A method for identifying unlicensed drones

Abstract

The present application relates to the field of unmanned aerial vehicle (UAV) control technology, and particularly to a method for identifying unlicensed UAVs, including: a line-of-sight range detection step of detecting UAVs within the line-of-sight range through a vision sensor, positioning the positions of the UAVs based on a preset target detection algorithm, and establishing a UAV position list; an unlicensed UAV identification step of continuously receiving ADS-B signals and Remote ID signals through a pre-installed signal receiving device, and based on the positions of the UAVs and the ADS-B signals and Remote ID signals, sequentially verifying whether the corresponding UAVs within the line-of-sight range send ADS-B signals and Remote ID signals, whether the signals are registered signals, and whether the registered signals are genuine signals. If any verification result is negative, it is determined as an unlicensed UAV. Through the comprehensive analysis and judgment of the UAV positions and signals in the present application, the reliability and robustness of identifying forged signals are improved.

Description

Technical Field

[0001] This application relates to the technical field of UAV control, and particularly to a method for identifying unlicensed UAVs. Background Art

[0002] UAV identity recognition technology is mainly an anti-measure against unlicensed flights. Unlicensed flight refers to flights by those without a private pilot's license or aircraft without legal identity, that is, unregistered flights, which are somewhat dangerous. Due to its characteristic of being outside the regulatory framework, it poses a serious threat to low-altitude traffic and public safety.

[0003] Its harm is reflected in: UAVs that do not send ADS-B signals and Remote ID signals cannot be recognized by other aircraft or air traffic control systems, resulting in a direct threat to the safe operation of aviation due to the risk of airspace collisions; there are also cases where forged Remote ID signals are sent through equipment to disguise as legitimate UAVs, misleading the regulatory system and enabling attackers to use this loophole to cover up malicious activities; in addition, UAVs deviating from the declared flight path may also pose a threat to public safety due to out-of-control or operational errors, such as falling in crowded areas.

[0004] Existing identification methods use ADS-B or Remote ID to identify the identity of UAVs, but this method relies on UAVs to actively broadcast identity information, cannot identify UAVs that deliberately turn off signals, and is vulnerable to attacks by false forged signals. There are also some identification methods that use devices such as radars and infrared sensors arranged on the ground to visually capture UAVs, but have low resolution for small-sized and low-altitude UAVs, are easily blocked by terrain and interfered by clutter, and are restricted by weather and light conditions and cannot accurately capture in real time.

[0005] In addition, the data of different monitoring and identification methods are not effectively integrated at present, resulting in fragmented information, and the deployment cost of high-precision detection equipment is too high, making it difficult to promote on a large scale. Summary of the Invention

[0006] An embodiment of this application provides a method for identifying unlicensed UAVs to at least solve the problem in related technologies of how to identify regular UAVs within the line of sight and various unlicensed flight behaviors.

[0007] An embodiment of this application provides a method for identifying unlicensed UAVs, including:

[0008] A line-of-sight range detection step, detecting UAVs within the line of sight through a vision sensor and positioning the positions of the UAVs based on a preset target detection algorithm to establish a UAV position list, where the vision sensor is mounted on a patrol UAV or other patrol equipment performing unlicensed flight patrol tasks in the target area.

[0009] Steps for identifying unlicensed drones: Continuously receive ADS-B signals and Remote ID signals through pre-installed signal receiving devices. Based on the position of the drone and the ADS-B signals and Remote ID signals, verify in sequence whether the corresponding drone within the line of sight sends ADS-B signals and Remote ID signals, whether the signals are registered signals, and whether the registered signals are genuine signals. If any verification result is negative, it is determined as an unlicensed drone. Among them, the ADS-B signals include first position information, altitude information, speed information, heading information, and ICAO address, and the Remote ID signals include the serial number of the drone, second position information, operator position, and status information.

