Unmanned aerial vehicle interference method and equipment based on identification system

Abstract

The invention discloses an unmanned aerial vehicle interference method and equipment based on an identification system. The method comprises the following steps: collecting multi-modal data of a monitoring area; inputting the multi-modal data of the monitoring area into a pre-trained unmanned aerial vehicle identification model, and judging whether a black flight unmanned aerial vehicle exists in the monitoring area; obtaining the spatial position of the black flight unmanned aerial vehicle on the time sequence according to the unmanned aerial vehicle identification model; predicting a flight path of the black flight unmanned aerial vehicle; according to the flight path of the black-flight unmanned aerial vehicle, a camouflage unmanned aerial vehicle follows and intercepts a communication signal of the black-flight unmanned aerial vehicle, analyzes the communication signal of the black-flight unmanned aerial vehicle to obtain a communication frequency band and a communication protocol, generates a camouflage signal based on the communication frequency band and the communication protocol of the black-flight unmanned aerial vehicle, and performs feedback by using the camouflage unmanned aerial vehicle; and interfering a communication signal of the black flight unmanned aerial vehicle until the black flight unmanned aerial vehicle flies away from the monitoring area. The method can effectively interfere with the enemy unmanned aerial vehicle and feed back the camouflage information to the enemy, thereby effectively protecting the important facilities of our part from being found.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicles, and more specifically, to a method and device for jamming unmanned aerial vehicles based on an identification system. Background Art

[0002] In recent years, drones have been used more and more widely in professional photography, filming, agricultural applications and disaster rescue, which has brought great convenience to people's work and life, but also caused a lot of personal privacy and public safety issues, such as drones being used for criminal activities such as invasion and reconnaissance. Therefore, a series of accidents caused by drones warn people that they need to quickly and effectively combat the "illegal flying" of drones.

[0003] At present, the countermeasures against drones mainly include four categories: interference, physical interception and software defense. Interference means mainly include: radio interference, navigation signal interference, and sound wave interference; physical interception means mainly include: capture nets, laser weapons and microwave weapons; software defense means include: setting up no-fly zones.

[0004] However, after using the above methods, the enemy will notice that there is something abnormal with the drone, and thus infer that there may be important facilities at the location where the abnormality occurs. Summary of the invention

[0005] The purpose of the present invention is to provide a UAV jamming method and device based on an identification system, which can effectively interfere with enemy UAVs and feedback camouflage information to the enemy, thereby effectively protecting our important facilities from being discovered.

[0006] In order to achieve these purposes and other advantages according to the present invention, a method for jamming a drone based on an identification system is provided, comprising: Collect multimodal data of the monitoring area; The multimodal data of the monitored area is input into the pre-trained drone recognition model to determine whether there are illegal drones in the monitored area; When there are illegal drones in the monitoring area, the spatial position of the illegal drones in the time series is obtained according to the drone recognition model; Predict the flight path of illegal drones based on their spatial positions in time series; According to the flight path of the illegal drone, a disguised drone is used to follow and intercept the communication signal of the illegal drone, the communication signal of the illegal drone is analyzed to obtain the communication frequency band and communication protocol, and a disguised signal is generated based on the communication frequency band and communication protocol of the illegal drone, and the disguised drone is used for feedback; Use camouflaged drones to interfere with the communication signals of illegal drones until the illegal drones fly away from the monitored area.

[0007] Preferably, the multimodal data includes: binocular camera images and millimeter wave radar data.

[0008] Preferably, the drone recognition model includes: a binocular camera network, a millimeter wave radar network and a fusion network; The camera network includes ResNet-101 and feature pyramid network to learn multi-scale features, and 6 Transformer layers to generate visual update queries, with each layer taking the output query of the previous layer as input; The millimeter wave radar network adopts a multi-layer perception network to learn millimeter wave radar features; The fusion network uses three cascaded Transformer cross-attention decoders to learn the association between visual update queries and millimeter-wave radar features.

[0009] Preferably, the method for analyzing the communication signal of the illegal drone to obtain the communication frequency band includes: sequentially generating interference signals of different frequencies and transmitting them to the illegal drone, obtaining flight status images of the illegal drone under the interference signals of different frequencies, and judging whether the interference signal hits the communication frequency band of the illegal drone according to whether the flight status of the illegal drone is normal or not.

