A method, system, electronic device and storage medium for comprehensive photoelectric platform for unmanned aerial vehicle detection

Abstract

The application provides a kind of integrated optoelectronic platform method, system, electronic equipment and storage medium for unmanned aerial vehicle detection, related to unmanned aerial vehicle detection and positioning technical field, the application generates the multi-channel signal data with accurate time stamp by collecting unmanned aerial vehicle remote control signal original signal, triggers optoelectronic platform to capture suspicious airspace high-definition video stream and builds the space-time corresponding database of both;Again, with high-precision atomic clock as time reference, synchronize each radio frequency receiving node, process multi-channel signal data to get time-aligned basic signal;Then extract the radio frequency characteristics of basic signal and correlate equipment identification, build its mapping relationship with remote control model, construct hardware feature library;Finally, combined with library identification result, signal time difference and optical data, use particle filtering to correct multipath error, predict and update operator position, realize unmanned aerial vehicle comprehensive detection and positioning after iterative convergence, can realize the integrated detection and positioning of unmanned aerial vehicle combining radio frequency detection and optical observation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of unmanned aerial vehicle detection and positioning, and particularly relates to a comprehensive optoelectronic platform method and system for unmanned aerial vehicle detection, an electronic device and a storage medium. BACKGROUND

[0002] In cities with high-rise buildings and complex electromagnetic environments, unmanned aerial vehicle (UAV) irregular flight occurs frequently. It is necessary to solve the signal problems caused by non-line-of-sight propagation and building multipath reflection. At the same time, it is necessary to accurately identify UAV remote control signals to exclude interference, and to realize real-time tracing of the operator and positioning error meeting the requirements of the city, thereby providing support for airspace supervision and public safety.

[0003] At present, the mainstream scheme is to use a single radio frequency node monitoring system to collect UAV remote control signals through fixed radio frequency equipment, analyze the source direction according to the signal strength, identify the signal type in combination with a preset frequency band database, and estimate the approximate area of the operator by a single point azimuth angle to assist in completing detection and preliminary tracing.

[0004] However, this system has three defects: first, the anti-interference is weak, and the signal fluctuation caused by city multipath reflection is easy to misjudge the interference signal; second, the positioning accuracy is low, the single point azimuth angle cannot solve the non-line-of-sight deviation, the tracing error is large, and it is difficult to meet the accurate management; third, there is a lack of data verification, and the signal is not verified in combination with optical observation, which is easy to appear false alarm, missed report, and no complete detection closed loop. SUMMARY

[0005] The present application aims to provide a comprehensive optoelectronic platform method and system for unmanned aerial vehicle detection, an electronic device and a storage medium to solve the problems of signal misjudgment and low positioning accuracy in the prior art.

[0006] To solve the above technical problems, in a first aspect, the present application provides a comprehensive optoelectronic platform method for unmanned aerial vehicle detection, comprising:

[0007] Collecting original signals of UAV remote control signals, generating multi-channel signal data with accurate time stamps based on the original signals, synchronously triggering an optoelectronic platform to capture high-definition video streams of suspicious airspace, and establishing a space-time correspondence database of the multi-channel signal data and the high-definition video streams;

[0008] Configuring a high-precision atomic clock as a unified time reference, achieving time synchronization of each radio frequency receiving node through a precise time synchronization protocol, performing frequency conversion and signal demodulation processing on the signals based on the frequency characteristics and modulation characteristics in the multi-channel signal data, and obtaining time-aligned basic signals;

[0009] extracting a radio frequency feature in the base signal, associating the radio frequency feature with device identification information in the multi-path signal data, establishing a mapping relationship between the radio frequency feature and the model of the unmanned aerial vehicle remote controller through feature digitization processing and classification analysis, and constructing a hardware feature library including a feature template, a threshold parameter and a matching algorithm according to the mapping relationship;

[0010] According to the recognition result of the hardware feature library, combining the signal time difference in the base signal of the multiple nodes and the optical observation data in the space-time corresponding database, the multi-path error is corrected and the next time position of the operator is predicted through the particle filtering technology, and then the predicted position is updated by using the compensation result of the corrected multi-path error, and the iteration is repeated until convergence, realizing the comprehensive detection and positioning of the unmanned aerial vehicle combining the radio frequency detection and the optical observation.

[0011] According to the recognition result of the hardware feature library, combining the signal time difference in the base signal of the multiple nodes and the optical observation data in the space-time corresponding database, the multi-path error is corrected and the next time position of the operator is predicted through the particle filtering technology, and then the predicted position is updated by using the compensation result of the corrected multi-path error, and the iteration is repeated until convergence, realizing the comprehensive detection and positioning of the unmanned aerial vehicle combining the radio frequency detection and the optical observation.