[0010] In some embodiments, the steps for identifying unlicensed drones include:

[0011] Step for judging the presence or absence of signals: Analyze the ADS-B signals and Remote ID signals, obtain the position information in the ADS-B signals and / or Remote ID signals, compare it with the drone position list, identify the drones that do not send signals as unlicensed drones, and the drones that send signals enter the registration status judgment step for registration verification;

[0012] Registration status judgment step: Analyze the ICAO address in the ADS-B signals and / or the serial number in the Remote ID signals, and conduct real-time query verification in the drone supervision database. Identify the drones that send unregistered signals as unlicensed drones, and the drones that send registered signals enter the genuine signal judgment step for authenticity verification;

[0013] Genuine signal judgment step: Conduct frequency feature analysis, signal strength analysis, and spatio-temporal consistency analysis on the ADS-B signals and Remote ID signals respectively. By verifying the similarity between the signal and the normal signal, verifying the matching degree between the signal strength and the distance and the signal strength change, the drone position change, and the signal frequency change, and the consistency of the three with the movement law of the aircraft in the air, comprehensively judge whether the signal is a genuine signal or a forged signal based on the verification results of the ADS-B signals and Remote ID signals. Identify the drones with forged signals as unlicensed drones. If any one of them is judged as a forged signal, it is identified as an unlicensed drone.

[0014] In the embodiments of the present application, comprehensive judgment is carried out through spectrum analysis, signal strength analysis, and spatio-temporal consistency analysis on the ADS-B signals and Remote ID signals respectively, improving the reliability and robustness of identifying forged signals and avoiding missed judgments or misjudgments that may occur in single-signal analysis.

[0015] In some embodiments, the genuine signal judgment step further includes:

[0016] Spectrum analysis step: Receive the signal to be verified , perform a fast Fourier transform on the signal to be verified to obtain the spectrum of the signal to be verified , calculate the cosine similarity and Euclidean distance with the spectrum template of the normal signal after normalizing it , determine whether it is an abnormal spectrum by combining the cosine similarity and Euclidean distance. If so, it is a forged signal;

[0017] Signal strength analysis step: Obtain the position of the target drone in the drone position list, determine the distance between the target drone and the signal receiving device, calculate the corresponding received power based on the distance using the free space path loss model, and determine whether the received power matches the actual power of the signal to be verified. If not, it is a forged signal;

[0018] Spatio-temporal consistency analysis step: Continuously obtain the position of the target drone and the signal to be verified , analyze whether the change in signal strength during its movement conforms to the free space path loss model, whether the speed change is smooth, and whether the signal frequency offset of the signal to be verified conforms to the expectation. If any one of the analysis results is negative, it is a forged signal.

[0019] In some embodiments, the black flight drone identification step further includes:

[0020] Abnormal flight judgment step: For the drone identified as a real signal in the real signal judgment step, analyze the heading information in its ADS-B signal, and determine whether its position in the drone position list is compliant, that is, conforms to its preset route. If not, it is judged as a black flight drone.

[0021] In some embodiments, in the spectrum analysis step, the abnormal spectrum is the signal spectrum with a cosine similarity lower than the first threshold and an Euclidean distance exceeding the second threshold; where the cosine similarity is expressed as:

[0022] ;

[0023] The Euclidean distance is expressed as:

[0024] .

[0025] In some embodiments, in the signal strength analysis step, the received power is expressed as the following calculation model:

[0026] ,

[0027] where, is the transmit power, and They are the transmitting antenna gain and the receiving antenna gain respectively, is the distance, is the frequency of the normal signal, .

[0028] In some of the embodiments, the spatio-temporal consistency analysis step further includes:

[0029] A signal strength consistency analysis step, according to the current position of the target UAV, calculates the expected received power at the current position based on the free space path loss model, determines the actual received power at the current position according to the signal to be verified, calculates the mean square error between the actual received power and the expected received power, and if it is greater than a set error value, determines that the signal is abnormal;

[0030] A position consistency analysis step, calculates the moving speed and moving acceleration of the target UAV according to its position, and if the moving acceleration is greater than an acceleration threshold, determines that the signal is abnormal;

[0031] A frequency consistency analysis step, measures the signal frequency of the signal to be verified and calculates the absolute difference between the actual frequency offset and the expected frequency offset based on its position, and if the absolute difference is greater than the offset threshold, determines that the signal is abnormal.

