a joint probability model constructed by a Bayesian network, and outputting posterior probability information. Markov chain Monte Carlo self-adaptive sampling is utilized, the classification confidence coefficient is calculated through iterative convergence, the signal category is judged according to a threshold value, and the unmanned aerial vehicle signal recognition accuracy can be improved in the complex electromagnetic environment.

Description

Technical Field

[0001] This application belongs to the field of unmanned aerial vehicles (UAVs), and particularly relates to a method and system for identifying UAV signals in a complex electromagnetic environment. Background Art

[0002] The technology for identifying UAV signals has important application values in fields such as public safety, disaster rescue, and military reconnaissance. With the development of UAVs towards clustering and intelligence, the demand for signal identification in complex electromagnetic environments is becoming increasingly urgent. For example, at large-scale event sites, it is necessary to quickly separate the UAV communication link from dense Wi-Fi, Bluetooth, and other interferences. An efficient and accurate identification technology is the core foundation for UAV countermeasures, collaborative operations, and covert communications, directly affecting the emergency response speed and mission success rate.

[0003] In the prior art, the identification of UAV signals in complex electromagnetic environments mainly relies on the method of combining antennas with machine learning. For example, a microstrip antenna or an array antenna is used to collect signals, and models such as support vector machines and convolutional neural networks are used for UAV signal identification.

[0004] However, the existing UAV signal identification methods in complex electromagnetic environments, such as at large-scale event sites and disaster rescue sites, have dense interference sources such as Wi-Fi, Bluetooth, and mobile phone signals in these scenarios, making it difficult for traditional signal identification methods to effectively separate the target signal from the interference, resulting in a high misjudgment rate and low identification accuracy of the existing UAV signal identification methods for UAV signal identification in complex electromagnetic environments. Summary of the Invention

[0005] This application actually provides a method, system, device, and computer storage medium for identifying UAV signals in a complex electromagnetic environment, which can improve the accuracy of UAV signal identification in a complex electromagnetic environment.

[0006] In a first aspect, this application provides a method for identifying UAV signals in a complex electromagnetic environment. The method includes: Reconstructing the target morphological parameters of the liquid metal antenna by adjusting the morphological form of the liquid metal channel of the liquid metal antenna, and collecting the omnidirectional electromagnetic signals of the characteristic frequency band of the UAV in the target area; Generating spatial distribution data including the directions of UAV signals and interference signals according to the phase difference of the electromagnetic signals in the omnidirectional electromagnetic signals and the preset signal characteristics of UAVs and interference signals; According to the spatial distribution data, controlling the deformation of the liquid metal droplets of the liquid metal antenna by adjusting the electro-wetting control voltage parameters, so that the main lobe direction of the liquid metal antenna points to the UAV signal direction, and the sidelobe suppression area of the liquid metal antenna covers the spatial distribution range of the interference signal direction; Using a liquid metal antenna with lobe adjusted, collect an electromagnetic signal set of the drone signal direction, and extract multi-dimensional signal features of each sub-electromagnetic signal from the electromagnetic signal set. The multi-dimensional signal features include time-domain fluctuation features, frequency-domain distribution features, and spatial-direction features; Input the multi-dimensional signal features of each sub-electromagnetic signal into the constructed joint probability distribution model to obtain electromagnetic signal posterior probability information. The joint probability distribution model is constructed by a Bayesian network based on historical drone signals and historical interference signals; Adopt Markov Chain Monte Carlo to adaptively sample based on the electromagnetic signal posterior probability information, obtain the signal classification confidence through iterative convergence, and determine the identification information of the sub-electromagnetic signal according to a preset confidence threshold.

[0007] In an implementable embodiment, before extracting the multi-dimensional signal features of each sub-electromagnetic signal from the electromagnetic signal set, the method further includes: Perform blind source separation processing on the electromagnetic signal set, and based on the signal time-frequency sparsity assumption and independent component analysis, separate out multiple independent signal source electromagnetic signals; Extracting the multi-dimensional signal features of each sub-electromagnetic signal from the electromagnetic signal set includes: Extract the multi-dimensional signal features of each independent signal source electromagnetic signal from multiple independent signal source electromagnetic signals.

[0008] In an implementable embodiment, before generating spatial distribution data including the drone signal direction and the interference signal direction according to the phase difference of the electromagnetic signals in the omnidirectional electromagnetic signal and the preset signal features of the drone and the interference signal, the method further includes: Demodulate and parse the protocol of each electromagnetic signal in the omnidirectional electromagnetic signal to obtain demodulation protocol information including the demodulation information and protocol information of each electromagnetic signal; Generating spatial distribution data including the drone signal direction and the interference signal direction according to the phase difference of the electromagnetic signals in the omnidirectional electromagnetic signal and the preset signal features of the drone and the interference signal includes: Generate spatial distribution data including the drone signal direction and the interference signal direction according to the phase difference and demodulation protocol information of each electromagnetic signal in the omnidirectional electromagnetic signal and the preset signal features of the drone and the interference signal.

[0009] In an implementable embodiment, before controlling the deformation of the liquid metal droplets of the liquid metal antenna by adjusting the electro-wetting control voltage parameter according to the spatial distribution data, so that the main lobe direction of the liquid metal antenna points to the drone signal direction and the side lobe direction of the liquid metal antenna points to the interference signal direction, the method further includes: Predict the moving direction of the UAV signal through Kalman filtering based on the preset historical spatial distribution data; According to the moving direction of the UAV signal, control the deformation of the liquid metal droplets of the liquid metal antenna by adjusting the electrowetting control voltage parameters, so that the main lobe direction of the liquid metal antenna points to the moving direction of the UAV signal.

[0010] In an implementable embodiment, reconstruct the target morphological parameters of the liquid metal antenna by adjusting the morphological form of the liquid metal channel of the liquid metal antenna, and collect the omnidirectional electromagnetic signals in the characteristic frequency band of the UAV in the target area, including: Based on the center frequency and bandwidth of the characteristic frequency band of the UAV, determine the target morphological parameters of the liquid metal channel. The target morphological parameters include the target continuous length and target branch angle of the liquid metal filling path of the liquid metal antenna; Generate an electrowetting control voltage sequence according to the target continuous length and target branch angle. By gradually loading the single-stage voltage value in the electrowetting control voltage sequence, drive the liquid metal droplets of the liquid metal antenna to deform along the filling path until the liquid metal distribution state meets the target continuous length and target branch angle; Under the condition that the liquid metal distribution state meets the target continuous length and target branch angle, collect the omnidirectional electromagnetic signals in all directions in the target area.

[0011] In an implementable embodiment, according to the spatial distribution data, control the deformation of the liquid metal droplets of the liquid metal antenna by adjusting the electrowetting control voltage parameters, so that the main lobe direction of the liquid metal antenna points to the UAV signal direction, and the sidelobe suppression area of the liquid metal antenna covers the spatial distribution range of the interference signal direction, including: Based on the UAV signal direction and interference signal direction in the spatial distribution data, determine the angular deviation between the main lobe direction of the liquid metal antenna and the UAV signal direction, and the spatial coverage deviation between the sidelobe suppression area of the liquid metal antenna and the interference signal direction; Generate a control voltage correction value for the electrowetting control voltage parameters according to the angular deviation and spatial coverage deviation. The control voltage correction value includes a main lobe direction alignment component and a sidelobe suppression enhancement component; Decompose the control voltage correction value into multiple voltage gradient values, and apply the voltage gradient values to the electrode unit of the liquid metal antenna in sequence until the angular difference between the main lobe direction and the UAV signal direction is less than the preset difference threshold, and the sidelobe suppression area covers the spatial distribution range of the interference signal direction.

[0012] In an implementable embodiment, use Markov chain Monte Carlo to adaptively sample based on the posterior probability information of the electromagnetic signal, obtain the signal classification confidence through iterative convergence, and determine the recognition information of the sub-electromagnetic signal according to the preset confidence threshold, including: Determine the initial sampling step size and the initial sampling density based on the posterior probability information of the electromagnetic signal. The initial sampling step size and the initial sampling density correspond to the dimensional distribution of the multi-dimensional signal features. Within the range defined by the initial sampling step size and the initial sampling density, perform probability sampling on the dimensional distribution of the multi-dimensional signal features to generate a set of candidate sampling points. The set of candidate sampling points includes the signal feature combinations corresponding to the posterior probability information of the electromagnetic signal. Adjust the initial sampling step size and the initial sampling density according to the distribution dispersion of the set of candidate sampling points, and repeat the probability sampling until the distribution dispersion of the set of candidate sampling points is lower than a preset convergence threshold. Calculate the signal classification confidence of each signal category based on the proportion of the number of each signal category in the set of candidate sampling points. When the signal classification confidence with the largest value is greater than the preset confidence threshold, determine the signal category corresponding to the signal classification confidence as the identification information of the sub-electromagnetic signal.

[0013] In a second aspect, the present application provides a drone signal recognition system in a complex electromagnetic environment. The system includes: An acquisition module, configured to reconstruct the target morphological parameters of the liquid metal antenna by adjusting the morphological state of the liquid metal channels of the liquid metal antenna, and acquire the omnidirectional electromagnetic signals in the characteristic frequency band of the drones in the target area. A generation module, configured to generate spatial distribution data including the directions of the drone signals and the directions of the interference signals according to the phase differences of the electromagnetic signals in the omnidirectional electromagnetic signals and the preset signal characteristics of the drones and the interference signals. An adjustment module, configured to control the deformation of the liquid metal droplets of the liquid metal antenna by adjusting the electro-wetting control voltage parameters according to the spatial distribution data, so that the main lobe direction of the liquid metal antenna points to the direction of the drone signals, and the sidelobe suppression area of the liquid metal antenna covers the spatial distribution range of the directions of the interference signals. The acquisition module is further configured to use the liquid metal antenna with adjusted lobes to acquire the set of electromagnetic signals in the direction of the drone signals, and extract the multi-dimensional signal features of each sub-electromagnetic signal from the set of electromagnetic signals. The multi-dimensional signal features include time-domain fluctuation features, frequency-domain distribution features, and spatial domain direction features. An input module, configured to input the multi-dimensional signal features of each sub-electromagnetic signal into the constructed joint probability distribution model to obtain the posterior probability information of the electromagnetic signal. The joint probability distribution model is constructed by a Bayesian network based on historical drone signals and historical interference signals. A determination module, configured to perform adaptive sampling on the basis of the posterior probability information of the electromagnetic signal by using Markov chain Monte Carlo, obtain the signal classification confidence through iterative convergence, and determine the identification information of the sub-electromagnetic signal according to the preset confidence threshold.

[0014] In a third aspect, the present application provides an electronic device, which includes a processor and a memory storing computer program instructions; the processor reads and executes the computer program instructions to implement the method for identifying UAV signals in a complex electromagnetic environment according to any one of the embodiments in the first aspect.