[0012] The collected base signal is input into the hardware feature library, and the recognition result of the model of the unmanned aerial vehicle remote controller is output through the matching algorithm, the time difference of the base signal between different radio frequency receiving nodes is calculated, and the initial position estimation area is generated in combination with the signal propagation speed;

[0013] The optical observation data corresponding to the recognition result is extracted from the space-time corresponding database, and the optical observation data includes the pixel coordinates of the unmanned aerial vehicle in the video frame, the initial position estimation area and the optical observation data are input into the position prediction unit, and the initial position prediction point of the operator is generated;

[0014] Based on the particle filtering model, the theoretical signal propagation time delay of the initial position prediction point and each radio frequency receiving node is calculated;

[0015] The theoretical signal propagation time delay is compared with the actual time difference to generate a multi-path error compensation amount, and the initial position prediction point is corrected based on the multi-path error compensation amount, and an updated predicted position is output;

[0016] The error compensation and position updating operation is repeatedly executed until the deviation of the predicted position is less than the convergence threshold value for two consecutive times, and the comprehensive detection and positioning result of the unmanned aerial vehicle combining the radio frequency detection and the optical observation is obtained.

[0017] Optionally, based on the particle filtering model simulating the multipath propagation path, the theoretical signal propagation time delay of the initial position prediction point and each radio frequency receiving node is calculated, including:

[0018] Based on the initial position prediction point and the radio frequency receiving node position, a particle set representing different signal reflection paths is generated, each particle containing a path length parameter and a reflection point position parameter;

[0019] The matching degree of the path length parameter of each particle in the particle set and the initial position prediction point is calculated, and a preset number of particles with the highest matching degree are retained to form a particle subset;

[0020] Particles in the particle subset are randomly paired to form a plurality of particle pairs, the reflection point position parameters of each particle pair are exchanged, a new reflection path is generated based on the exchanged reflection point position parameters, and new particles corresponding to the new reflection path form a new particle set;

[0021] A part of particles in the new particle set are randomly selected, and the reflection point position parameters of the selected particles are randomly disturbed to form an optimized particle set. Based on the optimized particle set, the path length of each propagation path from the initial position prediction point to each radio frequency receiving node is calculated;

[0022] According to the electromagnetic wave propagation speed, the path length of each propagation path is converted into single-path propagation time, and the single-path propagation time is weighted and combined according to the particle weight to generate the theoretical signal propagation time delay.

[0023] Optionally, the radio frequency features in the basic signal are extracted, the radio frequency features are associated with the device identification information in the multi-path signal data, the mapping relationship between the radio frequency features and the model of the unmanned aerial vehicle remote controller is established through feature digitization processing and classification analysis, and the hardware feature library containing the feature template, the threshold parameter and the matching algorithm is constructed according to the mapping relationship, including:

[0024] The signal amplitude, frequency offset and modulation period are separated from the basic signal as radio frequency features, and the radio frequency features are stored in the form of numerical vectors;

[0025] The preset device coding field in the multi-path signal data is associated with the numerical vector of the radio frequency features to form a binding data set of devices and features;

[0026] The binding data set is subjected to feature digitization processing, the numerical vector of the radio frequency features is quantized into a digital matrix of fixed dimension, and the digital matrix is subjected to clustering division to generate feature clustering clusters corresponding to different models of unmanned aerial vehicle remote controllers;

[0027] establish a mapping relationship between the feature clustering cluster and the model of the unmanned aerial vehicle remote controller, and extract a center vector corresponding to the feature clustering cluster of each model based on the mapping relationship to generate a feature template;

[0028] Based on the mapping relationship, a threshold parameter is set in combination with a clustering boundary distance, a matching algorithm based on similarity calculation is designed, the feature template, the threshold parameter and the matching algorithm are integrated, and a dynamically updated hardware feature library is constructed.

[0029] Optionally, the mapping relationship between the feature clustering cluster and the model of the unmanned aerial vehicle remote controller is established, and the center vector corresponding to the feature clustering cluster of each model is extracted based on the mapping relationship to generate a feature template, comprising:

[0030] The device code field is read from the binding data set, and each feature clustering cluster is mapped to a unique model of the unmanned aerial vehicle remote controller according to the model information contained in the device code field, forming a mapping relationship between the model and the feature clustering cluster;

[0031] Based on the mapping relationship, the feature clustering cluster associated with each model of the unmanned aerial vehicle remote controller is located, and the arithmetic mean of all numerical vectors in the feature clustering cluster is calculated to obtain a center vector representing the characteristics of each model;

[0032] The center vector of each model of the unmanned aerial vehicle remote controller is standardized to adjust the numerical precision of all vector elements to a unified standard, and the standardized center vector is packaged as a fixed format data structure to generate a feature template with unified dimensions and coding format.