[0032] In some of the embodiments, in the signal strength consistency analysis step, calculates the current distance between the target UAV and the signal receiving device at the current position, substitutes the current distance into the calculation model of the received power, and obtains the expected received power , combines with the actual received power of the signal receiving device Calculates the mean square error, and the mean square error MSE is expressed as:

[0033] .

[0034] In some of the embodiments, configures the first acceleration threshold as the maximum acceleration of the UAV, and if the absolute value of the moving acceleration is greater than the acceleration threshold, determines that the position change of the target UAV is not smooth and there is a signal abnormality.

[0035] In some of the embodiments, in the frequency consistency analysis step, the signal frequency is expressed as , and the expected frequency offset is calculated based on the following calculation model:

[0036] , where is the moving speed, is the included angle between the moving direction vector of the target UAV and the line of sight direction of the vision sensor, and the line of sight direction is the direction from the vision sensor to the target UAV;

[0037] The actual frequency offset is calculated based on the following calculation model:

[0038] .

[0039] Based on the above steps, the embodiments of the present application determine whether a signal conforms to the physical motion law of an aerial vehicle based on signal strength, position, and frequency change, so as to identify the abnormality of a forged signal.

[0040] The details of one or more embodiments of the present application are set forth in the following drawings and description, so that other features, objects, and advantages of the present application will become more comprehensible. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The drawings described herein are used to provide a further understanding of the present application and form a part of the present application. The illustrative embodiments and descriptions thereof are used to explain the present application and do not constitute an improper limitation of the present application. In the drawings:

[0042] Figure 1 is a schematic diagram of the judgment logic of the method for identifying unlicensed drones according to an embodiment of the present application;

[0043] Figure 2 is a schematic flowchart of the method for identifying unlicensed drones according to an embodiment of the present application;

[0044] Figure 3 is a schematic flowchart of sub-step S2 of the method for identifying unlicensed drones according to an embodiment of the present application;

[0045] Figure 4 Schematic flowchart of sub-step S23 of the method for identifying unlicensed drones according to an embodiment of the present application;

[0046] Figure 5 is a schematic flowchart of sub-step S233 of the method for identifying unlicensed drones according to an embodiment of the present application;

[0047] Figure 6 is another schematic flowchart of the method for identifying unlicensed drones according to an embodiment of the present application;

[0048] Figure 7 is a schematic diagram of the relative positions of drone A and drone B in the method for identifying unlicensed drones according to an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0049] In order to make the objectives, technical solutions and advantages of the present application more clear and understandable, the present application will be described and explained below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application. Based on the embodiments provided in the present application, all other embodiments obtained by those of ordinary skill in the art without making creative efforts belong to the scope of protection of the present application.

[0050] Obviously, the accompanying drawings in the following description are only some examples or embodiments of the present application. For those of ordinary skill in the art, the present application can also be applied to other similar scenarios based on these drawings without making creative efforts. In addition, it can also be understood that although the efforts made in such a development process may be complex and lengthy, for those of ordinary skill in the art related to the content disclosed in the present application, some design, manufacturing or production changes based on the technical content disclosed in the present application are only conventional technical means and should not be understood as the content disclosed in the present application being insufficient.

[0051] Unless otherwise defined, the technical terms or scientific terms involved in the present application should be the general meanings understood by those with ordinary skills in the technical field to which the present application belongs. The words such as "a", "an", "one kind", "the" and the like involved in the present application do not represent a limitation in quantity and can represent a singular or plural number. The terms "including", "comprising", "having" and any variations thereof involved in the present application are intended to cover non-exclusive inclusion; for example, a process, method, system, product or device including a series of steps or modules (units) is not limited to the listed steps or units, but may also include unlisted steps or units, or may further include other steps or units inherent to these processes, methods, products or devices. The terms "connected", "coupled" and the like involved in the present application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The "multiple" involved in the present application means two or more. "And / or" describes the association relationship of associated objects and indicates that three relationships may exist. For example, "A and / or B" may represent: A exists alone, A and B exist simultaneously, and B exists alone. The character " / " generally represents an "or" relationship between the front and rear associated objects. The terms "first", "second", "third" and the like involved in the present application are only used to distinguish similar objects and do not represent a specific order for the objects.