[0015] In a fourth aspect, the present application provides a computer-readable storage medium, on which computer program instructions are stored, and when the computer program instructions are executed by a processor, the method for identifying UAV signals in a complex electromagnetic environment according to any one of the embodiments in the first aspect is implemented.

[0016] The method, system, device, and computer storage medium for identifying UAV signals in a complex electromagnetic environment according to the present application, through the dynamic form reconstruction ability of the liquid metal antenna, realize the omnidirectional signal adaptive capture of the characteristic frequency band in a complex electromagnetic environment, and overcome the limitation of the traditional fixed antenna structure for interference suppression; based on the spatial distribution modeling of the phase difference and signal characteristics, it is able to separate the propagation directions of UAV signals and dense interference sources from the spatial domain dimension, enhancing the spatial directivity of the target signal; combining the electro-wetting regulation technology to dynamically optimize the antenna lobe pattern, so that the main lobe accurately tracks the UAV signal while using the side lobe suppression to form an interference shielding area, effectively improving the signal-to-noise ratio of signal acquisition; through the fusion of multi-dimensional feature analysis in the time domain, frequency domain, and spatial domain, combined with the joint probability model constructed by the Bayesian network, it is able to distinguish the essential differences of signals from the statistical characteristic level; finally, using the adaptive sampling algorithm to dynamically optimize and iterate the posterior probability, it solves the problem of blurred classification boundaries of traditional machine learning models in a complex electromagnetic coupling environment, thereby significantly reducing the misjudgment rate and improving the accuracy and anti-interference ability of UAV signal identification in complex scenarios.

[0017] Furthermore, by adding a signal demodulation and protocol analysis link before the spatial distribution modeling, it is able to deconstruct the characteristics of electromagnetic signals from the communication protocol layer, effectively identifying the differential characteristics between the UAV-specific communication protocol and interference protocols such as Wi-Fi / Bluetooth; combining the modulation type, coding format, and protocol interaction characteristics in the demodulation information, it can strengthen the protocol fingerprint matching degree of UAV signals during the spatial positioning process, avoiding the problem of being misled by interference signals in similar frequency bands in traditional phase difference direction finding; by fusing the physical layer signal characteristics and protocol layer semantic information, constructing spatial distribution data with protocol identification capabilities, enabling subsequent antenna beamforming to exclude non-target protocol interference based on multi-dimensional characteristics; finally, in the scenario of coexistence of dense heterogeneous signals, it significantly improves the reliability of UAV signal spatial positioning and the ability to resist protocol confusion, providing a dual verification mechanism for accurate identification in a complex electromagnetic environment. Description of the Drawings

[0018] To more clearly illustrate the technical solutions of the embodiments of the present application, the following briefly introduces the accompanying drawings required to be used in the embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0019] Figure 1 It is a schematic flowchart of a method for identifying UAV signals in a complex electromagnetic environment provided by an embodiment of the present application; Figure 2 It is a schematic flowchart of a method for adjusting a liquid metal antenna provided by an embodiment of the present application; Figure 3 It is a schematic flowchart of a method for determining identification information of UAV signals provided by an embodiment of the present application; Figure 4 It is a schematic structural diagram of a UAV signal identification system in a complex electromagnetic environment provided by an embodiment of the present application; Figure 5 It is a schematic hardware structure diagram of an electronic device provided by an embodiment of the present application. Detailed Embodiments

[0020] The following will describe in detail the features and exemplary embodiments of various aspects of the present application. To make the purpose, technical solutions and advantages of the present application clearer and more understandable, the following further describes the present application in detail in combination with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain the present application, rather than limiting the present application. For those skilled in the art, the present application can be implemented without some of these specific details. The following description of the embodiments is only to provide a better understanding of the present application by showing examples of the present application.

[0021] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or further includes elements inherent to such process, method, article or device. Without further limitations, the elements defined by the statement "including..." do not exclude the existence of additional identical elements in the process, method, article or device including the said elements.

[0022] In the prior art, the identification of UAV signals in a complex electromagnetic environment mainly relies on the method of combining antennas with machine learning. For example, a microstrip antenna or an array antenna is used to collect signals, and models such as support vector machines and convolutional neural networks are used for UAV signal identification. However, the existing UAV signal identification methods in a complex electromagnetic environment, such as scenarios like large-scale event sites and disaster rescue sites, have dense interference sources such as Wi-Fi, Bluetooth, and mobile phone signals in these scenarios, making it difficult for traditional signal identification methods to effectively separate the target signal from the interference, resulting in a high misjudgment rate and low identification accuracy of the existing UAV signal identification methods in a complex electromagnetic environment.

[0023] To solve the problems of the prior art, the embodiments of the present application provide a method, a system, a device, and a computer storage medium for UAV signal identification in a complex electromagnetic environment. First, the method for UAV signal identification in a complex electromagnetic environment provided by the embodiments of the present application will be introduced below.

[0024] Figure 1 The flowchart of the method for UAV signal identification in a complex electromagnetic environment provided by an embodiment of the present application is shown. As Figure 1 shown, it includes steps S110 to S160.

[0025] S110: Reconstruct the target morphological parameters of the liquid metal antenna by adjusting the morphological pattern of the liquid metal channels of the liquid metal antenna, and collect the omnidirectional electromagnetic signals in the characteristic frequency band of UAVs in the target area.

[0026] The liquid metal antenna is a programmable antenna device based on the dynamic reconstruction characteristics of liquid metal fluid. The liquid metal inside it can change its distribution pattern under the action of an electric field to dynamically adjust the radiation characteristics. The liquid metal channel pattern refers to the spatial arrangement mode of the liquid metal flow path defined by the microchannel structure inside the antenna. The resonant frequency of the antenna is adjusted by changing the filling length and branch angle of the liquid metal in the channel. The target morphological parameters are the liquid metal distribution parameters pre-calculated to match the characteristic frequency band of UAVs, including the target continuous length and target branch angle of the liquid metal filling path. The target continuous length refers to the total length of the liquid metal filling path in the microchannel, and the target branch angle refers to the geometric angle at the bifurcation of the microchannel. The target area refers to the three-dimensional space range to be monitored, usually the airspace range where UAVs may operate, such as areas like large-scale event sites and disaster rescue sites. The characteristic frequency band of UAVs refers to the specific frequency band range used by UAV remote control, video transmission, and other communication links, such as the 5.8 GHz or 2.4 GHz sub-bands in the ISM band. The omnidirectional electromagnetic signal refers to the original electromagnetic radiation data in all directions and all frequency points in the target area obtained through the omnidirectional reception mode of the antenna.

[0027] First, determine the center frequency and bandwidth of the characteristic frequency band of the target UAV communication protocol. Based on the correspondence between the center frequency and bandwidth of the preset UAV characteristic frequency band and the morphological parameters of the liquid metal antenna, determine the target morphological parameters, namely the target continuous length and the target branch angle. Among them, the correspondence between the center frequency and bandwidth of the preset UAV characteristic frequency band and the morphological parameters of the liquid metal antenna is established through an electromagnetic simulation model. Subsequently, generate an electro-wetting regulation voltage sequence, which contains multiple voltage gradient values, and each gradient value corresponds to the driving voltage for the deformation of the liquid metal droplets in a specific area of the electrode array. By gradually loading the voltage gradient values, use the electro-wetting effect to control the movement of the liquid metal droplets along the filling path of the microchannel, and monitor the liquid metal distribution state in real time until its continuous length and branch angle reach the target parameters. The electro-wetting effect refers to the physical phenomenon that the interfacial tension between the liquid metal and the substrate material is changed by applying a voltage, thereby driving the deformation or displacement of the liquid metal droplets in the microchannel. Finally, in the morphological state of the reconstructed liquid metal antenna, collect electromagnetic signals in all directions within the target area in an omnidirectional scanning mode to form an omnidirectional electromagnetic signal dataset including time-domain waveforms, frequency-domain spectral lines, and spatial direction information.

[0028] In one embodiment, the liquid metal antenna can be a multi-channel liquid metal antenna array.

[0029] S120: Generate spatial distribution data including the UAV signal direction and the interference signal direction according to the phase difference of the electromagnetic signals in the omnidirectional electromagnetic signals and the preset signal characteristics of the UAV and the interference signals.

[0030] The signal characteristics of the UAV and the interference signals refer to the set of differential characteristics between the UAV communication signals and common interference sources extracted from historical data, including modulation methods, spectrum occupancy characteristics, signal periodicity, and protocol interaction modes, etc. Common interference sources can include Wi-Fi, Bluetooth, mobile phone signals, etc. The spatial distribution data is a set of spatial coordinate information representing the source directions of electromagnetic signals, including the three-dimensional angular coordinates of the UAV signal direction and the azimuth-pitch angle distribution range of the interference signal direction, and is used to describe the spatial position relationship of different signal sources within the target area.

[0031] First, perform multi-channel phase difference analysis on omnidirectional electromagnetic signals. Calculate the time difference of arrival of signals based on the phase difference data received by the array antenna, and estimate the incident direction angle of each electromagnetic signal in combination with the multi-channel phase interferometer algorithm. Subsequently, extract the time-frequency domain characteristics of the signals, including the instantaneous frequency, modulation depth, and spectral expansion characteristics of the signals, match them with the pre-stored UAV signal feature database, and screen out candidate UAV signals. At the same time, use the interference signal feature library to determine the type of interference for the remaining signals. The interference signal feature library can include Wi-Fi beacon frame features, Bluetooth hopping patterns, etc. Finally, fuse the incident direction of the UAV candidate signals with the azimuth information of the interference signals to construct a spatial distribution matrix with azimuth angle and pitch angle as coordinates, and mark the spatial ranges of the main direction of the UAV signals and the dense regions of the interference signals.

[0032] Exemplarily, in the UAV signal recognition scenario at a large-scale event site, the omnidirectional electromagnetic signal dataset contains multiple signals in the 5.8 GHz frequency band. The phase difference data is collected through a four-channel receiving array of liquid metal antennas, and the time delay differences of each signal in the X / Y axis directions are calculated. The direction of arrival estimation algorithm is used to obtain the azimuth angle and pitch angle of the signals. For time-frequency feature analysis, the short-time Fourier transform spectrum and cyclostationary characteristics of the signals are extracted and matched with the orthogonal frequency division multiplexing features in the UAV signal feature library to identify two candidate UAV signals. At the same time, it is detected that 12 signals have the fixed time interval characteristics of Wi-Fi beacon frames and are determined as interference signals. Finally, the direction of the UAV signals is marked as an azimuth angle of 120 degrees and a pitch angle of 30 degrees, and the direction of the interference signals is clustered into a high-density region with an azimuth angle of 60 - 90 degrees and a pitch angle of 0 - 20 degrees.

