Unmanned aerial vehicle identification method and related device

Abstract

The invention discloses an unmanned aerial vehicle identification method and a related device, and relates to the field of unmanned aerial vehicles, and the method comprises the steps: obtaining a mixed unmanned aerial vehicle signal of a plurality of unmanned aerial vehicles, and separating the mixed unmanned aerial vehicle signal into a plurality of unmanned aerial vehicle signals; processing each separated unmanned aerial vehicle signal as follows: extracting a current feature vector and a signal basic parameter of the unmanned aerial vehicle signal; determining a non-black flight unmanned aerial vehicle signal most similar to the unmanned aerial vehicle signal according to the signal basic parameter, and obtaining a preset feature vector; calculating a feature deviation between each feature included in the current feature vector and each corresponding feature included in the preset feature vector; performing weighted summation on the feature deviation of each feature and the preset weight of each feature to obtain a judgment score of the unmanned aerial vehicle signal; and if the determination score is greater than a preset score threshold, determining that the unmanned aerial vehicle to which the unmanned aerial vehicle signal belongs is a black flight unmanned aerial vehicle. According to the method, whether the unmanned aerial vehicle signal is the black flight unmanned aerial vehicle or not is judged according to the multi-dimensional features, accurate control is performed on the unmanned aerial vehicle, and effective control of a coexistence scene of multiple unmanned aerial vehicles is realized.

Description

Technical Field

[0001] This application relates to the field of unmanned aerial vehicle (UAV) technology, and in particular to a UAV identification method and related apparatus. Background Technology

[0002] In scenarios where multiple drones coexist, several drones operate together or coordinate their operations in the same environment. However, among these drones, there are often unauthorized drones that interfere with the flight safety of other drones. Unauthorized drones can refer to drones that are not registered under their real names, do not have flight permits, or engage in illegal or irregular flight behaviors such as violating no-fly zones or altitude restrictions.

[0003] Currently, indiscriminate interference from unauthorized drones flying in multi-drone scenarios can often affect the flight of other drones. Therefore, how to effectively manage unauthorized drones in multi-drone coexistence scenarios is a technical problem that urgently needs to be solved. Summary of the Invention

[0004] In view of the above problems, this application provides a drone identification method and related device to effectively control unauthorized drones in scenarios where multiple drones coexist. The specific solution is as follows:

[0005] The first aspect of this application provides a method for identifying unmanned aerial vehicles (UAVs), the method comprising:

[0006] Acquire mixed drone signals from multiple drones, separate the mixed drone signals, and obtain multiple drone signals;

[0007] The signals from each separated UAV are processed as follows:

[0008] Extract the current feature vector and basic signal parameters of the UAV signal;

[0009] Based on the signal basic parameters, determine the non-black flight drone signal that is most similar to the drone signal stored in the preset database, and obtain the preset feature vector of the determined non-black flight drone signal.

[0010] Calculate the feature deviation between each feature included in the current feature vector and each corresponding feature included in the preset feature vector;

[0011] The feature deviations of each feature and the preset weights of each feature are weighted and summed to obtain the judgment score of the UAV signal;

[0012] If the determination score of the drone signal is greater than a preset score threshold, then the drone to which the drone signal belongs is determined to be an unauthorized drone.

[0013] One possible implementation also includes:

[0014] During the signal window period of the unauthorized drone, an interference signal that has been attenuated and has been modified with added noise is sent to the unauthorized drone.

[0015] In one possible implementation, the drone identification method further includes, prior to separating the hybrid drone signals:

[0016] The hybrid UAV signal is processed by interference removal, noise removal, distortion compensation, and signal correction.

[0017] In one possible implementation, separating the hybrid UAV signals includes:

[0018] Remove the zero-frequency signals from the hybrid drone signals. After removing the zero-frequency signals, construct an objective function based on the non-Gaussianity parameters of the hybrid drone signals.

[0019] Construct a signal separation matrix based on the separation requirements of the hybrid UAV signals;

[0020] The signal separation matrix is ​​optimized using an optimization algorithm;

[0021] The hybrid UAV signals are separated into a set of signals using an optimized signal separation matrix. The objective function value is then calculated based on this set of signals to evaluate the degree of independence of the set of signals.

[0022] Calculate the magnitude of change in the objective function value, and determine the relationship between the magnitude of change in the objective function value and the threshold.

[0023] If the change in the value of the objective function is not less than the threshold, then the optimized signal separation matrix is ​​used as the signal separation matrix, and the process returns to the step of optimizing the signal separation matrix using the optimization algorithm.

[0024] If the change in the objective function value is less than the threshold, an optimized signal separation matrix is ​​output, and the hybrid UAV signal is separated using the optimized signal separation matrix to obtain multiple separated UAV signals.

[0025] In one possible implementation, the drone identification method further includes, before processing each separated drone signal individually:

[0026] The signals of all separated drones are scaled to ensure that their amplitudes are within the same range, their formats are set to the same format, and their time steps are aligned.

[0027] In one possible implementation, extracting the current feature vector of the UAV signal includes:

[0028] Extract multiple frequency features and multiple time features from the UAV signal, and convert the multiple frequency features and multiple time features of the UAV signal into the current feature vector;

[0029] Among them, several frequency characteristics include the center frequency and effective bandwidth of the UAV signal, and several time characteristics include the baud rate, frame period and single frame signal duration of the UAV signal.