[0052] The Automatic Dependent Surveillance - Broadcast (ADS - B) system is an aviation surveillance technology based on satellite navigation and digital communication, used for real - time broadcasting of information such as the position, altitude, and speed of aircraft, relying on frequencies of 1090 MHz or 978 MHz and using a specific protocol.

[0053] Remote ID is the identity recognition system of the drone. Similar to an avionics license plate, it broadcasts identity information, location information, and flight status via wireless signals or networks for ground devices or other drones to receive.

[0054] Although ADS-B and Remote ID signals have different functions, they are both important bases for drone authentication, and forgers may counterfeit them separately. In scenarios such as civil aviation control airspace, drone logistics, and urban low-altitude transportation, high-precision airspace perception is required, and drones will be required to be equipped with both ADS-B and Remote ID simultaneously. Based on this, this application designs a method for identifying unlicensed drones with step-by-step screening.

[0055] Figure 1 It is a logical schematic diagram of the method for identifying unlicensed drones according to an embodiment of this application. Figures 2 to 6 It is a flowchart of the method for identifying unlicensed drones according to an embodiment of this application. As Figure 2 shown, this process includes the following steps:

[0056] Line-of-sight range detection step S1: Detect drones within the line of sight through a visual sensor and locate the positions of the drones based on a preset target detection algorithm to establish a drone position list. Among them, the visual sensor is mounted on a patrol drone or other patrol equipment performing unlicensed patrol tasks in the target area. The visual sensor includes, but is not limited to, a binocular camera or a depth camera. Model training is performed based on the image dataset collected within the line of sight of the drone. The trained target detection algorithm is used to obtain the real-time video stream detected by the visual sensor and output the bounding box, category, and image coordinates of the drone. Convert the image coordinates of each drone to the camera coordinate system and then to the world coordinate system to obtain the position of the drone. Specifically,

[0057] When using a binocular camera or a depth camera for detection, utilize the internal parameter matrix of the camera and the center point of the detected drone bounding box to convert the points in the image coordinate system to the camera coordinate system through depth information or parallax. Based on the known position, attitude, and camera installation position of the current patrol drone, further calculate and locate the position of the target drone.

[0058] When using a monocular camera for detection, the wing width of the target drone can be obtained based on the category of the drone. The distance D between the target drone and the current patrol drone can be calculated through the pixel width in the image. The azimuth angle and pitch angle of the target drone in the camera coordinate system can be calculated through the target key coordinates in the image and the camera internal parameters, and then its three-dimensional coordinates can be calculated in combination with the distance D and finally converted to the world coordinate system.

[0059] The preset object detection algorithm uses the CenterNet object detection network. In this embodiment, its backbone network structure is changed to UNet, which is suitable for feature extraction and upsampling and helps to capture small objects. The center detection loss and the bounding box regression loss are introduced to handle the problem of unbalanced positive and negative samples in small object detection. At the same time, an auxiliary branch is added during training and discarded during inference to improve the training accuracy and inference speed of the model.

[0060] For the identification step S2 of unlicensed drones, the ADS-B signal and the Remote ID signal are continuously received through a pre-installed signal receiving device. Based on the position of the drone and the ADS-B signal and the Remote ID signal, it is successively verified whether the corresponding drone within the line of sight sends the ADS-B signal and the Remote ID signal, whether the signal is a registered signal, and whether the registered signal is a real signal. If any verification result is negative, it is determined as an unlicensed drone. Among them, the ADS-B signal includes the first position information, altitude information, speed information, heading information, and ICAO address, and the Remote ID signal includes the serial number of the drone, the second position information, the operator position, and the status information.