[0033] S130: According to the spatial distribution data, control the deformation of the liquid metal droplets of the liquid metal antenna by adjusting the electro-wetting control voltage parameters, so that the main lobe direction of the liquid metal antenna points to the UAV signal direction, and the sidelobe suppression region of the liquid metal antenna covers the spatial distribution range of the interference signal direction.

[0034] The electro-wetting control voltage parameters are a set of voltage parameters used to control the deformation of liquid metal, including voltage amplitude, application timing, and spatial distribution pattern. By adjusting the voltage combination of the electrode array, the flow path and shape of the droplets can be precisely controlled. The liquid metal droplets refer to discrete or continuous liquid metal fluid units in the antenna microchannel, and their deformation behavior directly affects the radiation pattern characteristics of the antenna, such as main lobe directivity and sidelobe suppression ability.

[0035] First, based on the directions of the UAV signals and interference signals in the spatial distribution data, calculate the angular deviation between the main lobe direction of the current liquid metal antenna and the target direction, as well as the spatial coverage deviation between the sidelobe suppression region and the interference direction. Through the mapping relationship between the angular deviation, spatial coverage deviation, and electrowetting control voltage parameters, obtain the target electrowetting control voltage parameters, that is, the control voltage correction value including the main lobe direction alignment component and the sidelobe suppression enhancement component. Among them, the mapping relationship between the angular deviation, spatial coverage deviation, and electrowetting control voltage parameters is established through an electromagnetic field simulation model. Decompose the correction value into multiple voltage gradient values, and apply them to the corresponding electrodes in stages according to the spatial arrangement order of the electrode array, and use the electrowetting effect to drive the gradual deformation of the liquid metal droplets. Real-time detect the main lobe pointing accuracy and sidelobe suppression effect of the antenna pattern, and iteratively adjust the voltage gradient until the main lobe direction error is less than the preset threshold and the sidelobe suppression range completely covers the interference region.

[0036] S140: Use the liquid metal antenna with adjusted lobes to collect the electromagnetic signal set in the direction of the UAV signal, and extract the multi-dimensional signal features of each sub-electromagnetic signal from the electromagnetic signal set. The multi-dimensional signal features include time-domain fluctuation features, frequency-domain distribution features, and spatial-domain direction features.

[0037] The electromagnetic signal set refers to the electromagnetic signal data set collected in the main lobe direction of the UAV signal by the liquid metal antenna with adjusted lobes, including the time-domain waveforms, frequency-domain spectral lines, and spatial-domain positioning information of multiple signals in the target direction. The sub-electromagnetic signal is a single-channel electromagnetic signal entity separated and processed in the signal set, corresponding to an independent signal source or a resolvable component in a mixed signal.

[0038] First, use the liquid metal antenna with lobe adjusted to collect directional signals in the main lobe direction of the UAV signal. Through beamforming technology, enhance the signal reception sensitivity in the target direction, suppress the penetration of interference signals in the sidelobe direction, and form a set of electromagnetic signals with high signal-to-noise ratio. Subsequently, preprocess the signal set, including signal normalization and noise filtering, to eliminate the noise floor introduced by the residual interference of the antenna pattern sidelobe. For each sub-electromagnetic signal after preprocessing, perform time-domain fluctuation analysis respectively, extract the variance feature and zero-crossing rate of the signal envelope; perform short-time Fourier transform to obtain the frequency-domain energy distribution features, including the main frequency offset and the out-of-band attenuation slope; combine the direction-of-arrival estimation results of the antenna array, calculate the time-varying stability parameters of the signal azimuth and elevation angles, and form the spatial domain direction features. Finally, align and integrate the time-domain, frequency-domain, and spatial-domain feature vectors according to the signal source dimension to construct a multi-dimensional signal feature matrix. The multi-dimensional signal features are a set of cross-domain features extracted from single-channel signals, where the time-domain fluctuation features characterize the variation law of the signal amplitude over time, including the peak volatility and the envelope fluctuation characteristics; the frequency-domain distribution features describe the distribution form of the signal energy within the frequency band, such as the main lobe bandwidth and the harmonic component intensity; the spatial domain direction features reflect the spatial directivity of the signal source, including the azimuth stability and the elevation angle change rate.

[0039] Exemplarily, in the UAV signal recognition scenario at the large event site, the main lobe of the liquid metal antenna with lobe adjusted points to the azimuth angle of 120 degrees and the elevation angle of 30 degrees, and the sidelobe suppression area covers the interference source with the azimuth angle of 60 - 90 degrees. A set of 5 sub-electromagnetic signals is obtained by directional acquisition, among which two are UAV signals and three are residual interference signals. Perform time-domain analysis on each signal, extract the periodic pulse envelope features of the UAV signal, and the interference signal shows random amplitude fluctuations; frequency-domain analysis shows that the UAV signal has a narrowband spectrum of 5.8 GHz ± 10 MHz, while the interference signal shows broadband spectrum characteristics; spatial-domain analysis calculates through the array phase difference, and confirms that the azimuth angle fluctuation range of the UAV signal is less than ±2 degrees, and the azimuth angle of the interference signal has a jump of ±15 degrees. In the finally generated multi-dimensional feature matrix, the UAV signal shows a combined feature of high time-domain periodicity, narrowband spectrum, and stable spatial domain directivity, while the interference signal corresponds to random time-domain fluctuations, broadband spectrum, and spatial domain jump features.

[0040] S150: Input the multi-dimensional signal features of each sub-electromagnetic signal into the constructed joint probability distribution model to obtain the posterior probability information of the electromagnetic signal. The joint probability distribution model is constructed by a Bayesian network based on historical UAV signals and historical interference signals.

[0041] The joint probability distribution model is a statistical model constructed based on a Bayesian network, which is used to describe the conditional probability relationship between multi-dimensional signal features and signal categories. Its network nodes represent feature variables and category labels of different dimensions, and the edges represent the dependency relationships between features. The signal categories can include UAV signals, interference signals, etc. The electromagnetic signal posterior probability information refers to the probability distribution data of the signal belonging to the UAV or interference category under the condition of given multi-dimensional signal features, including the probability values of each category and their confidence intervals, which are used to quantify the credibility of signal classification.

[0042] Input the multi-dimensional signal features of each sub-electromagnetic signal into the constructed joint probability distribution model. For each input multi-dimensional feature vector of the sub-electromagnetic signal, calculate its joint probability values under the UAV category and the interference category through the Bayesian network inference algorithm, and perform Bayesian update in combination with the prior probability. Finally, output the posterior probability information, including the UAV signal probability, the interference signal probability, and the probability ratio of the two. Among them, the prior probability can be the probability of UAV appearance in the target area.

[0043] Exemplarily, taking two candidate UAV signals collected at the scene of a large-scale event as an example, its multi-dimensional features include a time-domain periodic pulse envelope (feature A), a narrowband spectrum of 5.8 GHz ± 10 MHz (feature B), and an azimuth stability of ±2 degrees (feature C). After inputting the feature vector into the joint probability distribution model, the Bayesian network calculates the posterior probability of the signal belonging to the UAV category as 0.92 and the interference category probability as 0.08 according to the high conditional probability correlation between feature A and feature B of the UAV signal in the historical data; for another signal with time-domain random fluctuations (feature D), broadband spectrum (feature E), and azimuth jump (feature F), the calculated UAV probability is 0.15 and the interference probability is 0.85.

[0044] In one embodiment, before step S150: input the multi-dimensional signal features of each sub-electromagnetic signal into the constructed joint probability distribution model to obtain the electromagnetic signal posterior probability information, the method further includes: Obtain a training sample set, where the training sample set includes multiple training samples, and each training sample includes drone signal information, interference signal information in historical data, and the corresponding true signal category label; for each training sample, perform the following steps respectively: input the drone signal information and interference signal information in each training sample into a preset joint probability distribution model to obtain a predicted signal category label; determine the loss function value of the joint probability distribution model according to the true signal category label and the predicted signal category label; in the case that the loss function value does not meet the training stop condition, adjust the model parameters of the joint probability distribution model to obtain an updated joint probability distribution model, and return to input the drone signal information and interference signal information into the preset joint probability distribution model to obtain a signal category label until the training stop condition is met, and obtain a trained joint probability distribution model.

[0045] Among them, the joint probability distribution model is constructed through a Bayesian network topology structure, and the conditional dependence relationship between each dimension feature is determined through feature correlation analysis, such as the joint distribution constraint between the airspace direction feature and the frequency domain feature. The maximum likelihood estimation method is used to learn the conditional probability table of the network nodes, and the joint probability distribution model parameters are obtained by training with historical data.

[0046] S160: Use Markov Chain Monte Carlo to perform adaptive sampling on the posterior probability information of electromagnetic signals, obtain the signal classification confidence through iterative convergence, and determine the recognition information of sub-electromagnetic signals according to the preset confidence threshold.

[0047] Markov Chain Monte Carlo is a random sampling method based on Markov chains. It performs iterative sampling in the probability space by constructing a state transition probability matrix, and is used to obtain a sample set from a complex probability distribution to approximately calculate statistics. The preset confidence threshold is a preset classification decision threshold value. When the signal classification confidence exceeds this threshold, it is determined as the corresponding category. For example, the confidence threshold for drone signals is set to 0.8, and if it is lower than this value, it is regarded as an interference signal. The recognition information is the finally determined signal category label, including drone signals, interference signals, or unrecognized signal types.

[0048] First, determine the initial sampling step size and sampling density of the multi-dimensional feature space based on the posterior probability information of the electromagnetic signal. The initial step size is determined by the variance of the posterior probability distribution, and the sampling density is dynamically adjusted according to the number of feature dimensions. Within the range of the initial parameters, perform Markov chain Monte Carlo sampling to generate a set of candidate sampling points, where each sampling point represents a possible combination of signal features. Evaluate the convergence state according to the distribution dispersion of the sampling points. If the dispersion exceeds the preset convergence threshold, reduce the sampling step size and increase the sampling density, and repeat the sampling process until the distribution of the sampling points tends to be stable. Finally, count the proportion of the number of samples of each category in the set of candidate sampling points, and take the category with the highest proportion as the signal classification confidence. If this confidence exceeds the preset confidence threshold, output the corresponding category label as the recognition information.

[0049] In this embodiment, through the dynamic morphology reconstruction ability of the liquid metal antenna, the omnidirectional signal adaptive capture of the characteristic frequency band is realized in a complex electromagnetic environment, overcoming the limitation of the traditional fixed antenna structure for interference suppression; based on the spatial distribution modeling of the phase difference and signal characteristics, the propagation directions of the UAV signal and the dense interference sources can be separated from the spatial domain dimension, enhancing the spatial directivity of the target signal; combined with the electro-wetting regulation technology, the antenna lobe pattern is dynamically optimized, so that the main lobe accurately tracks the UAV signal while using the side lobe suppression to form an interference shielding area, effectively improving the signal-to-noise ratio of signal acquisition; through the multi-dimensional feature analysis integrating time domain, frequency domain and spatial domain, combined with the joint probability model constructed by the Bayesian network, the essential differences of signals can be distinguished from the statistical characteristics level; finally, the adaptive sampling algorithm is used to dynamically optimize and iterate the posterior probability, solving the problem of fuzzy classification boundaries of traditional machine learning models in a complex electromagnetic coupling environment, thereby significantly reducing the misjudgment rate and improving the accuracy and anti-interference ability of UAV signal recognition in complex scenarios.