[0010] Preferably, the method of parsing the communication signal of the illegal drone to derive the communication protocol includes: inputting the communication signal into a logic analyzer, analyzing the timing diagram of the SPI protocol, finding the MOSI, MISO, CS and CLK signals, and then using the SPI protocol to analyze the instruction format, address width data and send address data.

[0011] Preferably, the method for interfering with the communication signal of the illegal drone includes: generating a high-power interference signal whose frequency matches the communication frequency band of the illegal drone and transmitting it to the illegal drone to suppress the communication signal of the illegal drone.

[0012] The present invention also provides a UAV jamming device based on an identification system, comprising: A data acquisition module, which is used to collect multimodal data of the monitoring area; The recognition module is used to input the multimodal data of the monitoring area into a pre-trained drone recognition model to determine whether there are illegal drones in the monitoring area; A positioning module, which is used to obtain the spatial position of the illegal drone in the time series according to the drone recognition model when there is an illegal drone in the monitoring area; A prediction module, which is used to predict the flight path of the illegal drone based on the spatial position of the illegal drone in the time series; A camouflage module is used to follow and intercept the communication signal of the illegal drone using a camouflage drone according to the flight path of the illegal drone, parse the communication signal of the illegal drone to obtain the communication frequency band and communication protocol, generate a camouflage signal based on the communication frequency band and communication protocol of the illegal drone, and use the camouflage drone for feedback; The jamming module is used to use a camouflaged drone to interfere with the communication signal of the illegal drone until the illegal drone flies away from the monitoring area.

[0013] The present invention also provides an electronic device, comprising: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the above-mentioned method.

[0014] The present invention also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the above method is implemented.

[0015] The present invention at least includes the following beneficial effects: the present invention can effectively improve the recognition accuracy of drones and reduce the misjudgment of non-drones by collecting multimodal data, fusing and complementing multiple data, and identifying drones. When it is identified that there are illegal drones in the monitoring area, the flight path of the illegal drones is quickly predicted, and a camouflaged drone is sent to track at close range, so that the communication signal of the illegal drone can be intercepted at close range, the attenuation of the communication signal is reduced, and the signal strength is guaranteed to be sufficient to meet the analysis requirements. After the communication frequency band and communication protocol of the illegal drone are analyzed, a camouflaged signal is immediately generated and the camouflaged drone is used for feedback, so that the controller of the illegal drone is difficult to find that the drone is interfered with. At the same time, the pre-set remote sensing information can be input into the camouflaged signal, and the pre-set remote sensing information can be information that does not contain any sensitive features, so that the controller of the illegal drone cannot find the real situation of its detection area. In addition, the use of camouflaged drones to interfere with the communication signals of illegal drones is to interfere in the small range where the illegal drones are located at close range, and to ensure the interference effect, and also to avoid the interference signal from affecting our communication signals. Finally, when the illegal drone completes its mission and needs to return, it is still allowed to leave, thus preventing the controller of the illegal drone from discovering any abnormalities.

[0016] Other advantages, objectives and features of the present invention will be embodied in part through the following description, and in part will be understood by those skilled in the art through study and practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 The figure is a flow chart of the UAV jamming method based on the identification system described in the present invention. DETAILED DESCRIPTION

[0018] The present invention is further described in detail below in conjunction with the accompanying drawings so that those skilled in the art can implement the invention with reference to the description.

[0019] It should be noted that the experimental methods described in the following embodiments are conventional methods unless otherwise specified, and the reagents and materials are commercially available unless otherwise specified; in the description of the present invention, the terms "lateral", "longitudinal", "upper", "lower", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention.

[0020] like Figure 1 As shown, the present invention provides a UAV interference method based on an identification system, comprising: Collect multimodal data of the monitoring area; The multimodal data of the monitored area is input into the pre-trained drone recognition model to determine whether there are illegal drones in the monitored area; When there are illegal drones in the monitoring area, the spatial position of the illegal drones in the time series is obtained according to the drone recognition model; Predict the flight path of illegal drones based on their spatial positions in time series; According to the flight path of the illegal drone, a disguised drone is used to follow and intercept the communication signal of the illegal drone, the communication signal of the illegal drone is analyzed to obtain the communication frequency band and communication protocol, and a disguised signal is generated based on the communication frequency band and communication protocol of the illegal drone, and the disguised drone is used for feedback; Use camouflaged drones to interfere with the communication signals of illegal drones until the illegal drones fly away from the monitored area.