[0061] Reference Figure 3 As shown, the identification step S2 of unlicensed drones includes:

[0062] The signal presence / absence judgment step S21: Analyze the ADS-B signal and the Remote ID signal, obtain the position information in the ADS-B signal and / or the Remote ID signal, compare it with the drone position list, identify the drone that does not send a signal as an unlicensed drone, and the drone that sends a signal enters the registration status judgment step for registration verification;

[0063] The registration status judgment step S22: Analyze the ICAO address in the ADS-B signal and / or the serial number in the Remote ID signal, and perform real-time query verification in the drone supervision database. Identify the drone that sends an unregistered signal as an unlicensed drone, and the drone that sends a registered signal enters the real signal judgment step for authenticity verification;

[0064] The real signal judgment step S23: Perform frequency feature analysis, signal strength analysis, and spatio-temporal consistency analysis on the ADS-B signal and the Remote ID signal respectively. By verifying the similarity between the signal and the normal signal, verifying the matching degree between the signal strength and the distance and the signal strength change, the drone position change, and the signal frequency change and the consistency with the movement law of the aerial vehicle, comprehensively judge whether the signal is a real signal or a forged signal based on the verification results of the ADS-B signal and the Remote ID signal. Identify the drone with a forged signal as an unlicensed drone. If any one of them is judged as a forged signal, it is identified as an unlicensed drone.

[0065] In the embodiments of the present application, the verification results of ADS-B signals and Remote ID signals can be that both signals need to pass registration verification and authenticity verification, or it can be considered that a non-black flying unmanned aircraft if either one passes the verification. It can be flexibly adjusted according to the strictness of the verification results in the control scenario. Among them, when forging the identity of a black flying unmanned aircraft for signals sent through low-cost devices and ground devices and disguising as a low-altitude unmanned aircraft, the embodiments of the present application comprehensively judge by performing spectrum analysis, signal strength analysis, and spatio-temporal consistency analysis on ADS-B signals and Remote ID signals respectively, improving the reliability and robustness of identifying forged signals and avoiding missed or misjudgments that may occur in single-signal analysis.

[0066] In some of the above embodiments, referring to Figure 4 as shown, the real signal judgment step S23 further includes:

[0067] Spectrum analysis step S231, receiving the signal to be verified , performing a fast Fourier transform on the signal to be verified to obtain the spectrum of the signal to be verified , calculating the cosine similarity and Euclidean distance with the spectrum template of the normal signal after normalizing it , and judging whether it is an abnormal spectrum based on the cosine similarity and Euclidean distance. If so, it is a forged signal; where , FFT is the fast Fourier transform. Among them, in the spectrum analysis step S231, the abnormal spectrum is the signal spectrum with a cosine similarity lower than the first threshold and an Euclidean distance exceeding the second threshold. The first threshold in this embodiment is set to 0.8, and the second threshold is set to 0.5, but it is not limited thereto. By using the cosine similarity to focus on the directional consistency of the spectrum shape and the Euclidean distance to analyze the absolute difference of the spectrum amplitude, the abnormal signal can be accurately analyzed; where the cosine similarity is expressed as:

[0068] ;

[0069] The Euclidean distance is expressed as:

[0070] .

[0071] The above steps use the normalization operation to eliminate the interference of the absolute amplitude, making the subsequent similarity calculation pay more attention to the relative relationship of the frequency distribution and making the local frequency distortion of the normalized signal to be verified easier to quantify.

[0072] Signal strength analysis step S232: Obtain the position of the target UAV in the UAV position list, determine the distance between the target UAV and the signal receiving device, calculate the corresponding received power based on this distance using the free space path loss model, and determine whether this received power matches the actual power of the signal to be verified. If not, it is a forged signal. Among them, the received power is expressed as the following calculation model:

[0073] ,

[0074] Among them, is the transmit power, and are the transmit antenna gain and receive antenna gain respectively, is the distance, is the frequency of the normal signal, . In the actual solution process, the transmit antenna gain, receive antenna gain, and normal signal frequency can be set accordingly according to whether the signal to be verified belongs to an ADS-B signal or a Remote ID signal. For example, the frequency of ADS-B is set to 1090 MHz, the transmit antenna gain is set to 2 - 3 dBi, the frequency of Remote ID is set to 2.4 GHz or 5.8 GHz, and the transmit antenna gain is set to 5 dBi.