[0050] In an implementable embodiment, before extracting the multi-dimensional signal features of each sub-electromagnetic signal from the electromagnetic signal set in step S140, the method further includes: Perform blind source separation processing on the electromagnetic signal set, and based on the signal time-frequency sparsity assumption and independent component analysis, separate multiple independent signal source electromagnetic signals.

[0051] The signal time-frequency sparsity hypothesis means that in the time-frequency joint domain, the time-frequency energy distributions of different signal sources have the characteristics of non-overlap or low overlap. This hypothesis assumes that at any moment and frequency point, only one signal source occupies the dominant energy, and the contributions of other signal sources can be ignored. In a complex electromagnetic environment, due to differences in modulation methods and protocols, the short-time spectra of UAV signals and interference signals have local sparse characteristics. For example, the periodic burst spectrum of UAV video transmission signals and the continuous broadband spectrum of Wi-Fi signals show an alternating occupancy characteristic in the time-frequency plane. Independent component analysis is a blind source separation method, and its core hypothesis is that the source signals in the mixed signal are statistically independent of each other. By finding a linear transformation matrix to maximize the independence between the output signals, the original independent signal sources can be recovered.

[0052] First, perform a time-frequency transformation on the electromagnetic signal set collected after lobe adjustment to generate a time-frequency energy distribution matrix. Based on the signal time-frequency sparsity hypothesis, use the short-time Fourier transform to calculate the spectral energy within each time window and construct a time-frequency domain mixed signal model. Decompose the time-frequency energy matrix through the non-negative matrix factorization algorithm to obtain the basis vectors and activation coefficient matrices representing different signal sources, and initially separate the signal components with significant time-frequency sparsity. Subsequently, apply independent component analysis to demix the time-domain mixed signal, optimize the separation matrix through the fixed-point iteration algorithm to maximize the non-Gaussianity of the output signals, and further separate the statistically independent signal sources. Finally, output the electromagnetic signals of multiple independent signal sources, where each signal source corresponds to a single physical emission source or a signal group with the same modulation characteristics.

[0053] Extract the multi-dimensional signal features of each sub-electromagnetic signal from the electromagnetic signal set, including: extract the multi-dimensional signal features of each independent signal source electromagnetic signal from the electromagnetic signals of multiple independent signal sources.

[0054] For each independent signal source electromagnetic signal after separation, perform feature extraction in the time domain, frequency domain, and spatial domain respectively. In time-domain analysis, use the Hilbert transform to extract the signal envelope and calculate its variance and zero-crossing rate as fluctuation features; in frequency-domain analysis, obtain the main lobe bandwidth and the proportion of harmonic components through power spectral density estimation; in spatial-domain analysis, combine the direction-of-arrival estimation results of the liquid metal antenna array to calculate the mean and standard deviation of the signal azimuth angle and elevation angle. Align the three types of features according to the signal source dimension to construct a multi-dimensional signal feature vector as the input for the subsequent classification model.

[0055] Exemplarily, in the scenario of identifying UAV signals at a large-scale event site, the hybrid electromagnetic signal set collected by the liquid metal antenna with lobe adjustment includes UAV video transmission signals, Wi-Fi hotspot signals, and Bluetooth device signals. First, perform a short-time Fourier transform on the hybrid signal to generate a time-frequency matrix and observe that there are periodic high-energy pulses (UAV signals) and continuous broadband energy regions (Wi-Fi signals) in the 5.8 GHz frequency band. Three groups of basis vectors corresponding to the typical time-frequency patterns of the three signal sources are separated by non-negative matrix factorization. Subsequently, apply independent component analysis to demix the time-domain signals, and use the kurtosis maximization criterion to separate two independent signals: one with high-kurtosis pulse modulation characteristics (UAV), and the other with low-kurtosis continuous characteristics (Wi-Fi). Features such as the time-domain envelope variance of 0.12, the frequency-domain main lobe bandwidth of 12 MHz, and the spatial azimuth angle standard deviation of 1.5 degrees are extracted from the separated UAV signals; the interference signals correspond to a time-domain variance of 0.35, a bandwidth of 40 MHz, and an azimuth angle standard deviation of 8 degrees. This process ensures that the signal features processed by the subsequent joint probability distribution model are derived from physically separated independent signal sources, improving the classification accuracy.

[0056] In an implementable embodiment, before step S120: generating spatial distribution data including the directions of UAV signals and interference signals according to the phase differences of the electromagnetic signals in the omnidirectional electromagnetic signal and the preset signal characteristics of UAVs and interference signals, the method further includes: Demodulate and parse the protocol of each electromagnetic signal in the omnidirectional electromagnetic signal to obtain demodulation protocol information including the demodulation information and protocol information of each electromagnetic signal.

[0057] Signal demodulation is the process of converting the received modulated signal into a baseband signal, and its core is to extract the original information waveform from the high-frequency carrier. Specifically, it includes identifying the modulation method, symbol rate, and carrier frequency of the signal, and restoring the baseband data stream through coherent demodulation or non-coherent demodulation algorithms. The modulation method can be FSK, QPSK, etc. Protocol parsing is to perform protocol layer analysis on the demodulated baseband signal to extract communication protocol features, including frame structure, synchronization header format, check mechanism, and interaction timing, etc. Demodulation protocol information is the joint output data of signal demodulation and protocol parsing, including the demodulated symbol sequence, modulation type identifier, protocol type identifier, and key protocol fields. The key protocol fields can be device address, frame type, etc. Demodulation protocol information is used to associate the physical layer features of the signal with the high-level protocol behavior, such as distinguishing UAV control signals from Wi-Fi data frames by protocol type.

[0058] First, perform adaptive demodulation processing on each electromagnetic signal in the omnidirectional electromagnetic signal. According to the signal spectrum characteristics and instantaneous amplitude distribution, identify the modulation method through the maximum likelihood estimation algorithm, and restore the symbol clock based on the phase-locked loop synchronization technology. Dynamically configure demodulation parameters for different modulation types. For example, use a frequency discriminator to demodulate FSK signals and use quadrature coherent demodulation for QPSK signals. After demodulation, output the baseband symbol sequence and modulation parameters, which can be symbol rate, frequency offset, etc. Subsequently, perform protocol parsing on the baseband symbol sequence and use the protocol feature library for pattern matching. The protocol feature library stores the frame structure templates of common communication protocols, including synchronization header features, address field positions, and CRC check algorithms. Common communication protocols can be Wi-Fi 802.11n, drone video transmission protocols, etc. Detect the synchronization header pattern in the baseband sequence through a sliding window, determine the protocol type, and extract the protocol fields in the payload. Finally, integrate the demodulation parameters and protocol fields to generate demodulation protocol information, including modulation method, symbol rate, protocol type, and key frame structure parameters.

[0059] Step S120: Generate spatial distribution data including the directions of the drone signal and the interference signal based on the phase difference of the electromagnetic signals in the omnidirectional electromagnetic signal and the preset signal characteristics of the drone and the interference signal, including: Generate spatial distribution data including the directions of the drone signal and the interference signal based on the phase difference of each electromagnetic signal in the omnidirectional electromagnetic signal, the demodulation protocol information, and the preset signal characteristics of the drone and the interference signal.

[0060] First, based on the received signal phase difference of the multi-channel liquid metal antenna array, use the multiple signal classification algorithm to calculate the direction of arrival of each electromagnetic signal, and obtain the preliminary azimuth and elevation angle estimation values. Subsequently, match the demodulation protocol information with the preset drone signal feature library: If the frame structure features of the drone-specific protocol are detected and the modulation parameters conform to the historical drone signal pattern, it is marked as a candidate drone signal; otherwise, in combination with the interference signal feature library, such as determining it as an interference signal through the Wi-Fi beacon frame interval feature. For candidate drone signals, calculate the similarity between its signal characteristics (such as time-domain pulse period, frequency-domain harmonic distribution) and the drone feature library, and weighted correct the direction estimation value to reduce the azimuth error caused by the multipath effect. The signal characteristics can be time-domain pulse period, frequency-domain harmonic distribution, etc. For interference signals, cluster the directions of the same type of interference sources according to the protocol parsing results and statistically analyze their spatial distribution density. Finally, integrate the corrected direction information and category labels of all signals to generate spatial distribution data, including the three-dimensional direction coordinates of the drone signal and the direction clustering area of the interference signal.

[0061] Exemplarily, in the scenario of identifying UAV signals at a large-scale event site, the omnidirectional electromagnetic signal contains multiple signals in the 5.8 GHz frequency band. First, demodulation is performed on each signal. One signal is identified as QPSK modulation with a symbol rate of 2 Msps. Protocol analysis detects that its frame structure conforms to the UAV video transmission protocol, the synchronization header is 0xAA55 and contains a UAV serial number field, and it is marked as a candidate UAV signal. Another signal is demodulated to OFDM modulation, and protocol analysis matches the Wi-Fi beacon frame characteristics, and it is determined as an interference signal. Through phase difference calculation, the initial direction of the UAV signal is estimated to be an azimuth angle of 125 degrees and a pitch angle of 28 degrees, but there is a fluctuation of ±5 degrees. Combining the protocol matching result with the historical UAV signal time-domain pulse characteristics (period 10 ms), the direction is corrected to an azimuth angle of 120 degrees and a pitch angle of 30 degrees, and the fluctuation range is reduced to ±2 degrees. The direction clustering of the interference signal shows a dense distribution in the area of azimuth angle 60 - 90 degrees and pitch angle 0 - 20 degrees. Finally, in the generated spatial distribution data, the direction of the UAV signal accurately points to the corrected coordinates, and the interference signal is marked as a high-density area, providing a basis for spatial suppression in subsequent beamforming.

[0062] In this embodiment, by adding a signal demodulation and protocol analysis link before spatial distribution modeling, the electromagnetic signal can be deconstructed in terms of characteristics from the communication protocol layer, effectively identifying the differential characteristics between the UAV dedicated communication protocol and interference protocols such as Wi-Fi / Bluetooth; combining the modulation type, coding format, and protocol interaction characteristics in the demodulation information can strengthen the protocol fingerprint matching degree of UAV signals during airspace positioning, avoiding the problem that traditional phase difference direction finding is misled by interference signals in similar frequency bands; by fusing the physical layer signal characteristics and protocol layer semantic information, a spatial distribution data with protocol identification ability is constructed, enabling subsequent antenna beamforming to exclude non-target protocol interference based on multi-dimensional characteristics; finally, in the scenario of coexistence of dense heterogeneous signals, the reliability of UAV signal spatial positioning and the ability to resist protocol confusion are significantly improved, providing a dual verification mechanism for accurate identification in complex electromagnetic environments.