[0033] Optionally, a high-precision atomic clock is configured as a unified time reference, and each radio frequency receiving node is time-synchronized through a precise time synchronization protocol. Based on the frequency characteristics and modulation characteristics in the multi-channel signal data, frequency conversion and signal demodulation processing are performed on the signal to obtain a time-aligned basic signal, comprising:

[0034] A high-stability atomic clock is deployed as a unified time reference. The atomic clock outputs a standard clock signal to all radio frequency receiving nodes, and the standard clock signal is distributed through a wired transmission link to make the deviation of the local clock of each radio frequency receiving node from the unified time reference less than a microsecond threshold, thereby completing the time synchronization between nodes;

[0035] The frequency characteristic parameters and the modulation waveform characteristic parameters are separated from the multi-channel signal data, and a frequency conversion unit is used to convert the original signals of different wave bands into unified intermediate frequency signals;

[0036] The demodulation operation is performed on the unified intermediate frequency signal to strip the carrier component and extract the baseband signal to obtain a basic signal containing timing information;

[0037] The base signals are time-aligned by timing calibration according to the time synchronization result between nodes, phase offset caused by transmission delay between nodes is eliminated, and time-aligned base signals are generated.

[0038] Optionally, original signals of the UAV remote control signals are collected, multi-channel signal data with accurate time stamps are generated based on the original signals, high-definition video streams of suspicious airspace are captured by the optoelectronic platform triggered synchronously, and a space-time corresponding database of the multi-channel signal data and the high-definition video streams is established, including:

[0039] Original signals transmitted by the UAV remote controller are captured at multiple radio frequency receiving nodes, the original signals contain frequency bands and modulation waveforms, and initial time information of the original signals is recorded by a signal capturing unit;

[0040] The initial time information is input into a time marking unit, a unique time identifier is added to each of the original signals in combination with a global clock reference, and multi-channel signal data with accurate time stamps are generated;

[0041] Based on the time identifier sequence in the multi-channel signal data, the optoelectronic sensor is controlled to perform directional scanning on the suspicious airspace, and high-definition video streams containing UAV images are captured;

[0042] The time identifier of the multi-channel signal data is matched and mapped with frame time codes of the high-definition video streams, and a space-time corresponding database containing signal feature fields and video frame coordinates is formed.

[0043] In a second aspect, the application provides a comprehensive optoelectronic platform system for UAV detection, including:

[0044] A collection module is configured to collect original signals of the UAV remote control signals, generate multi-channel signal data with accurate time stamps based on the original signals, trigger the optoelectronic platform to capture high-definition video streams of suspicious airspace synchronously, and establish a space-time corresponding database of the multi-channel signal data and the high-definition video streams;

[0045] A conversion module is configured to configure a high-precision atomic clock as a unified time reference, achieve time synchronization of each radio frequency receiving node through a precise time synchronization protocol, perform frequency conversion and signal demodulation processing on the signals based on frequency features and modulation features in the multi-channel signal data, and obtain time-aligned base signals;

[0046] A construction module is configured to extract radio frequency features in the base signals, associate the radio frequency features with device identification information in the multi-channel signal data, establish a mapping relationship between the radio frequency features and the UAV remote controller models through feature digitization processing and classification analysis, and construct a hardware feature library containing feature templates, threshold parameters and matching algorithms according to the mapping relationship;

[0047] The updating module is configured to correct and predict the position of the operator at the next moment according to the identification result of the hardware feature library, the time difference of the signals in the base signals of the plurality of nodes and the optical observation data in the space-time corresponding database, update the predicted position by using the corrected multipath error compensation result, and iteratively update until convergence, thereby realizing the comprehensive detection and positioning of the unmanned aerial vehicle by combining the radio frequency detection and the optical observation.

[0048] In a third aspect, the present application provides an electronic device, comprising:

[0049] a memory configured to store a computer program;

[0050] a processor configured to execute the computer program to implement the steps of the comprehensive optoelectronic platform method for unmanned aerial vehicle detection according to the first aspect.

[0051] In a fourth aspect, the present application provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program is executable by a processor to implement the steps of the comprehensive optoelectronic platform method for unmanned aerial vehicle detection according to the first aspect.

[0052] The comprehensive optoelectronic platform method for unmanned aerial vehicle detection provided by the present application can realize the space-time correlation of the radio frequency signal and the optical data by collecting the original signals of the unmanned aerial vehicle remote control signals, generating multi-channel signal data with accurate time stamps, synchronously triggering the high-definition video stream of the suspicious airspace captured by the optoelectronic platform, and establishing the space-time corresponding database of the two, thereby laying a foundation for subsequent multi-source data fusion analysis. By configuring a high-precision atomic clock as a unified time reference, the time of each radio frequency receiving node is synchronized, the time-aligned base signals are obtained by processing the multi-channel signal data, the time deviation between the nodes can be eliminated, the time consistency of the signal analysis is ensured, and reliable data is provided for subsequent time difference positioning. By extracting the radio frequency features of the base signals and associating the equipment identification information, the mapping relationship between the hardware feature library and the model of the unmanned aerial vehicle remote controller is established, which can realize the accurate identification and model matching of the unmanned aerial vehicle remote control signals, effectively distinguish the interference signals, and improve the accuracy of signal identification. By combining the identification result of the hardware feature library, the time difference of the signals of the plurality of nodes and the optical observation data, the multipath error is corrected and the position of the operator is iteratively updated by using the particle filtering technology, the advantages of radio frequency detection and optical observation can be combined, the positioning deviation caused by non-line-of-sight and multipath reflection can be effectively solved, and the high-precision comprehensive detection and positioning of the unmanned aerial vehicle and the operator can be realized. BRIEF DESCRIPTION OF DRAWINGS