[0075] Spatial and temporal consistency analysis step S233: Continuously obtain the position of the target UAV and the signal to be verified , analyze whether the change in signal strength during its movement conforms to the free space path loss model, whether the speed change is smooth, and analyze whether the signal frequency offset of the signal to be verified conforms to the expectation. If any one of the analysis results is negative, it is a forged signal.

[0076] Although low-cost devices and ground devices can forge signals, the forged signals may exhibit different spectral distributions from real signals due to hardware limitations (such as transmit power or modulation accuracy). Based on the above spectral analysis step S231, this embodiment identifies abnormal signals forged by low-cost devices by analyzing the frequency characteristics of the signals. Considering that the transmit position of the forged signal generated by the ground device is different from the position of the UAV in the air, this will result in a signal intensity difference. The signal strength analysis step S232 determines whether the signal intensity of the signal to be verified matches the distance of the target UAV based on the radio signal propagation model to eliminate forged signals sent by ground devices.

[0077] In the above embodiment, the spatial and temporal consistency analysis step S233 further includes:

[0078] Signal strength consistency analysis step S2331: Based on the current position of the target UAV, calculate the expected received power at the current position according to the free space path loss model, determine the actual received power at the current position according to the signal to be verified, calculate the mean square error between the actual received power and the expected received power. If it is greater than a set error value, it is determined that the signal is abnormal. In some embodiments, the set error value is 5 dB 2 Specifically, calculate the current distance between the target UAV and the signal receiving device at the current position, substitute the current distance into the calculation model of the received power, and obtain the expected received power , combined with the actual received power of the signal receiving device Calculate the mean square error, and the mean square error MSE is expressed as:

[0079] .

[0080] Position consistency analysis step S2332: Calculate the moving speed and moving acceleration of the target UAV according to its position. If the moving acceleration is greater than an acceleration threshold, it is determined that the signal is abnormal. Among them, the first acceleration threshold is configured as the maximum acceleration of the UAV. If the absolute value of the moving acceleration is greater than the acceleration threshold, it is determined that the position change of the target UAV is not smooth and there is signal abnormality;

[0081] Frequency consistency analysis step S2333: Measure the signal frequency of the signal to be verified and calculate the absolute difference between the actual frequency offset and the expected frequency offset based on its position. If the absolute difference is greater than the offset threshold, it is determined that the signal is abnormal. Optionally, the offset threshold is set to 100 Hz. Among them, the signal frequency is expressed as , and the expected frequency offset is calculated based on the following calculation model:

[0082] , where is the moving speed, is the angle between the moving direction vector of the target UAV and the line of sight direction of the vision sensor. The line of sight direction is the direction from the vision sensor of the inspection UAV A to the target UAV B, as shown in Figure 7 ;

[0083] The actual frequency offset is calculated based on the following calculation model:

[0084] .

[0085] Based on the above steps, the embodiments of the present application determine whether a signal conforms to the physical movement law of an aerial vehicle based on signal strength, position, and frequency change, perform multi-dimensional physical feature verification, thereby identifying the anomalies of forged signals, and can identify the deception of forged signals from the underlying signal characteristics, efficiently identify the behavior of low-cost devices simulating aircraft. The multi-index parallel detection can be completed in milliseconds, significantly improving the identification efficiency. The calculation of signal strength, position positioning, and frequency change detection can all be processed in real time by an embedded device, meeting the real-time requirements of the aircraft monitoring scenario.

[0086] In some other embodiments, the black flight drone identification step S2 further includes:

[0087] An abnormal flight judgment step S24, for the drones identified as real signals in the real signal judgment step S23, analyze the heading information in their ADS-B signals, and judge whether their positions in the drone position list are compliant, that is, conform to their preset flight routes. If not, it is judged as an abnormal flight and belongs to a black flight drone.

[0088] Based on the above steps, the embodiments of the present application can timely detect abnormal flight behaviors (such as deviating from the flight route and illegally entering the no-fly zone) by real-time monitoring of the heading, avoiding potential safety accidents. The flight of drones needs to comply with the preset flight route to ensure a safe interval from other aircraft and a reasonable allocation of airspace resources. By analyzing the heading information, it can be verified whether the drone is flying legally and avoid illegal operations. Moreover, complex ground equipment is not required, and only an on-board ADS-B device is needed to achieve the broadcast and reception of heading information, with flexible deployment.