[0063] In an implementable embodiment, before step S130: According to the spatial distribution data, control the deformation of the liquid metal droplets of the liquid metal antenna by adjusting the electro-wetting control voltage parameter, so that the main lobe direction of the liquid metal antenna points to the UAV signal direction and the side lobe direction of the liquid metal antenna points to the interference signal direction, the method further includes: According to the preset historical spatial distribution data, the moving direction of the UAV signal is predicted by Kalman filtering. The historical spatial distribution data is a set of UAV signal direction information recorded in the past time period, including parameters such as time stamps, three-dimensional azimuth and pitch angle coordinates, and signal strength, which are used to describe the movement trajectory and dynamic change law of UAV signals in the airspace.

[0064] First, construct the state vector and observation vector of the Kalman filter. The state vector contains the dynamic parameters of the UAV signal direction, such as azimuth, pitch, and their first-order derivatives, such as angular velocity. The observation vector is composed of the UAV direction coordinates in the current spatial distribution data. Initialize the Kalman filter parameters, including the state transition matrix, process noise covariance, and observation noise covariance. The state transition matrix is used to describe the linear relationship between the azimuth and pitch angles over time. The process noise covariance is used to characterize the uncertainty of the motion model, and the observation noise covariance reflects the direction measurement error.

[0065] Input the Kalman filter algorithm in chronological order through the historical spatial distribution data sequence. For each time step, execute the prediction stage. Based on the state estimate value at the previous moment and the state transition matrix, predict the current UAV direction. Execute the update stage. Combine the current actual measured direction, calculate the Kalman gain, and correct the predicted value to output the optimal estimated direction. After iteratively processing all historical data, extrapolate the moving direction of the UAV signal at future moments according to the latest state estimate value, including the azimuth change rate and pitch change rate, to form the moving direction prediction result.

[0066] According to the moving direction of the UAV signal, control the deformation of the liquid metal droplets of the liquid metal antenna by adjusting the electrowetting control voltage parameters, so that the main lobe direction of the liquid metal antenna points to the moving direction of the UAV signal.

[0067] Based on the predicted moving direction of the UAV signal, calculate the angle deviation between the current main lobe direction of the liquid metal antenna and the predicted direction. The angle deviation includes the azimuth deviation component and the pitch deviation component, corresponding to the direction differences in the horizontal and vertical planes respectively. Through the mapping relationship between the angle deviation and the electrowetting control voltage parameters, obtain the target electrowetting control voltage parameters, that is, the voltage correction value including the azimuth adjustment component and the pitch adjustment component. The mapping relationship between the angle deviation and the electrowetting control voltage parameters is established through the electromagnetic field simulation model. Decompose the voltage correction value into multiple voltage gradients and load them in stages according to the spatial distribution order of the electrode array. For example, the azimuth adjustment corresponds to the voltage gradient sequence of the horizontal electrode pairs, and the pitch adjustment corresponds to the voltage gradient sequence of the vertical electrode pairs. Drive the deformation of the liquid metal droplets along the microchannel through the electrowetting effect and monitor the change of the main lobe direction in real time. Stop the voltage adjustment when the deviation between the main lobe direction and the predicted direction is less than the preset threshold to complete the real-time tracking of the antenna beam.

[0068] Exemplarily, in the scenario of UAV signal recognition at the site of a large-scale event, the historical spatial distribution data shows that the azimuth angle of the UAV signal linearly increases from 100 degrees to 120 degrees within the past 5 seconds, and the pitch angle remains stable at 30 degrees. A state vector is constructed through Kalman filtering, including the azimuth angle, pitch angle, and their angular velocities. The state transition matrix is set to a uniform motion model, and the process noise covariance is set according to the historical angular velocity fluctuations. After inputting 10 consecutive historical direction data, Kalman filtering predicts that the azimuth angle will increase by 5 degrees to 125 degrees at the next moment, and the pitch angle remains 30 degrees. According to the prediction result, the azimuth deviation between the current main lobe pointing and the predicted direction is calculated to be 5 degrees. Through the mapping relationship between the angle deviation and the electro-wetting regulation voltage parameter, a voltage correction sequence for the horizontal direction electrode pair is obtained, and the gradient voltage is loaded in three steps, such as 3V, 5V, and 7V, to drive the liquid metal droplet to flow to the right. The change of the main lobe azimuth angle is detected in real time, and the voltage loading is stopped when it reaches 125 degrees, and the beam main lobe of the adjusted antenna continuously tracks the moving direction of the UAV.

[0069] In an implementable embodiment, step S110: Reconstruct the target morphological parameters of the liquid metal antenna by adjusting the morphological form of the liquid metal channel of the liquid metal antenna, and collect the omnidirectional electromagnetic signals in the characteristic frequency band of the UAV in the target area, including: Based on the center frequency and bandwidth of the characteristic frequency band of the UAV, determine the target morphological parameters of the liquid metal channel. The target morphological parameters include the target continuous length and target branch angle of the liquid metal filling path of the liquid metal antenna.

[0070] The center frequency of the characteristic frequency band of the UAV refers to the midpoint frequency of the frequency band used by the target UAV communication link. For example, the center frequency of the 5.8 GHz frequency band is 5.8 GHz; the bandwidth refers to the effective frequency range of this frequency band, such as ±10 MHz.

[0071] First, determine the center frequency and bandwidth parameters of the characteristic frequency band according to the target UAV communication protocol. Then, based on the center frequency and bandwidth parameters, use the corresponding relationship between the center frequency and bandwidth of the characteristic frequency band of the UAV established by the electromagnetic simulation model and the morphological parameters of the liquid metal antenna to calculate the liquid metal filling path length and branch angle required to meet the frequency band coverage. For example, for the 5.8 GHz ±10 MHz frequency band, according to the center frequency and bandwidth parameters, the target continuous length is 12 mm, and the target branch angle is 60 degrees. This process ensures the precise matching of the antenna morphological parameters with the target frequency band and provides a hardware basis for subsequent signal acquisition.

[0072] Generate an electro-wetting regulation voltage sequence according to the target continuous length and target branch angle, and drive the liquid metal droplet of the liquid metal antenna to deform along the filling path by gradually loading the single-stage voltage value in the electro-wetting regulation voltage sequence until the liquid metal distribution state meets the target continuous length and target branch angle.

[0073] The electrowetting control voltage sequence is a sequence composed of multiple voltage gradient values. Each gradient value corresponds to the driving voltage of a specific electrode region and is used to control the flow path and deformation amplitude of the liquid metal droplet in stages. The liquid metal distribution state refers to the filling form of the liquid metal in the microchannel.

[0074] First, convert the target continuous length and branch angle into the voltage configuration parameters of the electrode array through the correspondence between the continuous length and branch angle and the voltage configuration parameters of the electrode array. The correspondence between the continuous length and branch angle and the voltage configuration parameters of the electrode array is constructed based on the historical form parameters and voltage configuration parameters. Then, based on the non-linear relationship between voltage and liquid metal wettability in the electrowetting effect, determine the voltage gradient sequence loaded in stages. For example, when the target continuous length needs to increase by 5 mm, it is decomposed into 3 voltage gradients, such as 3 V, 5 V, and 7 V, and applied to the horizontal electrode pair in sequence to drive the droplet to extend step by step. At the same time, for the adjustment of the branch angle, apply different voltages to the bifurcated electrodes to control the droplet splitting. It is also possible to stop the voltage loading by monitoring the liquid metal distribution state when the error between the detected continuous length and branch angle is less than the preset threshold.

[0075] Under the condition that the liquid metal distribution state meets the target continuous length and target branch angle, collect the omnidirectional electromagnetic signals in all directions within the target area.

[0076] After the liquid metal distribution state meets the target parameters, activate the omnidirectional scanning mode of the antenna. Synchronously collect the signals in each direction through the multi-channel receiving array, and use the high-speed analog-to-digital converter to digitize the signals. During the collection process, the antenna maintains dynamic impedance matching to minimize the reflection loss and ensure the high-fidelity reception of the full-band signals. Finally, generate an omnidirectional electromagnetic signal dataset containing time-domain waveforms, spectral energy, and phase difference information.

[0077] Exemplarily, in the scenario of UAV signal recognition at a large-scale event site, the characteristic frequency band of the UAV is 5.8 GHz, with a center frequency of 5.8 GHz and a bandwidth of 20 MHz. First, based on the correspondence between the center frequency and bandwidth of the UAV characteristic frequency band and the morphological parameters of the liquid metal antenna, the target continuous length of the liquid metal antenna is determined to be 12.9 mm, and the branch angle is 30 degrees. Then, based on the non-linear relationship between the voltage and the wettability of the liquid metal in the electro-wetting effect, an electro-wetting control voltage sequence is generated. The horizontal electrode pair is sequentially loaded with voltages of 3V, 5V, and 7V, and the forked electrode is applied with a voltage of 2V to form a 30-degree branch. Real-time impedance detection shows that the liquid metal extends to 12.8 mm, and the branch angle is 29.5 degrees, meeting the error threshold. Subsequently, the antenna switches to the omnidirectional mode, and all-direction electromagnetic signals in the frequency band of 5.8 GHz ± 10 MHz are collected through a four-channel array to form an omnidirectional electromagnetic signal dataset including time-domain waveforms, frequency-domain spectral lines, and spatial direction information.

[0078] Figure 2 FIG. shows a schematic flow chart of a method for adjusting a liquid metal antenna provided by an embodiment of the present application. As Figure 1 shown, it includes steps S210 to S230.

[0079] In an implementable embodiment, step S130: According to the spatial distribution data, control the deformation of the liquid metal droplets of the liquid metal antenna by adjusting the electro-wetting control voltage parameters, so that the main lobe direction of the liquid metal antenna points to the UAV signal direction, and the sidelobe suppression area of the liquid metal antenna covers the spatial distribution range of the interference signal direction, including: S210: Based on the UAV signal direction and the interference signal direction in the spatial distribution data, determine the angular deviation between the main lobe direction of the liquid metal antenna and the UAV signal direction, and the spatial coverage deviation between the sidelobe suppression area of the liquid metal antenna and the interference signal direction.

[0080] The angular deviation refers to the azimuth and elevation angle differences between the current main lobe radiation direction of the liquid metal antenna and the target UAV signal direction. The main lobe direction is determined by the maximum radiation direction of the antenna pattern, and the target direction comes from the three-dimensional coordinates of the UAV signal marked in the spatial distribution data. The spatial coverage deviation refers to the spatial gap where the sidelobe suppression area of the antenna fails to fully cover the interference signal direction distribution range, manifested as insufficient azimuth and elevation angle overlap between the interference direction clustering area and the boundary of the sidelobe suppression area.