[0053] In order to more clearly illustrate the technical solutions of the embodiments of the present application or the prior art, the accompanying drawings required by the embodiments or the prior art description will be briefly introduced. Obviously, the accompanying drawings in the following description only constitute some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0054] Figure 1 A flowchart of a comprehensive optoelectronic platform method for unmanned aerial vehicle detection provided by an embodiment of the present application;

[0055] Figure 2 A detailed implementation flowchart of a comprehensive optoelectronic platform method for unmanned aerial vehicle detection provided by an embodiment of the present application;

[0056] Figure 3 A scene diagram of a comprehensive optoelectronic platform method for unmanned aerial vehicle detection provided by an embodiment of the present application;

[0057] Figure 4 A structural diagram of a comprehensive optoelectronic platform system for unmanned aerial vehicle detection provided by an embodiment of the present application. DETAILED DESCRIPTION

[0058] In the urban unmanned aerial vehicle supervision scene, the existing monitoring scheme relying on a single radio frequency node faces obvious bottlenecks: on the one hand, the multipath reflection of buildings in the city will make the signal unstable, and the system is easy to mistake other interference signals as unmanned aerial vehicle remote control signals, resulting in inaccurate identification; on the other hand, only relying on a single node to analyze the signal direction to estimate the operator position, the error is large in non-line-of-sight environment, and it is difficult to meet the precise positioning needs of urban management; at the same time, the system only looks at the radio frequency signal and does not combine with the optical observation data, and it cannot verify whether the signal is truly corresponding to the unmanned aerial vehicle, which is easy to cause false positives or false negatives, and cannot form a complete monitoring process.

[0059] In view of these problems, the present application provides a comprehensive optoelectronic platform method for unmanned aerial vehicle detection. The method first collects unmanned aerial vehicle remote control signals and generates multi-channel data with accurate time markers, and simultaneously triggers the device to shoot high-definition video of the suspicious airspace, and establishes an associated database of the two; then through a high-precision clock, multiple signal receiving nodes are time-synchronized, and the signal is processed to obtain basic data with a unified time reference; then the signal features are extracted and associated with device information, and a feature library capable of accurately identifying the model of the remote controller is established; finally, combined with the feature library recognition result, the multi-node signal time difference and the video data, the signal error is corrected by technical means, and the operator position estimation is repeatedly optimized until accurate. This scheme not only solves the signal misjudgment problem through the feature library and video data, but also improves the positioning accuracy with the help of multi-node synchronization and error correction, and forms a complete process of "signal collection-identification-positioning-verification", which fundamentally makes up for the defects of the existing single radio frequency node system.

[0060] In order to make the person skilled in the art better understand the scheme of the present application, the present application will be further described in detail below in combination with the drawings and specific embodiments. Obviously, the described embodiments are only part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of protection of the present application.

[0061] The core of the present application is to provide a comprehensive optoelectronic platform method for unmanned aerial vehicle detection, and a specific embodiment process schematic diagram of the present application is shown in Figure 1 The method comprises the following steps:

[0062] S101, collecting the original signal of the unmanned aerial vehicle remote control signal, generating multi-channel signal data with accurate time stamp based on the original signal, synchronously triggering the optoelectronic platform to capture the high-definition video stream of suspicious airspace, and establishing the space-time corresponding database of the multi-channel signal data and the high-definition video stream;

[0063] Optionally, step S101 can specifically include the following steps:

[0064] S1011, capturing the original signal emitted by the unmanned aerial vehicle remote controller at multiple radio frequency receiving nodes, the original signal containing frequency band and modulation waveform, and recording the initial time information of the original signal through the signal capture unit;

[0065] S1012, inputting the initial time information into the time marking unit, adding a unique time identifier to each original signal in combination with the global clock reference, and generating multi-channel signal data with accurate time stamp;

[0066] S1013, based on the time identifier sequence in the multi-channel signal data, controlling the optoelectronic sensor to perform directional scanning on the suspicious airspace, and capturing the high-definition video stream containing the image of the unmanned aerial vehicle;

[0067] S1014, matching and mapping the time identifier of the multi-channel signal data with the frame time code of the high-definition video stream, forming the space-time corresponding database containing the signal feature field and the video frame coordinates.

[0068] In the above steps, the radio frequency receiving node is a device for receiving signals transmitted by a drone remote controller, which can be deployed at different locations to cover a wider monitoring range; the original signal refers to an unprocessed electrical signal directly transmitted by the drone remote controller, which contains a frequency band and a modulated waveform, the frequency band is the frequency range of signal transmission, and the modulated waveform is the waveform form of signal information bearing; the signal capture unit is a module for collecting original signals and recording the initial time of signal reception, the initial time information is the time data when the original signal recorded by the signal capture unit is received; the time marking unit is a module for adding time identifiers to the original signal, the global clock reference is a reference source for providing a unified time standard for all monitoring devices, ensuring consistent time for each device; the time identifier is a unique time code generated by the time marking unit, used to identify the time information of each signal data; the multi-channel signal data is a set of signals collected by multiple radio frequency receiving nodes and added with time identifiers; the photoelectric sensor is a device for capturing spatial images and videos, which can realize directional scanning function; the suspicious airspace is the airspace range where the presence of a drone is monitored; the high-definition video stream is a sequence of high-definition continuous images captured by the photoelectric sensor, and the frame time code is the time code corresponding to each image in the high-definition video stream; the space-time correspondence database is a database that associates multi-channel signal data with high-definition video streams, wherein the signal feature field is information describing signal properties, and the video frame coordinate is the position information of the target in each frame of the high-definition video stream.