[0030] In one possible implementation, the basic signal parameters include frequency band, modulation scheme, and signal bandwidth;

[0031] The step of determining the non-black-flying drone signal most similar to the drone signal from the non-black-flying drone signals stored in a preset database based on the signal's basic parameters includes:

[0032] The drone signals and the non-black drone signals stored in the preset database are classified according to the classification criteria of the frequency band, the modulation method, and the signal bandwidth, respectively.

[0033] If the drone signal and the target non-black-flying drone signal are of the same type under at least two classification criteria in the frequency band, the modulation method, and the signal bandwidth, then the target non-black-flying drone signal is the non-black-flying drone signal most similar to the drone signal.

[0034] A second aspect of this application provides a drone identification device, the drone identification device comprising:

[0035] The acquisition unit is used to acquire mixed drone signals from multiple drones, separate the mixed drone signals, and obtain multiple drone signals.

[0036] For each separated UAV signal, the following unit is executed:

[0037] An extraction unit is used to extract the current feature vector and basic signal parameters of the UAV signal;

[0038] The classification unit is used to determine the non-black-flying drone signal that is most similar to the drone signal from the non-black-flying drone signals stored in the preset database according to the basic signal parameters, and to obtain the preset feature vector of the determined non-black-flying drone signal.

[0039] A calculation unit is used to calculate the feature deviation between each feature included in the current feature vector and each corresponding feature included in the preset feature vector;

[0040] The determination unit is used to perform a weighted summation of the feature deviations of each feature and the preset weights of each feature to obtain the determination score of the drone signal. If the determination score of the drone signal is greater than the preset score threshold, the drone to which the drone signal belongs is determined to be an unauthorized drone.

[0041] A third aspect of this application provides an electronic device, comprising at least one processor and a memory connected to the processor, wherein:

[0042] The memory is used to store computer programs;

[0043] The processor is used to execute the computer program so that the electronic device can implement the drone identification method of the first aspect or any implementation thereof.

[0044] The fourth aspect of this application provides a computer program product including computer-readable instructions that, when executed on an electronic device, cause the electronic device to implement the unmanned aerial vehicle (UAV) identification method described in the first aspect or any implementation thereof.

[0045] Based on the above technical solution, this application provides a drone identification method and related apparatus. The method acquires mixed drone signals from multiple drones, separates the mixed drone signals, and performs unified processing on each separated drone signal: extracting the current feature vector and basic signal parameters of the drone signal; determining the most similar non-black-flying drone signal from a preset database of non-black-flying drone signals based on the basic signal parameters, and obtaining a preset feature vector for the non-black-flying drone signal; calculating the feature deviations of each feature included in the current feature vector and the corresponding features included in the preset feature vector; weighted summing of the feature deviations of each feature and the preset weights of each feature to obtain a determination score for the drone signal; if the determination score of the drone signal is greater than a preset score threshold, the drone to which the drone signal belongs is determined to be a black-flying drone. This method first separates the mixed drone signals collected in a multi-drone coexistence scenario to facilitate subsequent processing. Then, it uses the feature vector of non-black-flying drone signals as the basis for judgment to determine whether a drone signal is a black-flying drone from multiple dimensions, reducing the possibility of misjudgment. Once a black-flying drone is identified among the multiple drones, it is easy to accurately control it, thereby reducing the impact on other drones and achieving effective control of black-flying drones in a multi-drone coexistence scenario. Attached Figure Description

[0046] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and the originals and elements are not necessarily drawn to scale.

[0047] Figure 1 A flowchart illustrating a drone identification method provided in an embodiment of this application;

[0048] Figure 2 This is a schematic diagram of the structure of a drone identification device provided in an embodiment of this application;

[0049] Figure 3 This application provides a hardware structure block diagram of an electronic device. Detailed Implementation

[0050] The embodiments of this application are described below with reference to the accompanying drawings. The terminology used in the implementation section of this application is for explaining specific embodiments only and is not intended to limit the scope of this application.

[0051] The embodiments of this application will now be described with reference to the accompanying drawings. Those skilled in the art will recognize that, with technological advancements and the emergence of new scenarios, the technical solutions provided in the embodiments of this application are equally applicable to similar technical problems.

[0052] The terms “comprising” and “having”, and any variations thereof, in the specification and accompanying drawings of this application are intended to cover non-exclusive inclusion, such that a process, method, system, product, or apparatus that comprises a series of elements is not necessarily limited to those elements, but may include other elements not expressly listed or inherent to such process, method, product, or apparatus.

[0053] To address the aforementioned problems, this application provides a method for identifying unmanned aerial vehicles (UAVs). The UAV identification method of this application embodiment will be described in detail below with reference to the accompanying drawings.

[0054] Reference Figure 1 , Figure 1 This is a flowchart illustrating a drone identification method provided in an embodiment of this application, as shown below. Figure 1 As shown in the figure, the drone identification method provided in this application embodiment may include steps S10 to S16, which are described in detail below.

[0055] S10. Acquire mixed drone signals from multiple drones, separate the mixed drone signals, and obtain multiple drone signals.

[0056] Hybrid UAV signals can refer to UAV signals collected from multiple frequency bands in a scenario where multiple UAVs coexist. Furthermore, signal acquisition must cover all commonly used frequency bands of the UAVs. UAV signals can refer to various wireless signals transmitted between the UAV and remote controllers, ground stations, satellites, base stations, or other devices during flight and operation to achieve functions such as control command interaction, data transmission, positioning and navigation, and status feedback. Specifically, the hybrid UAV signals collected in this embodiment can be I / Q signals. I / Q signals (In-phase / Quadrature Signals) can refer to a basic complex number representation of a signal, decomposing the signal into two real signal components with orthogonal phases and independent amplitudes (I-path is the in-phase component, Q-path is the quadrature component), and fully describing the amplitude, frequency, and phase information of the signal through a complex form (I+jQ, where j is the imaginary unit).