[0081] First, extract the three-dimensional direction coordinates of the UAV signal from the spatial distribution data, including the azimuth angle and the pitch angle. Call the current pattern parameters of the liquid metal antenna to obtain the maximum radiation direction of the main lobe. Calculate the azimuth deviation component and the pitch deviation component of the two in the horizontal plane and the vertical plane to obtain the angle deviation. At the same time, extract the boundary coordinates of the interference signal direction clustering region, such as the azimuth angle range and the pitch angle range. Compare the azimuth and pitch coverage ranges of the sidelobe suppression region with the interference region boundary, and calculate the angle span of the uncovered region, that is, the spatial coverage deviation.

[0082] S220: Generate a regulation voltage correction value of the electro-wetting regulation voltage parameter according to the angle deviation and the spatial coverage deviation. The regulation voltage correction value includes a main lobe direction alignment component and a sidelobe suppression enhancement component.

[0083] The main lobe direction alignment component is a set of voltage parameters required to adjust the main lobe direction of the liquid metal antenna. By changing the electrode voltage to drive the deformation of the liquid metal, the main lobe is directed towards the target UAV direction. The sidelobe suppression enhancement component is a set of voltage parameters for optimizing the coverage range of the sidelobe suppression region. By adjusting the distribution form of the liquid metal, the sidelobe energy is compressed, and the spatial coverage range of the suppression region is expanded.

[0084] Based on the angle deviation and the spatial coverage deviation, query the mapping relationship between the preset angle deviation, spatial coverage deviation and electro-wetting regulation voltage parameter to obtain the corresponding electro-wetting regulation voltage parameter, that is, the voltage correction value of the corresponding electrode region of the liquid metal antenna. The mapping relationship table of the angle deviation, spatial coverage deviation and electro-wetting regulation voltage parameter is established through electromagnetic simulation, and the voltage correction values of each electrode pair under different angle deviations and spatial coverage deviations are recorded. The voltage correction value includes the voltage parameters required to adjust the main lobe direction of the liquid metal antenna, that is, the main lobe direction alignment component; the voltage correction value also includes the voltage parameters for optimizing the coverage range of the sidelobe suppression region, that is, the sidelobe suppression enhancement component.

[0085] S230: Decompose the regulation voltage correction value into multiple voltage gradient values, and apply the voltage gradient values to the electrode units of the liquid metal antenna in sequence until the angle difference between the main lobe direction and the UAV signal direction is less than the preset difference threshold, and the sidelobe suppression region covers the spatial distribution range of the interference signal direction.

[0086] The voltage parameters of the main lobe alignment component and the sidelobe suppression component are decomposed into multiple sub-steps according to the electrode spatial distribution. For example, the horizontal-direction electrode pairs are loaded with voltages of 3V, 5V, and 7V in three steps according to the azimuth deviation, and the vertical-direction electrode pairs are loaded with voltages of 2V and 4V in two steps according to the pitch deviation. A voltage gradient is applied to the corresponding electrode units in stages, and the distribution pattern of the liquid metal is detected. The filling path length and branch angle of the liquid metal are used to determine the main lobe direction of the current antenna pattern and the coverage range of the sidelobe suppression area. If the main lobe angle deviation is still higher than the threshold or the proportion of the coverage range of the sidelobe suppression area covering the interference direction distribution range is less than the preset proportion threshold, the next gradient voltage is continuously loaded. Iterative adjustment is performed until the main lobe direction error is less than the preset threshold and the proportion of the coverage range of the sidelobe suppression area covering the interference direction distribution range is greater than or equal to the preset proportion threshold.

[0087] Exemplarily, in the scenario of identifying UAV signals at the site of a large-scale event, the spatial distribution data shows that the target UAV signal is located at an azimuth angle of 120 degrees and a pitch angle of 30 degrees, and the interference signals are densely distributed in the area of azimuth angles 60 - 90 degrees. The initial main lobe direction of the liquid metal antenna is at an azimuth angle of 115 degrees and a pitch angle of 28 degrees, and the sidelobe suppression area only covers the azimuth angles 70 - 85 degrees. First, according to the angle deviation (azimuth 5 degrees, pitch 2 degrees) and the spatial coverage deviation (not covered in the azimuth ranges 60 - 70 degrees and 85 - 90 degrees), a regulation voltage correction value is generated: the total increment of the horizontal electrode pairs corresponding to the main lobe alignment component is 8V and the vertical electrode pairs is 4V, and the sidelobe enhancement component corresponds to 6V for the outer horizontal electrodes. The correction value is decomposed into three-step loading for the horizontal electrodes (3V, 3V, 2V), two-step loading for the vertical electrodes (2V, 2V), and three-step loading for the outer electrodes (2V, 2V, 2V), and is applied to the electrode units in stages. After each step of loading, the liquid metal form is detected by an impedance sensor and the pattern is deduced inversely. After three iterations, the main lobe direction is adjusted to an azimuth angle of 120.3 degrees and a pitch angle of 30.1 degrees, that is, the deviation is < 0.5 degrees, and the sidelobe suppression area expands to the azimuth angles 60 - 90 degrees, completely covering the interference direction, and finally high-precision beamforming and interference shielding are achieved.

[0088] Figure 3 The flowchart showing the method for determining the identification information of UAV signals provided by an embodiment of the present application is shown. As Figure 1 shown, it includes steps S310 to S340.

[0089] In an implementable embodiment, step S160: Using Markov chain Monte Carlo to perform adaptive sampling on the posterior probability information based on electromagnetic signals, obtaining the signal classification confidence through iterative convergence, and determining the identification information of sub-electromagnetic signals according to a preset confidence threshold, including: S310: Determine the initial sampling step and the initial sampling density based on the electromagnetic signal posterior probability information. The initial sampling step and the initial sampling density correspond to the dimensional distribution of the multi-dimensional signal features.

[0090] The initial sampling step refers to the maximum jump distance for each state transition in the Markov Chain Monte Carlo algorithm, which appears as the maximum allowable eigenvalue change range for each dimension in the multi-dimensional feature space. The initial sampling density refers to the number of candidate points sampled within the unit feature space volume, reflecting the probability density coverage degree of the feature dimensional distribution. The dimensional distribution of the multi-dimensional signal features refers to the joint probability distribution form formed by the time-domain fluctuation features, the frequency-domain distribution features, and the spatial-domain direction features in the multi-dimensional space, characterizing the aggregation regions of different signal categories in the feature space.

[0091] According to the electromagnetic signal posterior probability information output by the Bayesian network, analyze the covariance matrix of the multi-dimensional signal features to determine the variance range of each feature dimension. The initial sampling step is set to the square root of the variance of each dimension to ensure that the step is proportional to the feature distribution range. The initial sampling density is set using an exponential decay rule according to the number of feature dimensions, with a lower density for a larger number of dimensions. For example, the initial density in a three-dimensional feature space is set to 10 sampling points per unit cube, and it is reduced to 5 in a four-dimensional space. Adjust the step direction through the eigenvalue decomposition of the covariance matrix to align the sampling direction with the principal components and improve the initial sampling efficiency.

[0092] S320: Within the range defined by the initial sampling step and the initial sampling density, perform probability sampling on the dimensional distribution of the multi-dimensional signal features to generate a set of candidate sampling points. The set of candidate sampling points includes the signal feature combinations corresponding to the electromagnetic signal posterior probability information.

[0093] The set of candidate sampling points refers to the set of points in the feature space generated by the Markov Chain Monte Carlo algorithm. Each point represents a possible signal feature combination, corresponding to a probability distribution sample of the UAV or interference category. The signal feature combination refers to a specific value combination of the time-domain, frequency-domain, and spatial-domain features, such as a time-domain variance of 0.1, a frequency band width of 15 MHz, and an azimuth angle fluctuation of ±2 degrees.

[0094] Use the Metropolis-Hastings algorithm to construct a Markov chain. Take the current posterior probability peak point as the initial state and generate candidate points according to the initial step. For each candidate point, calculate its acceptance probability: accept it if the posterior probability is higher than the current point, otherwise randomly accept it according to the probability ratio. In the three-dimensional feature space, sequentially execute the Gaussian distribution proposal function on the time-domain, frequency-domain, and spatial-domain dimensions to generate candidate points. For example, the time-domain feature is sampled from the current value with a normal distribution of step 0.05, and the frequency-domain feature is sampled with a step of 2 MHz. After iteratively generating 1000 candidate points, remove duplicate samples to form the set of candidate sampling points.

[0095] S330: Adjust the initial sampling step size and initial sampling density according to the distribution dispersion degree of the candidate sampling point set, and repeatedly perform probability sampling until the distribution dispersion degree of the candidate sampling point set is lower than the preset convergence threshold.

[0096] The distribution dispersion degree refers to the degree of dispersion of the candidate sampling point set in the feature space, which is quantified by calculating the weighted sum of the sample variances of each dimension. The preset convergence threshold refers to the upper limit of the dispersion degree for determining the stability of the sampling result. For example, in a three-dimensional space, the dispersion degree threshold is set to 0.01.

[0097] Calculate the covariance matrix of the candidate sampling point set, and obtain the dispersion degree index by weighted summing the variances of each dimension. If the dispersion degree exceeds the threshold, reduce the sampling step size to 0.8 times the original value and increase the sampling density by 1.5 times. For example, the initial step size of 0.1 is adjusted to 0.08, and the density increases from 10 points / unit to 15 points / unit. Use the adaptive Markov chain Monte Carlo algorithm to dynamically update the parameters of the proposal function. Repeatedly execute the sampling-evaluation-adjustment loop until the change rate of the dispersion degree in three consecutive iterations is less than 5%, and it is determined to converge.

[0098] S340: Calculate the signal classification confidence of each signal category based on the proportion of the number of each signal category in the candidate sampling point set. When the signal classification confidence with the largest value is greater than the preset confidence threshold, determine the signal category corresponding to the signal classification confidence as the identification information of the sub-electromagnetic signal.

[0099] The signal classification confidence refers to the normalized value of the proportion of sampling points of a certain category, which reflects the probability that the signal belongs to this category. For example, if the proportion of the drone category is 80%, the confidence is 0.8.

[0100] Count the proportion of the number of each category label in the candidate sampling point set. For the drone category, calculate the ratio of the number of its sampling points to the total number as the confidence. If the highest confidence exceeds the threshold, it is determined as this category. For example, among 1000 sampling points, the drone category accounts for 820, the confidence is 0.82, which exceeds the threshold of 0.75, and it is determined as a drone signal.