[0089] Combined with Figure 1 As shown, the embodiments of the present application determine the black flight drones that do not send identity information within the line of sight based on the signal presence / absence judgment step S21, determine the black flight drones with unregistered identity information within the line of sight based on the registration status judgment step S22, determine the black flight drones with forged identity information within the line of sight based on the real signal judgment step S23, and determine the abnormal flight drones within the line of sight based on the abnormal flight judgment step S24. The embodiments of the present application adopt the step-by-step screening method of steps S21~S23 to quantitatively analyze and identify black flight behaviors from various dimensions, not only improving the accuracy and efficiency of black flight monitoring and response, but also reducing safety risks and management costs.

[0090] It should be noted that the steps shown in the above process or the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. And although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here. For example, multiple steps of steps S231~S233 can be executed in parallel, and steps S2331~S2333 can be executed in parallel.

[0091] In addition, the black flight drone recognition method according to the embodiments of the present application described in combination with Figures 2 to 6 can be implemented by the drone device for aerial monitoring.

[0092] The drone device for monitoring may include a processor and a memory storing computer program instructions.

[0093] Specifically, the above-mentioned processor may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0094] Among them, the memory may include a mass memory for data or instructions. By way of example and not limitation, the memory may include a hard disk drive (HDD), a floppy disk drive, a solid state drive (SSD), a flash memory, an optical disc, a magneto-optical disc, a magnetic tape, or a universal serial bus (USB) drive, or a combination of two or more of these. In appropriate cases, the memory may include removable or non-removable (or fixed) media. In appropriate cases, the memory may be internal or external to the data processing device. In a particular embodiment, the memory is a non-volatile memory. In a particular embodiment, the memory includes a read-only memory (ROM) and a random access memory (RAM). In appropriate cases, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically alterable read-only memory (EAROM), or a flash memory (FLASH), or a combination of two or more of these. In appropriate cases, the RAM may be a static random access memory (SRAM) or a dynamic random access memory (DRAM), where the DRAM may be a fast page mode dynamic random access memory (FPMDRAM), an extended date out dynamic random access memory (EDODRAM), a synchronous dynamic random access memory (SDRAM), etc.

[0095] The memory can be used to store or cache various data files required for processing and / or communication, as well as possible computer program instructions executed by the processor.

[0096] The processor reads and executes the computer program instructions stored in the memory to implement any one of the black flying drone recognition methods in the above embodiments.

[0097] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification.

[0098] The above embodiments only represent several implementation manners of the present application. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the invention patent. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several deformations and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application should be subject to the appended claims.