[0069] In the embodiments of the present application, first, in step S1011, the original signals transmitted by the drone remote controller are captured at multiple radio frequency receiving nodes, which contain the frequency band of signal transmission and the modulated waveform bearing information, and the initial time information of each original signal received is recorded by the signal capture unit. For example, in the drone monitoring scene in area A, multiple radio frequency receiving nodes are deployed by the staff at different positions in the area, when the remote controller of a certain drone transmits a signal, multiple radio frequency receiving nodes receive this original signal at the same time, and the signal capture unit of each node records the initial time information of receiving the signal respectively, such as a certain node receives the signal at 9:00:01 am, and another node receives the same signal at 9:00:01 am.

[0070] Secondly, by step S1012, the initial time information recorded by each radio frequency receiving node in step S1011 is input into the time marking unit, and the time marking unit generates a unique time identifier for each original signal in combination with the pre-set global clock reference, and then integrates the multiple groups of original signals with the time identifier into multi-channel signal data with accurate time stamps. For example, in the scenario of the A area described above, the initial time information recorded by each radio frequency receiving node is input into the time marking unit, which generates a time identifier T001 for the original signal received at 9:00:01 am and a time identifier T002 for the original signal received at 9:00:02 am based on the global clock reference of the monitoring system of the A area, and then arranges the signal data with the time identifiers T001, T002, etc. into multi-channel signal data to ensure that each channel of data has an accurate time mark.

[0071] Next, by step S1013, based on the time identifier sequence in the multi-channel signal data generated in step S1012, the system sends a control instruction to the photoelectric sensor to control the photoelectric sensor to perform directional scanning on the suspicious airspace where the unmanned aerial vehicle may exist, and to capture a high-definition video stream containing the image of the unmanned aerial vehicle during the scanning process. For example, in the scenario of the A area, the system reads the time identifier sequence in the multi-channel signal data and finds that the signals corresponding to the time identifiers T001, T002, etc. all come from the airspace in the northeast direction of the area, and determines that the airspace is a suspicious airspace. Subsequently, the photoelectric sensor deployed in the area is controlled to turn to the northeast direction to perform directional scanning, and a high-definition video stream of the unmanned aerial vehicle in the airspace is successfully captured during the scanning process, and the flight state of the unmanned aerial vehicle can be clearly seen in the video stream.

[0072] Finally, by step S1014, the time identifiers of the multi-channel signal data generated in step S1012 are one-to-one matched and mapped with the frame time codes of the high-definition video stream captured in step S1013, the signal characteristics corresponding to each time identifier are arranged into a signal characteristic field, and the position information of the unmanned aerial vehicle in the video image corresponding to the time frame is recorded as a video frame coordinate, and finally a space-time correspondence database containing the signal characteristic field and the video frame coordinate is integrated. For example, in the scenario of the A area, the time identifier T001 is matched with the frame time code of the video frame corresponding to 9:00:01 am in the high-definition video stream, the signal frequency, modulation mode, etc. corresponding to T001 are arranged into a signal characteristic field, and the horizontal and vertical coordinates of the unmanned aerial vehicle in the image in the video frame are recorded as a video frame coordinate. Similarly, other time identifiers such as T002 are matched and processed, and finally a complete space-time correspondence database is constructed to realize the time and space correlation of the signal data and the video data.