[0101] Exemplarily, in the scenario of UAV signal recognition at the site of a large-scale event, multi-dimensional features of a certain sub-electromagnetic signal are extracted, including a time-domain variance of 0.15, a frequency-bandwidth of 18 MHz, and an azimuth-angle fluctuation of ±2 degrees. Based on the posterior probability information output by the Bayesian network, the initial sampling step sizes are set to 0.06 in the time domain, 1.2 MHz in the frequency domain, and 0.6 degrees in the spatial domain, and the sampling density is 10 points per unit space. After generating an initial candidate sampling point set through the Markov chain Monte Carlo algorithm, a distribution dispersion of 0.18 is detected, exceeding the preset threshold. The step sizes are dynamically adjusted to 0.04 in the time domain, 0.8 MHz in the frequency domain, and 0.4 degrees in the spatial domain, and the sampling density is increased to 15 points / unit. After three iterations of optimization, the dispersion is reduced to 0.009, and the proportion of UAV categories in the candidate point set reaches 85%. Since the confidence level of 0.85 exceeds the preset threshold of 0.8, it is determined that the signal is a UAV signal.

[0102] Based on the same concept, an embodiment of the present application provides a UAV signal recognition system in a complex electromagnetic environment. The following will combine Figure 4 to describe in detail the UAV signal recognition system in a complex electromagnetic environment provided by the embodiment of the present application.

[0103] Figure 4 is a structural block diagram of a UAV signal recognition system in a complex electromagnetic environment shown in an embodiment of the present application.

[0104] As Figure 4 shown, the UAV signal recognition system in a complex electromagnetic environment may include: An acquisition module 410, configured to reconstruct the target morphological parameters of the liquid metal antenna by adjusting the morphological form of the liquid metal channel of the liquid metal antenna, and acquire the omnidirectional electromagnetic signals in the characteristic frequency band of the UAV in the target area; A generation module 420, configured to generate spatial distribution data including the UAV signal direction and the interference signal direction according to the phase difference of the electromagnetic signals in the omnidirectional electromagnetic signals and the preset signal characteristics of the UAV and the interference signals; An adjustment module 430, configured to control the deformation of the liquid metal droplets of the liquid metal antenna by adjusting the electro-wetting control voltage parameters according to the spatial distribution data, so that the main lobe direction of the liquid metal antenna points to the UAV signal direction, and the sidelobe suppression area of the liquid metal antenna covers the spatial distribution range of the interference signal direction; The acquisition module 410 is further configured to use the liquid metal antenna with adjusted lobes to acquire the electromagnetic signal set in the UAV signal direction, and extract the multi-dimensional signal features of each sub-electromagnetic signal from the electromagnetic signal set. The multi-dimensional signal features include time-domain fluctuation features, frequency-domain distribution features, and spatial-domain direction features; An input module 440, configured to input the multi-dimensional signal features of each sub-electromagnetic signal into the constructed joint probability distribution model to obtain the posterior probability information of the electromagnetic signal, where the joint probability distribution model is constructed by a Bayesian network based on historical drone signals and historical interference signals; A determination module 450, configured to perform adaptive sampling on the basis of the posterior probability information of the electromagnetic signal by using Markov chain Monte Carlo, obtain the signal classification confidence level through iterative convergence, and determine the identification information of the sub-electromagnetic signal according to a preset confidence level threshold.

[0105] In one embodiment, the acquisition module 410 is further configured to perform blind source separation processing on the electromagnetic signal set before extracting the multi-dimensional signal features of each sub-electromagnetic signal from the electromagnetic signal set, and based on the signal time-frequency sparsity hypothesis and independent component analysis, separate multiple independent signal source electromagnetic signals; extract the multi-dimensional signal features of each independent signal source electromagnetic signal from the multiple independent signal source electromagnetic signals.

[0106] In one embodiment, the generation module 420 is further configured to perform signal demodulation and protocol parsing on each electromagnetic signal in the omnidirectional electromagnetic signal to obtain the demodulation protocol information including the demodulation information and protocol information of each electromagnetic signal before generating the spatial distribution data including the directions of the drone signal and the interference signal according to the phase difference of the electromagnetic signals in the omnidirectional electromagnetic signal and the preset signal features of the drone and the interference signal; generate the spatial distribution data including the directions of the drone signal and the interference signal according to the phase difference and demodulation protocol information of each electromagnetic signal in the omnidirectional electromagnetic signal and the preset signal features of the drone and the interference signal.

[0107] In one embodiment, the adjustment module 430 is further configured to predict the moving direction of the drone signal through Kalman filtering according to the preset historical spatial distribution data before controlling the deformation of the liquid metal droplets of the liquid metal antenna by adjusting the electro-wetting control voltage parameter according to the spatial distribution data, so that the main lobe direction of the liquid metal antenna points to the drone signal direction and the side lobe direction of the liquid metal antenna points to the interference signal direction; control the deformation of the liquid metal droplets of the liquid metal antenna by adjusting the electro-wetting control voltage parameter according to the moving direction of the drone signal, so that the main lobe direction of the liquid metal antenna points to the moving direction of the drone signal.

[0108] In one embodiment, the acquisition module 410 is specifically configured to determine target morphological parameters of the liquid metal channel based on the center frequency and bandwidth of the UAV characteristic frequency band. The target morphological parameters include the target continuous length of the liquid metal filling path of the liquid metal antenna and the target branch angle. An electrowetting control voltage sequence is generated according to the target continuous length and the target branch angle. By gradually loading the single-stage voltage value in the electrowetting control voltage sequence, the liquid metal droplets of the liquid metal antenna are driven to deform along the filling path until the liquid metal distribution state meets the target continuous length and the target branch angle. Under the condition that the liquid metal distribution state meets the target continuous length and the target branch angle, omnidirectional electromagnetic signals in all directions within the target area are acquired.

[0109] In one embodiment, the adjustment module 430 is specifically configured to determine the angular deviation between the main lobe direction of the liquid metal antenna and the UAV signal direction, and the spatial coverage deviation between the sidelobe suppression area of the liquid metal antenna and the interference signal direction based on the UAV signal direction and the interference signal direction in the spatial distribution data. A control voltage correction value of the electrowetting control voltage parameter is generated according to the angular deviation and the spatial coverage deviation. The control voltage correction value includes a main lobe direction alignment component and a sidelobe suppression enhancement component. The control voltage correction value is decomposed into multiple voltage gradient values, and the voltage gradient values are sequentially applied to the electrode units of the liquid metal antenna until the angular difference between the main lobe direction and the UAV signal direction is less than a preset difference threshold, and the spatial distribution range of the interference signal direction is covered by the sidelobe suppression area.

[0110] In one embodiment, the determination module 450 is specifically configured to determine an initial sampling step and an initial sampling density based on the electromagnetic signal posterior probability information. The initial sampling step and the initial sampling density correspond to the dimensional distribution of the multi-dimensional signal features. Within the range defined by the initial sampling step and the initial sampling density, probability sampling is performed on the dimensional distribution of the multi-dimensional signal features to generate a candidate sampling point set. The candidate sampling point set includes a signal feature combination corresponding to the electromagnetic signal posterior probability information. The initial sampling step and the initial sampling density are adjusted according to the distribution dispersion of the candidate sampling point set, and probability sampling is repeatedly performed until the distribution dispersion of the candidate sampling point set is lower than a preset convergence threshold. The signal classification confidence of each signal category is calculated based on the proportion of the number of each signal category in the candidate sampling point set. When the signal classification confidence with the largest value is greater than a preset confidence threshold, the signal category corresponding to the signal classification confidence is determined as the identification information of the sub-electromagnetic signal.

[0111] Figure 4 Each module in the system shown has the function of implementing Figures 1 to 3 each step in and can achieve its corresponding technical effects. For the sake of brevity, it will not be elaborated here.

[0112] Figure 5The figure shows a schematic diagram of the hardware structure of an electronic device provided by an embodiment of the present application.

[0113] The electronic device may include a processor 510 and a memory 520 storing computer program instructions.

[0114] Specifically, the above-mentioned processor 510 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0115] The memory 520 may include a mass storage for data or instructions. By way of example and not limitation, the memory 520 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disc, a magneto-optical disc, a magnetic tape, or a universal serial bus (USB) drive or a combination of two or more of these. In a suitable case, the memory 520 may include a removable or non-removable (or fixed) medium. In a suitable case, the memory 520 may be internal or external to the integrated gateway disaster recovery device. In a particular embodiment, the memory 520 is a non-volatile solid-state memory.

[0116] The memory may include a read-only memory (ROM), a random access memory (RAM), a magnetic disk storage media device, an optical storage media device, a flash memory device, an electrical, optical, or other physical / tangible memory storage device. Thus, in general, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method according to the first aspect of the present disclosure.

[0117] The processor 510 reads and executes the computer program instructions stored in the memory 520 to implement any one of the methods for identifying UAV signals in a complex electromagnetic environment in the above embodiments.

[0118] In one example, the electronic device may further include a communication interface 530 and a bus 540. Among them, as Figure 5 shown, the processor 510, the memory 520, and the communication interface 530 are connected through the bus 540 to complete communication with each other.

[0119] The communication interface 530 is mainly used to implement communication between the modules, devices, units, and / or devices in the embodiments of the present application.

[0120] The bus 540 includes hardware, software, or both, and couples the components of the online data flow metering device to each other. By way of example and not limitation, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a memory bus, a MicroChannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses or a combination of two or more of these. Where appropriate, the bus 540 may include one or more buses. Although the embodiments of the present application describe and illustrate specific buses, the present application contemplates any suitable bus or interconnect.

[0121] The electronic device can execute the method for identifying drone signals in a complex electromagnetic environment in the embodiments of the present application, so as to implement the combination Figures 1 to 3 of the method for identifying drone signals in the described complex electromagnetic environment.

[0122] In addition, in combination with the method for identifying drone signals in a complex electromagnetic environment in the above embodiments, the embodiments of the present application can be implemented by providing a computer-readable storage medium. Computer program instructions are stored on the computer-readable storage medium; when the computer program instructions are executed by a processor, any one of the methods for identifying drone signals in the above embodiments is implemented.

[0123] It should be clear that the present application is not limited to the specific configurations and processes described above and illustrated in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and illustrated as examples. However, the method process of the present application is not limited to the specific steps described and illustrated, and those skilled in the art can make various changes, modifications, and additions, or change the order between steps after understanding the spirit of the present application.

[0124] The functional blocks shown in the above-described structural block diagrams can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, it can be, for example, an electronic circuit, an application-specific integrated circuit (ASIC), appropriate firmware, a plug-in, a functional card, and so on. When implemented in software, the elements of the present application are programs or code segments for performing the required tasks. The program or code segment can be stored in a machine-readable medium or transmitted over a transmission medium or communication link via a data signal carried in a carrier wave. "Machine-readable medium" can include any medium that can store or transmit information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical discs, hard disks, fiber optic media, radio frequency (RF) links, and so on. The code segment can be downloaded via a computer network such as the Internet, an intranet, and so on.