Claims

1. A method for identifying illegal drones, characterized in that: include: A line-of-sight range detection step, detecting drones within the line-of-sight range through a visual sensor and locating the positions of the drones based on a preset target detection algorithm, and establishing a drone position list; The step of identifying illegal drones is to continuously receive ADS-B signals and Remote ID signals through a pre-installed signal receiving device, and based on the position of the drone and the ADS-B signal and Remote ID signal, sequentially verify whether the corresponding drone within the line of sight sends ADS-B signals and Remote ID signals, whether the signals are registered signals, and whether the registered signals are real signals. If any of the verification results is negative, it is determined to be an illegal drone; The steps of identifying illegal drones include: A signal existence judgment step is to parse the ADS-B signal and Remote ID signal, obtain the location information in the ADS-B signal and / or Remote ID signal, compare it with the drone location list, identify the drone that does not send a signal as an illegal drone, and the drone that sends a signal enters the registration status judgment step for registration verification; The registration status judgment step is to parse the ICAO address in the ADS-B signal and / or the serial number in the Remote ID signal, perform real-time query verification in the drone supervision database, identify the drone sending the unregistered signal as an illegal drone, and the drone sending the registered signal enters the real signal judgment step for authenticity verification; The real signal judgment step is to respectively perform frequency characteristic analysis, signal strength analysis and time-space consistency analysis on the ADS-B signal and Remote ID signal, and to judge whether the signal is a real signal or a forged signal based on the verification results of the ADS-B signal and Remote ID signal by verifying the similarity between the signal and the normal signal, the matching degree between the signal strength and the distance, and the consistency between the signal strength change, the position change of the UAV, and the signal frequency change and the movement law of the aerial vehicle, and to identify the UAV with a forged signal as an illegal UAV; the real signal judgment step includes: Spatiotemporal consistency analysis step, continuously obtaining the location of the target drone and the signal to be verified , analyzing whether the signal strength change during its movement conforms to the free space path loss model, whether the speed change is smooth, and analyzing whether the signal frequency offset of the signal to be verified is consistent with expectations. If any of the analysis results is no, it is a forged signal; the spatiotemporal consistency analysis step further includes: The signal strength consistency analysis step is to calculate the expected received power at the current position of the target UAV based on the free space path loss model, determine the actual received power at the current position based on the signal to be verified, and calculate the mean square error between the actual received power and the expected received power. If the error is greater than a set error value, the signal is judged to be abnormal. A position consistency analysis step, calculating the moving speed and moving acceleration of the target UAV according to the position of the target UAV, and if the moving acceleration is greater than a first acceleration threshold, determining that the signal is abnormal; The frequency consistency analysis step measures the signal frequency of the signal to be verified and calculates the absolute difference between the actual frequency offset and the expected frequency offset based on its position. If the absolute difference is greater than the offset threshold, the signal is judged to be abnormal. In the signal strength consistency analysis step, the current distance between the target drone and the signal receiving device at the current position is calculated, and the current distance is substituted into the calculation model of the received power to obtain the expected received power. , combined with the actual receiving power of the signal receiving device The mean square error (MSE) is calculated as: 。 2. The method for identifying illegal drones according to claim 1, characterized in that: The real signal determination step further comprises: Spectrum analysis step, receiving the signal to be verified , perform fast Fourier transform on the signal to be verified to obtain the spectrum of the signal to be verified , and then normalize it to calculate the spectrum template of the normal signal The cosine similarity and Euclidean distance of the spectrum are combined to determine whether it is an abnormal spectrum, and if so, it is a forged signal; The signal strength analysis step obtains the position of the target drone in the drone position list, determines the distance between the target drone and the signal receiving device, calculates the corresponding received power based on the distance using the free space path loss model, and determines whether the received power is consistent with the actual power of the signal to be verified. If not, it is a fake signal.

3. The method for identifying illegal drones according to claim 1 or 2, characterized in that: The step of identifying illegal drones also includes: In the abnormal flight judgment step, for the drone identified as a real signal in the real signal judgment step, the heading information in its ADS-B signal is analyzed, and its position in the drone position list is judged to be compliant, that is, consistent with its preset route. If not, it is judged to be an illegal drone.

4. The method for identifying illegal drones according to claim 2, characterized in that: In the spectrum analysis step, the abnormal spectrum is a signal spectrum whose cosine similarity is lower than a first threshold and whose Euclidean distance exceeds a second threshold.

5. The method for identifying illegal drones according to claim 2, characterized in that: In the signal strength analysis step, the received power is expressed as the following calculation model: ， in, is the transmission power, and are the transmitting antenna gain and the receiving antenna gain, respectively. For distance, is the frequency of the normal signal, .

6. The method for identifying illegal drones according to claim 1, characterized in that: The first acceleration threshold is configured as the maximum acceleration of the drone. If the absolute value of the moving acceleration is greater than the acceleration threshold, it is determined that the position change of the target drone is not smooth and there is a signal abnormality.

7. The method for identifying illegal drones according to claim 5, characterized in that: In the frequency consistency analysis step, the signal frequency is expressed as , the expected frequency offset is calculated based on the following calculation model: ,in is the moving speed, is the angle between the target UAV's motion direction vector and the visual sensor's sight line direction, where the visual sensor points to the target UAV; The actual frequency offset is calculated based on the following calculation model: 。

Images