[0057] Specifically, this embodiment utilizes a USRP device to acquire hybrid drone signals. The USRP device is connected to a host device (used for subsequent signal processing and identification of unauthorized drones) via USB 3.0. A full-band antenna is connected to the USRP device's RF interface. The USRP device acquires drone signals across multiple frequency bands. During signal acquisition, the USRP device can cycle through each frequency band at switching intervals, effectively avoiding signal interference during band switching. The USRP (Universal Software Radio Peripheral) device refers to a software radio core hardware platform capable of bidirectional conversion between RF signals and digital baseband signals, working in conjunction with the host device (computer) to capture wireless signals.

[0058] In this embodiment, after acquiring the mixed drone signal, before separating the mixed drone signal, it is necessary to perform interference removal, noise removal, distortion compensation, and signal correction on the mixed drone signal to improve signal quality. Optionally, this embodiment can perform the aforementioned processing on the mixed drone signal sequentially.

[0059] Specifically, this embodiment addresses interference signal removal from hybrid drone signals. It performs FFT analysis on the hybrid drone signals to locate the center frequency of strong interference signals (such as Wi-Fi signals, a type of wireless local area network radio frequency signal), filtering out these strong interference signals. Furthermore, this embodiment uses a Chebyshev Type I bandstop filter to suppress specific narrowband interference in the hybrid drone signals. FFT (Fast Fourier Transform) refers to an algorithm that rapidly transforms a time-domain signal to a frequency-domain signal to determine its composition (the superposition of sine waves with different frequencies, amplitudes, and phases). A Chebyshev Type I bandstop filter is a linear time-invariant filter based on Chebyshev polynomial design, which can accurately attenuate interference signals (such as Bluetooth signals) in the stopband while preserving the amplitude and phase characteristics of the target signal (such as drone signals) in the passband to the maximum extent, ultimately outputting a signal with suppressed interference and controllable target signal distortion.

[0060] Noise removal for hybrid drone signals. Since broadband thermal noise (inherent random noise generated by electron thermal motion) exists in the hybrid drone signal, this embodiment uses wavelet transform to decompose the hybrid drone signal into low-frequency components (core signal features) and high-frequency components (noise, signal edges, etc.). Wavelet thresholding is then applied to further threshold the high-frequency components, retaining the signal-related high-frequency components and suppressing the noise-related high-frequency components. The noise-suppressed high-frequency components and low-frequency components are then reconstructed using inverse wavelet transform to obtain the denoised hybrid drone signal.

[0061] Distortion compensation for hybrid UAV signals. Due to the Doppler effect during high-speed UAV flight, the acquired signal exhibits a frequency offset. Therefore, this embodiment employs FFT frequency offset estimation based on pilot signals for correction. The pilot position of the hybrid UAV signal is located using "sliding window matching," the pilot band signal is extracted, and FFT analysis is performed on it to calculate the actual frequency and frequency offset. Frequency offset compensation is then applied to the entire hybrid UAV signal based on this value. After compensation, the frequency of the pilot band signal is re-monitored to confirm the existence of any frequency offset error.

[0062] Signal correction for hybrid UAV signals. Since there are two receiving channels (I-channel and Q-channel) during hybrid UAV signal acquisition, subtle differences in the hardware of the acquisition equipment can cause amplitude and phase imbalances in the two real signal components. Therefore, this embodiment uses statistical analysis to calculate the differences between the two receiving channels and uses an algorithm to correct the Q-channel signal to make it consistent with the I-channel, ensuring that the signals from the two receiving channels are synchronized and balanced.

[0063] After performing the above processing on the hybrid drone, the hybrid drone signal is then separated. The specific separation process can be described in steps one through seven below:

[0064] Step 1: Remove zero-frequency signals from the preprocessed hybrid UAV signal. After removing zero-frequency signals, construct an objective function based on the non-Gaussianity parameters of the hybrid UAV signal; where,

[0065] Step 2: Construct a signal separation matrix based on the separation requirements of the mixed drone signals. The separation requirements of the mixed drone signals can refer to the number of drone signals required after separation. This separation requirement is related to the matrix dimension of the signal separation matrix. For example, if a mixed drone signal A needs to be separated into independent drone signals B, then the matrix dimension of the signal separation matrix can be B×A.

[0066] Step 3: Optimize the signal separation matrix using an optimization algorithm;

[0067] Step 4: Separate the hybrid UAV signals into a set of signals using the optimized signal separation matrix, and calculate the objective function value based on the set of signals to evaluate the degree of independence of the set of signals;

[0068] Step 5: Calculate the magnitude of change in the objective function value and determine the relationship between the magnitude of change in the objective function value and the threshold.

[0069] Step 6: If the change in the objective function value is not less than the threshold, then use the optimized signal separation matrix as the signal separation matrix and return to step 3.

[0070] Step 7: If the change in the objective function value is less than the threshold, output the optimized signal separation matrix, and use the optimized signal separation matrix to separate the mixed UAV signals to obtain multiple separated UAV signals.

[0071] In step one, in order to eliminate the constant offset in the hybrid drone signal and make the hybrid drone signal fluctuate around the zero point, this embodiment removes the signal (DC component) with a frequency of zero from the hybrid drone signal.