[0073] In practical application, the airspace management department of A city strengthens the unmanned aerial vehicle supervision in key areas such as core business circles and traffic hubs in the jurisdiction by deploying unmanned aerial vehicle detection systems in three layers around these key areas: the first layer deploys eight radio frequency receiving nodes within a one-kilometer range around the key areas, which are evenly distributed on the roofs of high-rise buildings and high places such as street lamp poles to ensure that there is no signal monitoring blind area; the second layer deploys one high-precision time tagging unit on the roof of a regional central office, which accesses the global clock reference of the city government network, which is synchronized with the national time center and can provide microsecond-level time accuracy; the third layer deploys one high-definition photoelectric sensor on each of the four high-rise buildings at the corners of the key area, and the sensor has a 360-degree rotating scanning function, supports 1080P resolution video capture, and can respond to millisecond-level control instructions. When the system backend detects a suspected unmanned aerial vehicle remote control signal fluctuation in the airspace east of the core business circle, it immediately triggers the S101 process: first, three of the eight radio frequency receiving nodes simultaneously capture the original signal of the remote control signal, which contains a 2.4 GHz frequency band and an FSK modulated waveform, and each node's signal capture unit records the initial time information through the built-in time module - the east node is 9:12:03:150 milliseconds, the north node is 9:12:03:152 milliseconds, and the south node is 9:12:03:151 milliseconds; then, the initial time information of these three nodes is transmitted to the central time tagging unit through a dedicated data link, and the time tagging unit generates unique time identifiers T20240828091203150, T20240828091203152, and T20240828091203151 for the three original signals based on the global clock reference, and integrates the time identifiers with corresponding signal frequencies, modulation waveforms, and other data to generate three routes of signal data with accurate time stamps; subsequently, the system backend reads the time identifier sequence of the three routes of data, analyzes that the signal source is approximately in the east airspace, and immediately sends directional scanning instructions to the two photoelectric sensors in the east and south, which adjust to the east 30-degree angle direction within one second, and start capturing high-definition video streams at a rate of 25 frames per second, and within 10 seconds, an image of a small unmanned aerial vehicle is captured in the video; finally, the system matches the time identifiers of the three routes of signal data with the frame time codes of the video stream, such as matching T20240828091203150 with the 375th frame corresponding to 9:12:03:150 milliseconds in the video stream, extracting the pixel coordinates (X:850, Y:420) of the unmanned aerial vehicle in that frame as the video frame coordinates, and recording the frequency, modulation method, and other signal characteristics of that signal as signal characteristic fields, and finally constructing a space-time correspondence database containing 20 matching data groups to prepare data for subsequent signal identification and operator positioning.

[0074] In the overall scheme of step S101, by capturing the original signal and generating multi-channel signal data with accurate time stamp, the high-definition video stream is synchronously acquired and the space-time corresponding database is established, realizing the accurate correlation of the unmanned aerial vehicle remote control signal and the optical image data in time and space dimensions. This correlation not only provides a data basis for subsequent differentiation of interference signals and improvement of signal recognition accuracy, but also creates conditions for fusion of radio frequency detection and optical observation results and improvement of operator positioning accuracy, at the same time making the data of the entire detection process traceable, which is helpful to form a complete detection process and provide reliable preliminary data support for unmanned aerial vehicle supervision in urban dense environment.

[0075] S102, configure a high-precision atomic clock as a unified time reference, synchronize the time of each radio frequency receiving node through a precise time synchronization protocol, and based on the frequency characteristics and modulation characteristics in the multi-channel signal data, perform frequency conversion and signal demodulation processing on the signal to obtain a time-aligned basic signal;

[0076] Optionally, step S102 can specifically include the following steps:

[0077] S1021, deploy a high-stability atomic clock as a unified time reference, the atomic clock outputs a standard clock signal to all radio frequency receiving nodes, and distributes the standard clock signal through a wired transmission link, so that the deviation of the local clock of each radio frequency receiving node from the unified time reference is less than a microsecond threshold, to complete the time synchronization between nodes;

[0078] S1022, separate the frequency characteristic parameters and the modulation waveform characteristic parameters from the multi-channel signal data, and use a frequency conversion unit to convert the original signals of different wave bands into unified intermediate frequency signals;

[0079] S1023, perform demodulation operation on the unified intermediate frequency signal, strip the carrier component and extract the baseband signal to obtain a basic signal containing timing information;

[0080] S1024, according to the time synchronization result between nodes, perform timing calibration on the basic signal to eliminate the phase offset caused by the transmission delay between nodes, and generate a time-aligned basic signal.

[0081] In the above steps, the high-precision atomic clock is a clock device with high stability time output capability, used to provide a unified and accurate time reference standard; the unified time reference is a time standard followed by all radio frequency receiving nodes to ensure consistency of the nodes; the precise time synchronization protocol is a communication rule for aligning the time of each node with the unified time reference; the radio frequency receiving node is a device for receiving unmanned aerial vehicle remote control signals, and the local clock is a time recording module carried by the radio frequency receiving node itself; the microsecond threshold is the maximum allowed value for measuring the deviation of the node local clock from the unified time reference, and a deviation less than the value is considered to be time synchronized; the wired transmission link is a physical line for transmitting the standard clock signal of the atomic clock to ensure stable signal transmission; the multi-channel signal data is a signal set with accurate time stamps, the frequency characteristic parameter is data describing the frequency attribute of the signal, and the modulation waveform characteristic parameter is data describing the modulation mode waveform of the signal; the frequency conversion unit is a module for converting signals of different frequency bands to fixed frequency signals; the original signal is an untreated signal transmitted by the unmanned aerial vehicle remote controller; the unified intermediate frequency signal is a fixed frequency signal obtained by converting different band original signals, facilitating subsequent unified processing; the demodulation operation is a process of extracting original information from a modulated signal; the carrier component is a high-frequency signal part carrying signal information; the baseband signal is a low-frequency original information signal obtained after demodulation, containing the timing information of the signal; the basic signal is a signal with timing information after demodulation; the timing calibration is an operation of adjusting the time sequence of the signal according to the time synchronization result; the transmission delay is the time delay generated during the transmission of the signal between nodes; the phase shift is a phenomenon of inconsistent signal phases due to transmission delay; and the time-aligned basic signal is a basic signal whose signals of each node remain consistent in the time dimension after timing calibration.