[0021] In the above embodiment, multimodal data is collected, multiple data are fused and complemented, and the drone is identified, which can effectively improve the recognition accuracy of the drone and reduce the misjudgment of non-drones. When it is identified that there is an illegal drone in the monitoring area, the flight path of the illegal drone is quickly predicted, and a disguised drone is sent to track it at close range, so that the communication signal of the illegal drone can be intercepted at close range, the attenuation of the communication signal is reduced, and the signal strength is guaranteed to be sufficient to meet the analysis requirements. After the communication frequency band and communication protocol of the illegal drone are analyzed, a disguised signal is immediately generated and the disguised drone is used for feedback, so that the controller of the illegal drone is difficult to find that its drone is interfered with. At the same time, the pre-set remote sensing information can be input into the disguised signal, and the pre-set remote sensing information can be information that does not contain any sensitive features, so that the controller of the illegal drone cannot find the real situation of its detection area. In addition, the use of a disguised drone to interfere with the communication signal of the illegal drone is to interfere in the small range where the illegal drone is located at close range, while ensuring the interference effect, and also avoiding the interference signal from affecting our communication signal. Finally, when the illegal drone completes its mission and needs to return, it is still allowed to leave, thus preventing the controller of the illegal drone from discovering any abnormalities.

[0022] Furthermore, the device for collecting multimodal data of the monitoring area may be a binocular camera image and a millimeter-wave radar data acquired by a binocular fisheye camera and a millimeter-wave radar respectively.

[0023] For example, the binocular camera image can be represented as Where R is the stereo camera image set, N is the number of time frames, V is the number of views, H is the height, W is the width, and 3 is the number of channels; Each image has an extrinsic and intrinsic matrix that together define the mapping from reference coordinates to local pixel coordinates of the camera image.

[0024] The detection result of the millimeter-wave radar is represented by a three-dimensional point in a three-dimensional coordinate system centered on the drone. This three-dimensional point in the millimeter-wave radar point cloud is parameterized as P = (x, y, z, v x ,v y ,v z ,t) Where (x, y, z) is the position of the drone, (v x ,v y ,v z ) are the speeds of the drone in the x, y, and z directions respectively. Due to the high sparsity of the point cloud, 3 to 5 time scans are usually aggregated, so a time dimension t is added here.

[0025] Furthermore, the drone identification model includes: a binocular camera network, a millimeter wave radar network and a fusion network; The camera network includes ResNet-101 and feature pyramid network to learn multi-scale features, and 6 Transformer layers to generate visual update queries, with each layer taking the output query of the previous layer as input; Specifically, for the first Transformer layer, the learnable 3D object queries are randomly initialized within the surveillance area. The camera network learns the distribution of these 3D query locations from the training data. For the following layers, the input query is the output query from the previous layer. Each 3D object query encodes the 3D position center of a potential object. Through bilinear interpolation, these 3D center points are projected into the image feature pyramid based on the camera internal and external parameters to extract image features. Assuming that there are k layers in the image feature pyramid, for a 3D point, the sampled image feature is the sum of the sampled features of all k layers. The final Transformer layer is used to learn the interactions between all 3D object queries and generate attention scores, and then the object query is combined with the adopted image features weighted by the attention score to form an updated object query.

[0026] The millimeter wave radar network adopts a multi-layer perception network to learn millimeter wave radar features; Specifically, the radar data of the previous 5 frames are accumulated and converted into the current frame. Since each radar point has multiple data channels, including 3D position x, y, z, radial velocity v x and v y , motion compensation speed v xc and v yc , dynamic attribute channel and other state channels. In order to make the state channel learnable to the network, the state channel is converted into a one-hot vector. In addition, since the current frame uses 5 accumulated frames, the time offset of each frame relative to the current time is useful for indicating the position offset, so a time offset channel is added for each radar point. The radar features and radar point position encoding are learned using a multi-layer perception network, and the learned features and position encoding are combined to form the final millimeter-wave radar features.

[0027] The fusion network uses three cascaded Transformer cross-attention decoders to learn the association between visual update queries and millimeter-wave radar features.