[0072] This embodiment constructs a clear objective function as an indicator for signal separation. During the construction of the objective function, this embodiment can use the core characteristics of the hybrid UAV signals as a basis to convert the independence of the hybrid UAV signals into a quantifiable computational indicator. Specifically, the core characteristics of the hybrid UAV signals can be the non-Gaussianity parameters of the hybrid UAV signals. Non-Gaussianity parameters can refer to indicators used to quantify the difference between the signal distribution and a Gaussian distribution, including kurtosis, negative entropy, etc. Since the non-Gaussianity of the hybrid UAV signals is low, and increases after signal separation, this embodiment can construct an objective function based on the non-Gaussianity parameters of the hybrid UAV signals to measure the degree of independence of the separated hybrid UAV signals.

[0073] In this embodiment, the core function of the objective function is to quantify signal independence. Its value change can directly reflect the optimization progress of the signal separation effect. When the change of the objective function value is not less than the threshold, it can be said that the parameter adjustment of the signal separation matrix can further improve the signal independence, and the signal separation matrix can be further optimized by the optimization algorithm. However, when the change of the objective function value is less than the threshold, it can be said that the parameter adjustment of the signal separation matrix is ​​difficult to significantly improve the signal independence. The output optimized signal separation matrix has already achieved a near-optimal separation result for the mixed UAV signals. Further optimization of the signal separation matrix by the optimization algorithm is no longer meaningful. Therefore, the optimized signal separation matrix can be output, and the mixed UAV signals can be separated using the optimized signal separation matrix to obtain multiple UAV signals. The multiple UAV signals obtained after separation can also be I / Q signals.

[0074] Before processing multiple drone signals, signal scaling can be performed on all separated drone signals to ensure that the amplitude of all separated drone signals is within the same range, the format of all separated drone signals can be set to the same format, and the time steps of all separated drone signals can be aligned.

[0075] Specifically, due to the varying distances between each drone and the signal acquisition device, there are differences in signal strength and format among the multiple drone signals. Therefore, this embodiment selects the maximum amplitude among the multiple drone signals and scales the amplitudes of other signals proportionally to unify the amplitude range of each drone signal, thus resolving the signal strength differences caused by distance. The signal format of each drone signal is then converted to a more precise 32-bit format (to accurately represent the waveform details), segmented into fixed lengths, and each segment is ensured to contain complete signal content. During time step alignment, this embodiment can locate the frame start time of each drone signal based on its frame header marker (a fixed symbol at the beginning of each drone signal segment), adjusting the frame start times of all drone signals to a unified time axis reference (start point alignment), achieving time dimension alignment of multiple drone signals and effectively avoiding time errors that could affect subsequent analysis.

[0076] S11. Process each separated UAV signal as follows:

[0077] S12. Extract the current feature vector and basic signal parameters of the UAV signal;

[0078] S13. Based on the signal basic parameters, determine the non-black-flying drone signal that is most similar to the drone signal from the non-black-flying drone signals stored in the preset database, and obtain the preset feature vector of the determined non-black-flying drone signal.

[0079] Among them, the basic signal parameters can refer to the core physical and timing characteristics of UAV communication, which are fixed by hardware design, general technical standards or transmission protocols. These can include frequency band (the frequency range dimension of the signal), modulation method (the encoded transmission form of the signal) and signal bandwidth (the spectral width occupied by the signal).

[0080] The preset database can refer to a database containing information on multiple legitimate drones. When constructing the preset database, legitimate drone signals can be collected, and their signal features can be extracted and stored in the preset database in the form of preset feature vectors. Legitimate drones can include non-illegal drones in this embodiment.

[0081] The current feature vector can refer to the feature set of time-frequency features of the UAV signal, which may include multiple frequency features and multiple time features. Specifically, in this embodiment, the UAV signal can be subjected to short-time Fourier transform to generate a time-frequency map of the UAV signal. Multiple frequency features and multiple time features of the UAV signal are extracted from the time-frequency map of the UAV signal from both the frequency and time domains. The multiple frequency features and multiple time features of each UAV signal are then converted into the current feature vector.

[0082] This embodiment constructs an individual identifier for each drone by collecting signals from each drone and combining the signal analysis results. This individual identifier facilitates subsequent determination of whether a drone is flying illegally. Furthermore, multi-dimensional feature recognition can effectively improve the robustness of illegal drone identification.

[0083] Each drone signal has a frequency, and it is frequency-hopping rather than fixed. The frequency variation pattern of each drone signal can be a unique characteristic of the drone. In this embodiment, multiple frequency characteristics of each drone signal can include: the center frequency and effective bandwidth of the drone signal. The center frequency can refer to the reference frequency at the center of energy concentration of the drone signal in the frequency domain, and the effective bandwidth can refer to the minimum frequency range that ensures the normal implementation of the core functions of the drone signal (transmission, demodulation, etc.).

[0084] In a time-frequency graph, a drone signal can be represented as a continuous bright frequency band, with the center frequency being the geometric center of this bright frequency band. Specifically, in this embodiment, when obtaining the center frequency, the time-frequency graph can be mapped along a frequency axis, mapping the FFT frequency index to the actual frequency. The frequency-weighted average of the energy within each frame of the drone signal is calculated, and the center frequencies of all frames in the drone signal are statistically analyzed to obtain the center frequency set of the drone signal. The average value of this center frequency set can then be taken as the center frequency of the drone signal. Here, the FFT frequency index can refer to a discrete frequency signal, where each index uniquely corresponds to an actual physical frequency.