[0125] It should also be noted that the exemplary embodiments mentioned in the present application describe some methods or systems based on a series of steps or devices. However, the present application is not limited to the order of the above steps, that is, the steps can be executed in the order mentioned in the embodiments, or different from the order in the embodiments, or several steps can be executed simultaneously.

[0126] The aspects of the present application have been described above with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present application. It should be understood that each block in the flowcharts and / or block diagrams, and the combinations of blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing devices to produce a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices enable the implementation of the functions / actions specified in one or more blocks of the flowcharts and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It should also be understood that each block in the block diagrams and / or flowcharts, and the combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by dedicated hardware for performing the specified functions or actions, or by a combination of dedicated hardware and computer instructions.

[0127] As described above, this is only the specific implementation manner of the present application. Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, modules, and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein. It should be understood that the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of various equivalent modifications or substitutions, and these modifications or substitutions should all be covered within the protection scope of the present application.

Claims

1. A method for identifying UAV signals in a complex electromagnetic environment, characterized in that, Including: Reconstructing the target morphological parameters of the liquid metal antenna by adjusting the morphological form of the liquid metal channel of the liquid metal antenna, and collecting omnidirectional electromagnetic signals in the characteristic frequency band of the unmanned aerial vehicle in the target area; Generating spatial distribution data including the directions of the unmanned aerial vehicle signals and the interference signals according to the phase differences of the electromagnetic signals in the omnidirectional electromagnetic signals and the preset signal characteristics of the unmanned aerial vehicle and the interference signals; According to the spatial distribution data, controlling the deformation of the liquid metal droplets of the liquid metal antenna by adjusting the electro-wetting regulation voltage parameters, so that the main lobe direction of the liquid metal antenna points to the direction of the unmanned aerial vehicle signals, and the sidelobe suppression area of the liquid metal antenna covers the spatial distribution range of the interference signal direction; Using the liquid metal antenna with the lobe adjusted, collecting the electromagnetic signal set in the direction of the unmanned aerial vehicle signals, and extracting the multi-dimensional signal characteristics of each sub-electromagnetic signal from the electromagnetic signal set, where the multi-dimensional signal characteristics include time-domain fluctuation characteristics, frequency-domain distribution characteristics, and spatial-domain direction characteristics; Inputting the multi-dimensional signal characteristics of each sub-electromagnetic signal into the constructed joint probability distribution model to obtain the posterior probability information of the electromagnetic signals, where the joint probability distribution model is constructed by a Bayesian network according to historical unmanned aerial vehicle signals and historical interference signals; Adopting Markov chain Monte Carlo to perform adaptive sampling on the basis of the posterior probability information of the electromagnetic signals, obtaining the signal classification confidence through iterative convergence, and determining the identification information of the sub-electromagnetic signals according to the preset confidence threshold.

2. The method according to claim 1, wherein Before extracting the multi-dimensional signal characteristics of each sub-electromagnetic signal from the electromagnetic signal set, the method further includes: Performing blind source separation processing on the electromagnetic signal set, and separating out electromagnetic signals of multiple independent signal sources based on the assumption of signal time-frequency sparsity and independent component analysis; The extracting the multi-dimensional signal characteristics of each sub-electromagnetic signal from the electromagnetic signal set includes: Extracting the multi-dimensional signal characteristics of each independent signal source electromagnetic signal from the electromagnetic signals of the multiple independent signal sources.

3. The method according to claim 2, characterized in that, Before generating the spatial distribution data including the directions of the unmanned aerial vehicle signals and the interference signals according to the phase differences of the electromagnetic signals in the omnidirectional electromagnetic signals and the preset signal characteristics of the unmanned aerial vehicle and the interference signals, the method further includes: Performing signal demodulation and protocol parsing on each electromagnetic signal in the omnidirectional electromagnetic signals to obtain demodulation protocol information including the demodulation information and protocol information of each electromagnetic signal; The generating the spatial distribution data including the directions of the unmanned aerial vehicle signals and the interference signals according to the phase differences of the electromagnetic signals in the omnidirectional electromagnetic signals and the preset signal characteristics of the unmanned aerial vehicle and the interference signals includes: Generating the spatial distribution data including the directions of the unmanned aerial vehicle signals and the interference signals according to the phase differences of each electromagnetic signal in the omnidirectional electromagnetic signals, the demodulation protocol information, and the preset signal characteristics of the unmanned aerial vehicle and the interference signals.

4. The method according to claim 3, wherein Before controlling the deformation of the liquid metal droplets of the liquid metal antenna by adjusting the electro-wetting control voltage parameters according to the spatial distribution data, so that the main lobe direction of the liquid metal antenna points to the UAV signal direction and the side lobe direction of the liquid metal antenna points to the interference signal direction, the method further includes: Predicting the moving direction of the UAV signal through Kalman filtering according to the preset historical spatial distribution data; Controlling the deformation of the liquid metal droplets of the liquid metal antenna by adjusting the electro-wetting control voltage parameters according to the moving direction of the UAV signal, so that the main lobe direction of the liquid metal antenna points to the moving direction of the UAV signal.

5. The method according to claim 1, characterized in that, Reconstructing the target morphological parameters of the liquid metal antenna by adjusting the morphological form of the liquid metal channels of the liquid metal antenna, and collecting the omnidirectional electromagnetic signals of the UAV characteristic frequency band in the target area, including: Determining the target morphological parameters of the liquid metal channels based on the center frequency and bandwidth of the UAV characteristic frequency band, where the target morphological parameters include the target continuous length and target branch angle of the liquid metal filling path of the liquid metal antenna; Generating an electro-wetting control voltage sequence according to the target continuous length and the target branch angle, and driving the deformation of the liquid metal droplets of the liquid metal antenna along the filling path by gradually loading the single-stage voltage value in the electro-wetting control voltage sequence until the liquid metal distribution state meets the target continuous length and the target branch angle; Collecting the omnidirectional electromagnetic signals in all directions in the target area under the condition that the liquid metal distribution state meets the target continuous length and the target branch angle.

6. The method according to claim 1, wherein Controlling the deformation of the liquid metal droplets of the liquid metal antenna by adjusting the electro-wetting control voltage parameters according to the spatial distribution data, so that the main lobe direction of the liquid metal antenna points to the UAV signal direction, and the spatial distribution range of the side lobe suppression area of the liquid metal antenna covers the interference signal direction, including: Based on the UAV signal direction and the interference signal direction in the spatial distribution data, determining the angle deviation between the main lobe direction of the liquid metal antenna and the UAV signal direction, and the spatial coverage deviation between the side lobe suppression area of the liquid metal antenna and the interference signal direction; Generating a regulation voltage correction value of the electro-wetting control voltage parameters according to the angle deviation and the spatial coverage deviation, where the regulation voltage correction value includes a main lobe direction alignment component and a side lobe suppression enhancement component; Decomposing the regulation voltage correction value into multiple voltage gradient values, and sequentially applying the voltage gradient values to the electrode units of the liquid metal antenna until the angle difference between the main lobe direction and the UAV signal direction is less than a preset difference threshold, and the spatial distribution range of the side lobe suppression area covers the interference signal direction.

7. The method according to claim 1, characterized in that Adaptive sampling is performed on the posterior probability information of the electromagnetic signal by using Markov chain Monte Carlo, the signal classification confidence is obtained through iterative convergence, and the identification information of the sub-electromagnetic signal is determined according to the preset confidence threshold, including: Determine an initial sampling step size and an initial sampling density based on the posterior probability information of the electromagnetic signal, where the initial sampling step size and the initial sampling density correspond to the dimensional distribution of the multi-dimensional signal features; Within the range defined by the initial sampling step size and the initial sampling density, perform probability sampling on the dimensional distribution of the multi-dimensional signal features to generate a set of candidate sampling points, where the set of candidate sampling points includes signal feature combinations corresponding to the posterior probability information of the electromagnetic signal; Adjust the initial sampling step size and the initial sampling density according to the distribution dispersion of the set of candidate sampling points, and repeatedly perform probability sampling until the distribution dispersion of the set of candidate sampling points is lower than a preset convergence threshold; Calculate the signal classification confidence of each signal category based on the proportion of the number of each signal category in the set of candidate sampling points. In the case where the signal classification confidence with the largest value is greater than the preset confidence threshold, determine the signal category corresponding to the signal classification confidence as the identification information of the sub-electromagnetic signal.

8. A drone signal recognition system under complex electromagnetic environment, characterized in that, The system includes: An acquisition module, configured to reconstruct the target morphological parameters of the liquid metal antenna by adjusting the morphological parameters of the liquid metal channels of the liquid metal antenna, and acquire omnidirectional electromagnetic signals in the characteristic frequency band of the unmanned aerial vehicle in the target area; A generation module, configured to generate spatial distribution data including the signal direction of the unmanned aerial vehicle and the signal direction of the interference signal according to the phase difference of the electromagnetic signals in the omnidirectional electromagnetic signals and the preset signal characteristics of the unmanned aerial vehicle and the interference signal; An adjustment module, configured to control the deformation of the liquid metal droplets of the liquid metal antenna by adjusting the electro-wetting control voltage parameters according to the spatial distribution data, so that the main lobe direction of the liquid metal antenna points to the signal direction of the unmanned aerial vehicle, and the sidelobe suppression area of the liquid metal antenna covers the spatial distribution range of the signal direction of the interference signal; The acquisition module is further configured to use the liquid metal antenna after lobe adjustment to acquire a set of electromagnetic signals in the signal direction of the unmanned aerial vehicle, and extract multi-dimensional signal features of each sub-electromagnetic signal from the set of electromagnetic signals, where the multi-dimensional signal features include time-domain fluctuation features, frequency-domain distribution features, and spatial-domain direction features; An input module, configured to input the multi-dimensional signal features of each sub-electromagnetic signal into a constructed joint probability distribution model to obtain posterior probability information of the electromagnetic signal, where the joint probability distribution model is constructed by a Bayesian network according to historical unmanned aerial vehicle signals and historical interference signals; A determination module, configured to perform adaptive sampling on the basis of the posterior probability information of the electromagnetic signal by using Markov chain Monte Carlo, obtain the signal classification confidence through iterative convergence, and determine the identification information of the sub-electromagnetic signal according to the preset confidence threshold.

9. An electronic device, characterized in that, The device includes: a processor and a memory storing computer program instructions; When the processor executes the computer program instructions, the method for identifying unmanned aerial vehicle signals in a complex electromagnetic environment according to any one of claims 1-7 is implemented.

10. A computer-readable storage medium, characterized in that, Computer program instructions are stored on the computer-readable storage medium, and when the computer program instructions are executed by a processor, the method for identifying an unmanned aerial vehicle signal in a complex electromagnetic environment according to any one of claims 1-7 is implemented.

Images