[0028] Specifically, since each visual query object is only associated with radar features near its location, it is not necessary for each visual query to interact with all radar features. Therefore, a shielding mechanism is set here, and the visual query object and the radar feature are allowed to interact only when the Euclidean distance between the visual query object and the radar feature is less than a threshold. Different decoders can set different threshold radiuses.

[0029] For the first Transformer cross-attention decoder, the object query output by the camera network and the millimeter-wave radar features output by the millimeter-wave radar network are used as input, and the cross-attention layer in the decoder outputs an attention score matrix. The M elements in the i-th row of the matrix represent the attention scores between the i-th object query and all the M radar features, and the sum is 1. Then, the attention-weighted radar features are associated with the object query, and these attention-weighted radar features are combined with the object query and enhanced through the feedforward network to form a fused query. The subsequent Transformer cross-attention decoder takes the fused query output by the previous decoder and the attention-weighted radar features calculated based on the threshold radius of this decoder as input.

[0030] Finally, the Hungarian loss function is used to guide the training process of the drone recognition model. There are two parts in the loss function, one for classification and the other for bounding box regression. Focal loss is used for classification to solve the class imbalance, and L1 loss is used for bounding box regression. Assume that N and K represent the number of predictions and ground truths in the same frame. When N is greater than K, φ (no object) is used to fill the ground truth set. The Hungarian algorithm is used to solve the bipartite matching problem between predictions and ground truths: in, is the optimal arrangement, Θ is the arrangement set, σ(i) ( c i ) is the permutation index σ ( i )of c i Class probability, L box is the L1 difference of the bounding box, b i and σ(i) They are the basic fact box and the prediction box respectively; The final Hungarian loss is expressed as: in, and is the parameter of the focal loss.

[0031] The drone recognition model is trained using the pre-prepared training set, and then tested using the test set. Compared with single binocular camera image recognition and millimeter wave radar recognition, the recognition accuracy of this embodiment is greatly improved, and the results are shown in the following table: mAP mATE mASE 1mOE mAVE mAAE Image Recognition 0.674 0.526 0.156 0.347 0.456 0.356 Radar recognition 0.581 0.482 0.139 0.482 0.389 0.218 This embodiment 0.893 0.217 0.067 0.214 0.142 0.145 Note: mAP is the average accuracy, mATE is the average translation error, mASE is the average scale error, mAOE is the average direction error, mAVE is the average velocity error, and mAAE is the average attribute error.

[0032] Furthermore, an autoregressive algorithm can be used to predict the flight path of illegal drones based on their spatial positions in the time series, as follows: In the formula, X t is the predicted position at the next moment, c is a constant term, p is the order of the model, indicating how many past positions have been used. is a parameter of the model, indicating the i The weight of the past moment, It is i The position of a past moment, is the error term, which represents the random fluctuations of the model.

[0033] Specifically, p Determined by information criteria or cross-validation methods, Determined using the least squares method or maximum likelihood method.

[0034] Furthermore, the method for analyzing the communication signal of the illegal drone to obtain the communication frequency band includes: generating interference signals of different frequencies in sequence and transmitting them to the illegal drone, obtaining flight status images of the illegal drone under the interference signals of different frequencies, and judging whether the interference signal hits the communication frequency band of the illegal drone according to whether the flight status of the illegal drone is normal or not.

[0035] Specifically, a signal generator for generating interference signals of different frequencies, a high-definition camera or radar system for capturing and recording the flight status of illegal drones, and a signal access device for obtaining the communication signal of illegal drones can be set on the camouflaged drone. After the signal access device obtains the communication signal of the illegal drone, it sends it to the signal receiving device set on the ground. The signal receiving device then transmits the communication signal of the illegal drone to the spectrum analyzer connected to it for communication. The spectrum analyzer scans within a preset frequency range (such as the common drone communication frequency bands of 2.4GHz, 5.8GHz, etc.), and records the frequency band of suspected drone communication to narrow the frequency range of subsequent interference tests and improve efficiency. According to the preliminary scanning results, frequency points are selected from low to high as the starting point of the interference signal, and reasonable intervals are set between each frequency point to ensure that all possible communication frequency bands are covered. The signal generator is used to generate interference signals according to the selected frequency. The signal strength should be set within the threshold that will not cause unnecessary interference to the surrounding environment. The generated interference signal is transmitted to the illegal drone through the antenna set on the camouflaged drone, and the frequency, time and signal strength of each transmission are recorded. While transmitting the interference signal, use drone monitoring equipment to continuously record the flight status of the illegal drone, including but not limited to flight altitude, speed, direction changes, and whether there are abnormal behaviors such as loss of control and falling. Compare the flight status images of the illegal drone under different frequency interferences, and identify which frequencies of interference have caused significant changes in the drone's flight status (such as loss of control, signal loss, etc.). These frequencies are candidate frequencies that hit the communication frequency band of the illegal drone. Further refine the test and narrow the frequency range until the most accurate communication frequency band is determined.