[0085] In a time-frequency graph, a drone signal can be represented as a continuous bright band of frequencies, and the effective bandwidth can be the width of this bright band. Specifically, in this embodiment, for the time-frequency graph of a drone, the energy of all time frames corresponding to each frequency index is accumulated, and the two-dimensional energy distribution of the drone signal in time and frequency is converted into a one-dimensional total energy distribution of frequency, thus obtaining the global energy spectrum of the drone signal. The maximum value of the global energy spectrum is determined, and based on this maximum value, the frequency value corresponding to the energy dropping from the maximum value to 3 dB is determined. Within the frequency range formed by the frequency value corresponding to the maximum value and the frequency value corresponding to 3 dB, the maximum frequency and the minimum frequency are determined, and the absolute value of the difference between the maximum frequency and the minimum frequency is taken as the effective bandwidth.

[0086] In this embodiment, several temporal characteristics of each UAV signal include: the baud rate, frame period, and duration of a single frame. The baud rate can refer to the number of symbols (the basic signal units carrying information in communication) transmitted per unit time in UAV communication. The frame period can refer to the time interval between two consecutive frames in UAV communication. The duration of a single frame can refer to the time consumed from the start of transmission of a complete frame until all content of that frame is transmitted.

[0087] Specifically, regarding the baud rate of the UAV signal, this embodiment can obtain it through time periodicity calculation of the UAV signal's time-frequency plot: extract the time-domain envelope of the UAV signal, calculate the autocorrelation function of the time-domain envelope, select the delay time (symbol period) corresponding to the first non-zero peak of the autocorrelation function, and take the reciprocal of this delay time to obtain the baud rate of the UAV signal. Here, the time-domain envelope can refer to the amplitude variation trajectory of a complex signal (such as an I / Q signal) in the time dimension. The time-domain envelope can eliminate interference from signal phase information, thereby focusing on the amplitude variation characteristics of the symbol. The autocorrelation function of the time-domain envelope can be a function describing the similarity of the same time-domain envelope signal at different delay times. It can quantify the repetitive pattern of the signal itself. If the signal has periodicity, the autocorrelation function can have a significant peak at a specific delay.

[0088] Regarding the frame period of the UAV signal, this embodiment obtains the frequency point corresponding to the center frequency from the time-frequency graph of the UAV signal, obtains a series of data corresponding to the frequency point, obtains a one-dimensional time series of center frequency-time, performs Fourier transform on the one-dimensional time series, converts the one-dimensional time series into a frequency domain spectrum, identifies the frequency (frame repetition frequency) corresponding to the spectral peak in the frequency domain spectrum, and takes the reciprocal of the frequency to obtain the frame period of the UAV signal.

[0089] Regarding the duration of a single frame of UAV signal, this embodiment can divide multiple signal frames into valid signal frames and invalid signal frames (noise or no signal segments) based on the total energy and energy threshold of each frame signal, locate the signal start frame and signal end frame, and calculate and determine the duration of a single frame signal based on the signal start frame and signal end frame.

[0090] Of course, in another alternative embodiment, in addition to the frequency features and time features mentioned above, other frequency features and time features can be extracted as features in the feature vector, such as signal strength, encrypted frame structure, etc.

[0091] This embodiment can classify UAV signals and non-illegal UAV signals stored in a preset database based on basic signal parameters, thereby determining the non-illegal UAV signal most similar to the UAV signal. Specifically, this embodiment can classify UAV signals and non-illegal UAV signals stored in the preset database according to classification standards such as frequency band, modulation method, and signal bandwidth. If the UAV signal and the target non-illegal UAV signal belong to the same category under at least two of the classification standards (frequency band, modulation method, and signal bandwidth), then the target non-illegal UAV signal is the non-illegal UAV signal most similar to the UAV signal.

[0092] According to frequency band, drone signals can be divided into ultra-low frequency, low frequency, high frequency, ultra-high frequency, etc., or specifically into frequency bands such as 2.5GHz and 5.8GHz; according to modulation method, drone signals can be divided into orthogonal frequency division multiplexing, phase keying, chirped spread spectrum, etc.; according to signal bandwidth, drone signals can be divided into narrowband, wideband, ultra-wideband, etc.

[0093] This embodiment can classify signals according to multiple classification criteria sequentially, without limiting the order of classification. For example, multiple signals (including drone signals and multiple non-illegal drone signals) can be classified first by frequency band. Signals in the same frequency band can be further classified by modulation method, and signals in the same frequency band and with the same modulation method can be further classified by signal bandwidth. Alternatively, multiple signals can be classified first by modulation method, signals with the same modulation method can be further classified by signal bandwidth, and signals with the same modulation method and signal bandwidth can be further classified by frequency band. Of course, in another optional embodiment, this embodiment can also further classify signals by basic signal parameters such as period and frame structure characteristics.

[0094] In this embodiment, after classifying the drone signal and multiple non-black-flying drone signals according to multiple classification criteria, the non-black-flying drone signal that most frequently belongs to the same category as the drone signal can be selected as the most similar non-black-flying drone signal. For example, a drone signal and a non-black-flying drone signal C are in the same category only under the frequency band classification criterion, while they are in the same category as another non-black-flying drone signal D not only under the frequency band classification criterion, but also under the modulation method classification criterion and the signal bandwidth classification criterion. Therefore, non-black-flying drone signal D has the most frequent occurrences of being in the same category as the drone signal, and non-black-flying drone signal D is the most similar non-black-flying drone signal to the drone signal.

[0095] In this embodiment, after determining the non-unauthorized drone signal that is most similar to the drone signal, the preset feature vector of the non-unauthorized drone signal can be obtained from the preset database and compared with the current feature vector of the drone signal to determine whether the drone signal is from an unauthorized drone.