[0082] In the embodiments of the present application, first, a high-stability atomic clock is deployed as a unified time reference through step S1021. The atomic clock continuously outputs a standard clock signal, which is distributed to all radio frequency receiving nodes through a wired transmission link. After receiving the signal, each node automatically calibrates its local clock to control the deviation of the local clock from the unified time reference within a microsecond threshold, thereby completing the time synchronization between nodes. For example, in an unmanned aerial vehicle detection system in a key area of City A, a rubidium atomic clock is deployed as a unified time reference. The time stability of the clock can reach a daily deviation of less than 1 microsecond. The standard clock signal is sent to 8 radio frequency receiving nodes through a shielded network cable (wired transmission link). Each node is built-in with a time calibration module, which compares the difference between the local clock and the standard clock after receiving the signal. If the local clock of a certain node is 2 microseconds slower than the standard clock, the calibration module automatically adjusts the local clock to be 2 microseconds faster. Finally, the deviation of all nodes from the unified time reference is controlled within 0.5 microseconds, achieving time synchronization.

[0083] Secondly, by step S1022, the frequency characteristic parameters and the modulation waveform characteristic parameters of each signal are separated from the multi-channel signal data generated by S101 using a signal feature extraction algorithm, and then the parameters are input into a frequency conversion unit. The frequency conversion unit converts the original signals of different wave bands into unified intermediate frequency signals through a frequency mixer according to a preset intermediate frequency value. For example, in the above scenario, the multi-channel signal data contains three original signals of different wave bands, 2.4 GHz, 5.8 GHz and 915 MHz. The frequency value of each signal, i.e. the frequency characteristic parameter, and the modulation type, i.e. the modulation waveform characteristic parameter, such as FSK and OFDM, are separated by the feature extraction algorithm. Then the frequency conversion unit sets the preset intermediate frequency value to 70 MHz, mixes the 2.4 GHz signal with the 2.33 GHz local oscillator signal, the 5.8 GHz signal with the 5.73 GHz local oscillator signal, and the 915 MHz signal with the 845 MHz local oscillator signal through the frequency mixer, and finally converts the three original signals of different wave bands into unified intermediate frequency signals of 70 MHz.

[0084] Next, by step S1023, a demodulation operation is performed on the unified intermediate frequency signal obtained by step S1022, and a coherent demodulation algorithm is used to strip the carrier component in the signal to extract a baseband signal containing information such as unmanned aerial vehicle control instructions. The baseband signal carries the timing information of the signal over time, i.e. a basic signal containing timing information is obtained. For example, in the above scenario, for the unified intermediate frequency signal of 70 MHz, a coherent demodulation algorithm is used. First, the carrier extraction module recovers the carrier signal of the same frequency and phase as the transmitting end from the intermediate frequency signal. Then, the intermediate frequency signal is multiplied by the recovered carrier signal. After filtering the high frequency component through a low pass filter, the 70 MHz carrier component is stripped off, and finally a baseband signal with a frequency of 0-10 MHz is extracted. The baseband signal contains the timing information of the “rise” and “turn” control instructions sent by the unmanned aerial vehicle remote controller, forming a basic signal.

[0085] Finally, by step S1024, the timing adjustment algorithm is used to perform timing adjustment on the basic signal obtained by step S1023 according to the inter-node time synchronization result obtained by step S1021. The time difference between the basic signals of each node caused by transmission delay is calculated, the phase of the signal is compensated according to the time difference, the phase shift caused by the transmission delay between the nodes is eliminated, and finally the time-aligned basic signal is generated. For example, in the above scenario, it is known that the eight radio frequency receiving nodes have achieved time synchronization. Through the timing calibration algorithm, it is found that the basic signal of the north node arrives 0.3 microseconds later than that of the east node. The algorithm automatically advances the basic signal of the north node by 0.3 microseconds on the time axis, and adjusts the phase of the signal to match the east node signal. After calibration, the starting time and signal period of the basic signal of all nodes on the time axis remain consistent, and the phase shift is completely eliminated, obtaining the time-aligned basic signal.

[0086] In practical applications, the airspace management department of City A deploys a complete S102 processing flow in the core business district monitoring system to improve the signal processing accuracy of unmanned aerial vehicle detection in key areas. First, a brand A rubidium atomic clock is deployed in the system control center as a unified time reference. The standard clock signal is distributed to eight radio frequency receiving nodes around the business district through a six-class shielding network. Each node synchronizes with the reference clock through the built-in time calibration software, ensuring a deviation of less than 0.5 microseconds. When the system captures three unmanned aerial vehicle remote control original signals (2.4 GHz, 5.8 GHz, 915 MHz), the signal processing software separates the frequency and modulation characteristics of each signal, and the hardware frequency conversion module converts them into 70 MHz intermediate frequency signals. Then, the coherent demodulation module processes the intermediate frequency signals to obtain the baseband signal containing the control command after stripping the carrier wave. Finally, according to the node time synchronization data, the time and phase of each node baseband signal are adjusted through the timing calibration tool to eliminate the transmission delay deviation caused by the difference in node location. Ultimately, three time-aligned basic signals are output, providing a unified and accurate signal data source for subsequent radio frequency feature extraction and operator positioning.