[0036] In the above embodiment, the communication frequency band of the illegal drone can be found quickly and effectively through the interference test of interference signals of different frequency bands in sequence.

[0037] Furthermore, the method of parsing the communication signal of the illegal drone to derive the communication protocol includes: inputting the communication signal into a logic analyzer, analyzing the timing diagram of the SPI protocol, finding the MOSI, MISO, CS and CLK signals, and then using the SPI protocol to analyze the instruction format, address width data and send address data.

[0038] Specifically, a signal access device, such as an antenna, a signal receiver or a direct access point of a drone communication module, can be set on the camouflaged drone to obtain the communication signal of the illegal drone, and then send the acquired communication signal of the illegal drone to a signal receiving device set on the ground. The signal receiving device then transmits the communication signal of the illegal drone to a logic analyzer connected to it for communication, and the logic analyzer is connected to a host computer running data analysis software.

[0039] Start the capture function of the logic analyzer and set the appropriate sampling rate and capture time length to cover the complete cycle of the drone communication. Monitor and wait for the unmanned drone to communicate to ensure that the logic analyzer can capture valid SPI communication signals. In the logic analyzer software, view the captured communication signal timing diagram. Identify and mark the MOSI, MISO, CS, and CLK signals to confirm that their timing relationship meets the SPI protocol standard. Observe the changes in the CS signal to determine the start and end time of each communication. According to the SPI protocol standard, analyze the data bits on the MOSI and MISO signals to identify instructions, addresses, and data. Pay attention to the format of the instructions, including instruction length, address width, and data length. Use the logic analyzer software tools, such as data decoders or protocol parsers, to help automatically identify and parse SPI protocol data. Based on the parsed instruction format, determine the address width (usually 8 bits, 16 bits, or 32 bits). Find the corresponding address data on the MOSI signal and record it. Analyze the meaning of the address data, which may include the address of the drone's internal registers, configuration parameters, or command codes. Compare the parsed SPI protocol instructions and address data with the actual drone's communication behavior to verify the correctness of the parsing results. If possible, try to use the parsed protocol instructions to perform simple control or query operations on the drone to further verify the validity of the protocol.

[0040] Furthermore, the method for interfering with the communication signal of the illegal drone includes: generating a high-power interference signal whose frequency matches the communication frequency band of the illegal drone and transmitting it to the illegal drone to suppress the communication signal of the illegal drone.

[0041] Specifically, select a signal generator that can generate high-power, stable frequency output, and its frequency range should cover the communication frequency band that illegal drones may use. According to the interference range, select a suitable antenna to transmit the interference signal. Directional antennas are suitable for accurately interfering with drones in a specific direction, while omnidirectional antennas are suitable for wide coverage areas. If the output power of the signal generator is not enough to suppress the communication signal of the drone, an additional power amplifier may be required to enhance the strength of the interference signal. Before and during the interference, the frequency and power of the interference signal can be monitored and verified with a spectrum analyzer. First set the parameters of the interference signal such as frequency, power, and waveform on the signal generator. Make sure that the frequency of the interference signal completely overlaps or is very close to the communication frequency band of the drone to achieve the best interference effect. Initially, set the power of the interference signal to a low level, and then gradually increase it until the drone communication is observed to be significantly suppressed. Be careful not to set the power too high to avoid unnecessary interference to the surrounding environment or other legitimate communications. Continuously monitor the communication status of the drone to ensure that the interference signal effectively suppresses the communication signal of the drone. During the interference process, use a spectrum analyzer to continuously monitor the frequency and power of the interference signal to ensure that they remain within the predetermined range.