[0096] S14. Calculate the feature deviations of each feature included in the current feature vector and the corresponding features included in the preset feature vector.

[0097] S15. The feature deviations of each feature and the preset weights of each feature are weighted and summed to obtain the judgment score of the UAV signal.

[0098] S16. If the determination score of the drone signal is greater than the preset score threshold, then the drone to which the drone signal belongs is determined to be an unauthorized drone.

[0099] The preset score threshold can be set in advance and can be flexibly set according to the actual situation.

[0100] In this embodiment, for the current feature vector and the preset feature vector, the numerical deviation between the corresponding features of the current feature vector and the preset feature vector can be calculated to obtain the feature deviation of multiple features between the drone signal and the non-black flight drone. The drone signal is then weighted and summed with the preset weights of multiple features and the feature deviation to generate a judgment score for the drone signal. Based on the judgment score, it is determined whether the drone signal belongs to the black flight drone.

[0101] The preset weights are primarily allocated based on the importance of signal features, which are mainly determined by stability and discriminability. Stability refers to the stability of a feature across different drones. For example, a feature might be highly stable and almost unchanged in a legitimate drone, but exhibit significant deviations in a different drone signal, suggesting that the latter is likely a signal from an unauthorized drone. Therefore, features with higher stability receive higher weights. For instance, center frequency is more stable than instantaneous signal strength (which is more susceptible to distance), so its preset weight should be higher than that of instantaneous signal strength. Discriminability refers to the degree of difference between a feature and unauthorized drones. Higher discriminability indicates a greater difference between the feature and unauthorized drones, resulting in a higher weight. For example, encrypted frame structures are more discriminative than signal bandwidth (where there is overlap between legitimate and unauthorized drones), so their preset weight should be higher than that of signal bandwidth.

[0102] The core function of preset feature weights is to amplify the influence of important features, allowing stable and highly discriminative features to play a leading role in the judgment, avoiding misjudgments caused by accidental deviations of secondary features, and making the judgment results closer to the priorities of the actual scenario.

[0103] In this embodiment, when determining that the drone signal belongs to an unauthorized drone, the start time and duration of the unauthorized drone's signal window can be located based on the frame period and single-frame signal duration of the signal. Within the unauthorized drone's signal window (the effective period of communication transmission), the signal is attenuated according to the actual distance. The attenuation magnitude can be dynamically adjusted according to the unauthorized drone's receiving sensitivity, ensuring that the signal strength drops to a threshold that is receivable but cannot be effectively demodulated. Gaussian white noise or random noise (narrowband Gaussian noise) is added, and the noise bandwidth must be consistent with the target command link bandwidth of the unauthorized drone to prevent noise from spreading to other frequency bands. The attenuated and noise-added interference signal is sent to the unauthorized drone through the antenna. When sending the interference signal, its transmission power can be dynamically adjusted, automatically matching the transmission power according to the distance to the unauthorized drone to prevent excessive power from causing the interference signal to spread. After receiving the interference signal, the unauthorized drone, due to the extremely poor signal quality of the interference signal, causes errors or loss in command parsing. Therefore, the unauthorized drone cuts off the normal communication link with the remote controller, forcing the unauthorized drone into a safety response state (such as hovering or returning to home), thus achieving control of the unauthorized drone.

[0104] This embodiment employs a combination of dynamic power attenuation and Gaussian noise addition. The attenuation amplitude is adjusted in real time according to the receiving sensitivity of the unauthorized drone, ensuring that the signal can be received but cannot be effectively demodulated. Furthermore, it only matches the communication frequency and signal window of the unauthorized drone. Legitimate drones are not affected by the interference signal due to the mismatch in characteristics. Moreover, the unauthorized drone only switches links after receiving the interference signal without losing control, effectively avoiding safety risks.

[0105] This application provides a drone identification method. The method acquires mixed drone signals from multiple drones, separates the mixed drone signals, and processes each separated drone signal uniformly: extracting the current feature vector and basic signal parameters of the drone signal; determining the most similar non-black-flying drone signal from a preset database of non-black-flying drone signals based on the basic signal parameters, and obtaining a preset feature vector for the non-black-flying drone signal; calculating the feature deviations of each feature included in the current feature vector and the corresponding features included in the preset feature vector; weighted summing of the feature deviations and preset weights of each feature to obtain a determination score for the drone signal; if the determination score of the drone signal is greater than a preset score threshold, the drone to which the drone signal belongs is determined to be a black-flying drone. This method first separates the mixed drone signals collected in a multi-drone coexistence scenario to facilitate subsequent processing. Then, it uses the feature vector of non-black-flying drone signals as the basis for judgment to determine whether a drone signal is a black-flying drone from multiple dimensions, reducing the possibility of misjudgment. Once a black-flying drone is identified among the multiple drones, it is easy to accurately control it, thereby reducing the impact on other drones and achieving effective control of black-flying drones in a multi-drone coexistence scenario.

[0106] The above describes a drone identification method provided by the embodiments of this application. The following describes the apparatus for performing the above drone identification method.

[0107] Please see Figure 2 , Figure 2 This is a schematic diagram of the structure of a drone identification device provided in an embodiment of this application. Figure 2 As shown, the drone identification device may include:

[0108] The acquisition unit 100 is used to acquire mixed drone signals from multiple drones, separate the mixed drone signals, and obtain multiple drone signals.