[0042] Furthermore, based on the previous technical solution, it also includes: In the process of the disguised drone following the illegal drone, the radio signal strength of the illegal drone is obtained in real time, and when the variance of the radio signal strength within a predetermined time is less than a predetermined threshold, interference is initiated, otherwise, the following is continued; Specifically, when the disguised drone closely follows the illegal drone, the radio signal receiving device installed on the disguised drone continuously and accurately obtains the radio signal strength value from the illegal drone; in order to reasonably decide when to start the interference action against the illegal drone, a predetermined threshold value of the predetermined time interval will be set in advance based on a large amount of practical experience. In each complete predetermined time interval, a series of values ​​of the radio signal strength of the illegal drone are collected and sorted, and then the variance of these values ​​is calculated. When the variance of the radio signal strength of the illegal drone obtained within this predetermined time is less than the predetermined threshold, it means that during this period, the radio signal strength emitted by the illegal drone presents a relatively stable state with very small fluctuations. This may mean that the illegal drone is in a relatively stable flight stage, maintaining a relatively fixed distance from the disguised drone, and the electromagnetic environment in which it is located has not changed significantly. At this time, all conditions for starting the interference action are met. Therefore, the disguised drone will quickly carry out targeted interference on the communication signals sent by the illegal drone. If the variance of the radio signal strength of the illegal drone within the predetermined time is calculated to be greater than or equal to the predetermined threshold, it means that the fluctuation of its radio signal strength is more obvious and frequent during this period of time. This may be because the illegal drone is performing complex flight maneuvers, resulting in constant changes in the relative position relationship between it and the disguised drone, such as the distance and angle, or there may be other electromagnetic interference sources in the surrounding area, which have an impact on the radio signal of the illegal drone. At this time, if the interference action is initiated, it is likely that the expected interference effect will not be achieved, and it may even cause unpredictable problems in subsequent operations due to factors such as unstable signals.

[0043] Based on the same inventive concept, the present invention also provides a drone jamming device based on an identification system, which can be a personal computer, a server, or other devices that implement the aforementioned drone jamming method based on the identification system.

[0044] The UAV jamming equipment based on the identification system includes: A data acquisition module, which is used to collect multimodal data of the monitoring area; The recognition module is used to input the multimodal data of the monitoring area into a pre-trained drone recognition model to determine whether there are illegal drones in the monitoring area; A positioning module, which is used to obtain the spatial position of the illegal drone in the time series according to the drone recognition model when there is an illegal drone in the monitoring area; A prediction module, which is used to predict the flight path of the illegal drone based on the spatial position of the illegal drone in the time series; A camouflage module is used to follow and intercept the communication signal of the illegal drone using a camouflage drone according to the flight path of the illegal drone, parse the communication signal of the illegal drone to obtain the communication frequency band and communication protocol, generate a camouflage signal based on the communication frequency band and communication protocol of the illegal drone, and use the camouflage drone for feedback; The jamming module is used to use a camouflaged drone to interfere with the communication signal of the illegal drone until the illegal drone flies away from the monitoring area.

[0045] All relevant contents of each step involved in the embodiment of the aforementioned identification system-based drone jamming method can be referred to the functional description of the functional modules corresponding to the identification system-based drone jamming device in the embodiment of the present application, and will not be repeated here.

[0046] The division of modules in the embodiments of the present application is schematic and is only a logical function division. There may be other division methods in actual implementation. In addition, each functional module in each embodiment of the present invention may be integrated into a processor, or may exist physically separately, or two or more modules may be integrated into one module. The above-mentioned integrated modules may be implemented in the form of hardware or in the form of software functional modules.

[0047] In the drawings of the system embodiment provided by the present invention, the connection relationship between modules indicates that there is a communication connection between them, which can be specifically implemented as one or more communication buses or signal lines.

[0048] The present invention also provides an electronic device, comprising: at least one processor, and a memory connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor executes the above-mentioned UAV jamming method based on the identification system. The electronic device can be any terminal device including a mobile phone, a laptop computer, a desktop computer, a tablet computer, a PDA (Personal Digital Assistant), a POS (Point of Sales), a vehicle-mounted computer, etc.

[0049] The present invention also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the above-mentioned drone interference method based on the identification system is implemented.

[0050] Through the description of the above implementation mode, the technicians in the relevant field can clearly understand that the present invention can be implemented by means of software plus necessary general hardware, and of course, it can also be implemented by special hardware including special integrated circuits, special CPUs, special memories, special components, etc. In general, all functions completed by computer programs can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be various, such as analog circuits, digital circuits or special circuits. However, for the present invention, software program implementation is a better implementation mode in more cases. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a readable storage medium, such as a computer floppy disk, a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, RandomAccess Memory), a disk or an optical disk, etc., including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0051] Although the embodiments of the present invention have been disclosed as above, they are not limited to the applications listed in the specification and the implementation modes, and they can be fully applied to various fields suitable for the present invention. For those familiar with the art, additional modifications can be easily implemented. Therefore, without departing from the general concept defined by the claims and the scope of equivalents, the present invention is not limited to the specific details and the illustrations shown and described herein.

Claims

1. A UAV jamming method based on an identification system, characterized in that: include: Collect multimodal data of the monitoring area; The multimodal data of the monitored area is input into the pre-trained drone recognition model to determine whether there are illegal drones in the monitored area; When there are illegal drones in the monitoring area, the spatial position of the illegal drones in the time series is obtained according to the drone recognition model; Predict the flight path of illegal drones based on their spatial positions in time series; According to the flight path of the illegal drone, a disguised drone is used to follow and intercept the communication signal of the illegal drone, the communication signal of the illegal drone is analyzed to obtain the communication frequency band and communication protocol, and a disguised signal is generated based on the communication frequency band and communication protocol of the illegal drone, and the disguised drone is used for feedback; Use camouflaged drones to interfere with the communication signals of illegal drones until the illegal drones fly away from the monitored area.

2. The UAV jamming method based on the identification system as claimed in claim 1, characterized in that: The multimodal data includes: binocular camera images and millimeter wave radar data.

3. The UAV jamming method based on the identification system as claimed in claim 2, characterized in that: The drone recognition model includes: a binocular camera network, a millimeter wave radar network and a fusion network; The camera network includes ResNet-101 and feature pyramid network to learn multi-scale features, and 6 Transformer layers to generate visual update queries, with each layer taking the output query of the previous layer as input; The millimeter wave radar network adopts a multi-layer perception network to learn millimeter wave radar features; The fusion network uses three cascaded Transformer cross-attention decoders to learn the association between visual update queries and millimeter-wave radar features.

4. The UAV jamming method based on the identification system as claimed in claim 1, characterized in that: The method for analyzing the communication signal of the illegal UAV to obtain the communication frequency band includes: sequentially generating interference signals of different frequencies and transmitting them to the illegal UAV, obtaining flight status images of the illegal UAV under the interference signals of different frequencies, and judging whether the interference signal hits the communication frequency band of the illegal UAV according to whether the flight status of the illegal UAV is normal or not.

5. The UAV jamming method based on the identification system as claimed in claim 1, characterized in that: The method of parsing the communication signal of the illegal drone to derive the communication protocol includes: inputting the communication signal into a logic analyzer, analyzing the timing diagram of the SPI protocol, finding the MOSI, MISO, CS and CLK signals, and then using the SPI protocol to analyze the instruction format, address width data and send address data.

6. The UAV jamming method based on the identification system as claimed in claim 1, characterized in that: The method for interfering with the communication signal of the illegal drone includes: generating a high-power interference signal with a frequency matching the communication frequency band of the illegal drone and transmitting it to the illegal drone to suppress the communication signal of the illegal drone.

7. A drone jamming device based on an identification system, characterized in that: include: A data acquisition module, which is used to collect multimodal data of the monitoring area; The recognition module is used to input the multimodal data of the monitoring area into a pre-trained drone recognition model to determine whether there are illegal drones in the monitoring area; A positioning module, which is used to obtain the spatial position of the illegal drone in the time series according to the drone recognition model when there is an illegal drone in the monitoring area; A prediction module, which is used to predict the flight path of the illegal drone based on the spatial position of the illegal drone in the time series; A camouflage module is used to follow and intercept the communication signal of the illegal drone using a camouflage drone according to the flight path of the illegal drone, parse the communication signal of the illegal drone to obtain the communication frequency band and communication protocol, generate a camouflage signal based on the communication frequency band and communication protocol of the illegal drone, and use the camouflage drone for feedback; The jamming module is used to use a camouflaged drone to interfere with the communication signal of the illegal drone until the illegal drone flies away from the monitoring area.

8. An electronic device, characterized in that: include: At least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor performs the method according to any one of claims 1 to 6.

9. A storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method described in any one of claims 1 to 6 is implemented.

Images