[0109] For each separated UAV signal, the following unit is executed:

[0110] Extraction unit 110 is used to extract the current feature vector and basic signal parameters of the UAV signal;

[0111] The classification unit 120 is used to determine the non-black flying drone signal that is most similar to the drone signal from the non-black flying drone signals stored in the preset database according to the basic signal parameters, and to obtain the preset feature vector of the determined non-black flying drone signal.

[0112] The calculation unit 130 is used to calculate the feature deviation between each feature included in the current feature vector and each corresponding feature included in the preset feature vector;

[0113] The determination unit 140 is used to perform a weighted summation of the feature deviation of each feature and the preset weight of each feature to obtain the determination score of the drone signal. If the determination score of the drone signal is greater than the preset score threshold, the drone to which the drone signal belongs is determined to be an unauthorized drone.

[0114] In one possible implementation, the drone identification device may also include a control unit:

[0115] The control unit is used to send interference signals, which have been attenuated and have noise added, to the unauthorized drones during their signal window.

[0116] In one possible implementation, the drone identification device may further include a preprocessing unit before separating the mixed drone signals:

[0117] The preprocessing unit is used to perform interference removal, noise removal, distortion compensation, and signal correction on the mixed UAV signals.

[0118] In one possible implementation, the acquisition unit 100 separates the hybrid UAV signals, which can be specifically configured as follows:

[0119] The process involves removing zero-frequency signals from the mixed drone signals. After removing zero-frequency signals, an objective function is constructed based on the non-Gaussianity parameter of the mixed drone signals. A signal separation matrix is ​​then constructed based on the separation requirements of the mixed drone signals. This signal separation matrix is ​​optimized using an optimization algorithm. The optimized signal separation matrix is ​​used to separate the mixed drone signals into a group of signals. The objective function value is calculated based on this group of signals to evaluate the independence of the group of signals. The magnitude of the change in the objective function value is calculated, and its relationship with a threshold is determined. If the magnitude of the change in the objective function value is not less than the threshold, the optimized signal separation matrix is ​​used as the signal separation matrix, and the process returns to the step of optimizing the signal separation matrix using the optimization algorithm. If the magnitude of the change in the objective function value is less than the threshold, the optimized signal separation matrix is ​​output, and the mixed drone signals are separated using the optimized signal separation matrix to obtain multiple separated drone signals.

[0120] In one possible implementation, the drone identification device may further include a standardization unit before processing each separated drone signal individually:

[0121] The standardization unit is used to perform signal scaling on all separated UAV signals to ensure that the amplitude of all separated UAV signals is within the same range, set the format of all separated UAV signals to the same format, and align the time steps of all separated UAV signals.

[0122] In one possible implementation, the extraction unit 110 can be specifically configured as follows:

[0123] Extract multiple frequency features and multiple time features from the UAV signal, and convert these features into a current feature vector.

[0124] Among them, several frequency characteristics include the center frequency and effective bandwidth of the UAV signal, and several time characteristics include the baud rate, frame period and single frame signal duration of the UAV signal.

[0125] In one possible implementation, the basic signal parameters include frequency band, modulation scheme, and signal bandwidth;

[0126] In classification unit 120, the non-black flight drone signal most similar to the drone signal is determined from the non-black flight drone signals stored in the preset database based on the signal's basic parameters. This can be specifically configured as follows:

[0127] The drone signal and the non-black drone signal stored in the preset database are classified according to the classification criteria of the frequency band, the modulation method, and the signal bandwidth, respectively. If the drone signal and the target non-black drone signal are of the same type under at least two of the classification criteria of the frequency band, the modulation method, and the signal bandwidth, then the target non-black drone signal is the non-black drone signal most similar to the drone signal.

[0128] This application also provides an electronic device in its embodiments. (See reference...) Figure 3 The diagram illustrates a structural schematic suitable for implementing the electronic device in the embodiments of this application. The electronic device in the embodiments of this application may include, but is not limited to, fixed terminals such as mobile phones, laptops, PDAs (personal digital assistants), PADs (tablet computers), desktop computers, etc. Figure 3 The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0129] like Figure 3 As shown, the electronic device may include a processing unit (e.g., a central processing unit, a graphics processing unit, etc.) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage device 308 into a random access memory (RAM) 303. When the electronic device is powered on, the RAM 303 also stores various programs and data required for the operation of the electronic device. The processing unit 301, ROM 302, and RAM 303 are interconnected via a bus 304. An input / output interface (I / O interface) 305 is also connected to the bus 304.

[0130] Typically, the following devices can be connected to I / O interface 305: input devices 306 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 307 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 308 including, for example, memory cards, hard drives, etc.; and communication devices 309. Communication device 309 allows electronic devices to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 3 Electronic devices with various devices are shown, but it should be understood that it is not required to implement or have all of the devices shown. More or fewer devices may be implemented or have alternatively.

[0131] This application also provides a computer program product including computer-readable instructions, which, when executed on an electronic device, cause the electronic device to perform any of the drone identification methods provided in this application.

[0132] This application also provides a computer-readable storage medium that carries one or more computer programs. When the one or more computer programs are executed by an electronic device, the electronic device is able to perform any of the drone identification methods provided in this application.

[0133] It should also be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. In addition, in the device embodiment drawings provided in this application, the connection relationship between modules indicates that they have a communication connection, which can be implemented as one or more communication buses or signal lines.

[0134] Through the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware, or it can be implemented by special-purpose hardware including application-specific integrated circuits, special-purpose CPUs, special-purpose memory, special-purpose components, etc. Generally, any function performed by a computer program can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be diverse, such as analog circuits, digital circuits, or special-purpose circuits. However, for this application, software program implementation is more often the preferred implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a readable storage medium, such as a computer floppy disk, USB flash drive, mobile hard disk, ROM, RAM, magnetic disk, or optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, training equipment, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0135] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product.

[0136] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, training device, or data center to another website, computer, training device, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a training device or data center that integrates one or more available media. The available media may be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid-state drives (SSDs)).

[0137] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the apparatus embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0138] It is understood that before using the technical solutions disclosed in the various embodiments of this disclosure, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in this disclosure in an appropriate manner in accordance with relevant laws and regulations, and user authorization should be obtained.

[0139] The above are merely embodiments of this application and are not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for identifying unmanned aerial vehicles (UAVs), characterized in that, The drone identification method includes: Acquire mixed drone signals from multiple drones, separate the mixed drone signals, and obtain multiple drone signals; The signals from each separated UAV are processed as follows: Extract the current feature vector and basic signal parameters of the UAV signal; Based on the signal basic parameters, determine the non-black flight drone signal that is most similar to the drone signal stored in the preset database, and obtain the preset feature vector of the determined non-black flight drone signal. Calculate the feature deviation between each feature included in the current feature vector and each corresponding feature included in the preset feature vector; The feature deviations of each feature and the preset weights of each feature are weighted and summed to obtain the judgment score of the UAV signal; If the determination score of the drone signal is greater than a preset score threshold, then the drone to which the drone signal belongs is determined to be an unauthorized drone.

2. The UAV identification method according to claim 1, characterized in that, Also includes: During the signal window period of the unauthorized drone, an interference signal that has been attenuated and has been modified with added noise is sent to the unauthorized drone.

3. The UAV identification method according to claim 1, characterized in that, Before separating the hybrid drone signals, the drone identification method further includes: The hybrid UAV signal is processed by interference removal, noise removal, distortion compensation, and signal correction.

4. The UAV identification method according to claim 1, characterized in that, The separation of the hybrid UAV signals includes: Remove the zero-frequency signals from the hybrid drone signals. After removing the zero-frequency signals, construct an objective function based on the non-Gaussianity parameters of the hybrid drone signals. Construct a signal separation matrix based on the separation requirements of the hybrid UAV signals; The signal separation matrix is ​​optimized using an optimization algorithm; The hybrid UAV signals are separated into a set of signals using an optimized signal separation matrix. The objective function value is then calculated based on this set of signals to evaluate the degree of independence of the set of signals. Calculate the magnitude of change in the objective function value, and determine the relationship between the magnitude of change in the objective function value and the threshold. If the change in the value of the objective function is not less than the threshold, then the optimized signal separation matrix is ​​used as the signal separation matrix, and the process returns to the step of optimizing the signal separation matrix using the optimization algorithm. If the change in the objective function value is less than the threshold, an optimized signal separation matrix is ​​output, and the hybrid UAV signal is separated using the optimized signal separation matrix to obtain multiple separated UAV signals.

5. The UAV identification method according to claim 1, characterized in that, Before processing each separated UAV signal individually, the UAV identification method further includes: The signals of all separated drones are scaled to ensure that their amplitudes are within the same range, their formats are set to the same format, and their time steps are aligned.

6. The UAV identification method according to claim 1, characterized in that, The extraction of the current feature vector of the UAV signal includes: Extract multiple frequency features and multiple time features from the UAV signal, and convert the multiple frequency features and multiple time features of the UAV signal into the current feature vector; Among them, several frequency characteristics include the center frequency and effective bandwidth of the UAV signal, and several time characteristics include the baud rate, frame period and single frame signal duration of the UAV signal.

7. The UAV identification method according to claim 1, characterized in that, The basic signal parameters include frequency band, modulation scheme, and signal bandwidth; The step of determining the non-black-flying drone signal most similar to the drone signal from the non-black-flying drone signals stored in a preset database based on the signal's basic parameters includes: The drone signals and the non-black drone signals stored in the preset database are classified according to the classification criteria of the frequency band, the modulation method, and the signal bandwidth, respectively. If the drone signal and the target non-black-flying drone signal are of the same type under at least two classification criteria in the frequency band, the modulation method, and the signal bandwidth, then the target non-black-flying drone signal is the non-black-flying drone signal most similar to the drone signal.

8. A drone identification device, characterized in that, The drone identification device includes: The acquisition unit is used to acquire mixed drone signals from multiple drones, separate the mixed drone signals, and obtain multiple drone signals. For each separated UAV signal, the following unit is executed: An extraction unit is used to extract the current feature vector and basic signal parameters of the UAV signal; The classification unit is used to determine the non-black-flying drone signal that is most similar to the drone signal from the non-black-flying drone signals stored in the preset database according to the basic signal parameters, and to obtain the preset feature vector of the determined non-black-flying drone signal. A calculation unit is used to calculate the feature deviation between each feature included in the current feature vector and each corresponding feature included in the preset feature vector; The determination unit is used to perform a weighted summation of the feature deviations of each feature and the preset weights of each feature to obtain the determination score of the drone signal. If the determination score of the drone signal is greater than the preset score threshold, the drone to which the drone signal belongs is determined to be an unauthorized drone.

9. An electronic device, characterized in that, It includes at least one processor and a memory connected to the processor, wherein: The memory is used to store computer programs; The processor is used to execute the computer program to enable the electronic device to implement the drone identification method as described in any one of claims 1 to 7.

10. A computer program product, characterized in that, It includes computer-readable instructions that, when executed on an electronic device, cause the electronic device to implement the drone identification method as described in any one of claims 1 to 7.

Images