[0087] In the overall scheme of the above step S102, the time synchronization of each radio frequency receiving node is achieved by deploying high-precision atomic clocks and precise synchronization protocols, ensuring the time consistency of signal processing. The frequency conversion unifies different band signals into intermediate frequency signals, reducing the complexity of subsequent signal processing. The demodulation operation extracts the baseband signal containing the core information, providing effective data for feature analysis. The timing calibration eliminates the phase shift caused by transmission delay, obtaining time-aligned basic signals. The entire process solves the problems of different node signal time synchronization, large band differences, and multiple signal clutter, laying a reliable signal foundation for subsequent construction of hardware feature library and implementation of high-precision positioning, while improving the efficiency and accuracy of signal processing, making the subsequent unmanned aerial vehicle detection process smoother and the results more reliable.

[0088] S103, extract the radio frequency features from the basic signal, associate the radio frequency features with the device identification information in the multi-channel signal data, establish the mapping relationship between the radio frequency features and the unmanned aerial vehicle remote controller model through feature digitization processing and classification analysis, and construct a hardware feature library containing feature templates, threshold parameters, and matching algorithms according to the mapping relationship;

[0089] Optionally, step S103 can specifically include the following steps:

[0090] S1031, separate the signal amplitude, frequency offset, and modulation period from the basic signal as radio frequency features, and store the radio frequency features in the form of numerical vectors;

[0091] S1032, associate the preset device code field in the multi-channel signal data with the numerical vector of the radio frequency feature to form a binding data set of device and feature;

[0092] S1033, perform feature digitization processing on the binding data set, quantize the numerical vector of the radio frequency feature into a fixed-dimension digital matrix, cluster and divide the digital matrix, and generate feature clustering clusters corresponding to different models of unmanned aerial vehicle remote controllers;

[0093] S1034, establish a mapping relationship between the feature clustering clusters and the models of unmanned aerial vehicle remote controllers, and extract the center vectors of the feature clustering clusters corresponding to each model based on the mapping relationship to generate feature templates;

[0094] The step S1034 can specifically include the following processes: reading the device code field from the binding data set, mapping each feature clustering cluster to a unique model of unmanned aerial vehicle remote controller according to the model information contained in the device code field, forming a mapping relationship between the model and the feature clustering cluster; based on the mapping relationship, locating the feature clustering cluster associated with each model of unmanned aerial vehicle remote controller, calculating the arithmetic mean of all numerical vectors in the feature clustering cluster to obtain the center vector representing the feature of each model; standardizing the center vector of each model of unmanned aerial vehicle remote controller, adjusting the numerical precision of all vector elements to a unified standard, encapsulating the standardized center vector into a fixed-format data structure, and generating a feature template with unified dimension and code format.

[0095] S1035, based on the mapping relationship combined with the clustering boundary distance, set a threshold parameter, and design a matching algorithm based on similarity calculation, integrate the feature template, the threshold parameter and the matching algorithm, and construct a dynamically updated hardware feature library.

[0096] In the above steps, the radio frequency features are key parameters describing the radio frequency properties of the base signal, including signal amplitude, frequency offset, and modulation period, the signal amplitude refers to the strength of the signal vibration, the frequency offset refers to the deviation value of the actual frequency of the signal from the reference frequency, and the modulation period refers to the time for the signal to complete one modulation change; the numerical vector is a data structure that stores the specific numerical values of the radio frequency features in an ordered array form, used to quantitatively represent the radio frequency features; the device identification information is information in the multi-channel signal data that is used to distinguish different drone remote controllers, wherein the device code field is a string or numerical code that uniquely identifies the remote controller; the binding data set is a data set formed by associating the device code field with the numerical vector of the corresponding radio frequency features, realizing one-to-one correspondence between the device and the features; the feature digitization process is the process of converting the numerical vector of the radio frequency features into a fixed-dimension numerical matrix, the numerical matrix is a numerical collection arranged by rows or columns, and the dimension is determined by the number of radio frequency features; the clustering division is an operation that uses a clustering algorithm to group similar numerical matrices into the same group, and the feature clustering cluster is a set formed after clustering division, containing similar numerical matrices, representing a feature group of a certain type of remote controller; the mapping relationship is the corresponding association between the feature clustering cluster and the model of the drone remote controller; the center vector is the arithmetic mean of all numerical vectors in the feature clustering cluster, used to represent the core features of the cluster; the feature template is a fixed-format data structure formed after standardizing the center vector, with a unified dimension and coding format; the threshold parameter is a judgment standard set based on the boundary distance between the feature clustering clusters, used to distinguish the feature differences of different models; the matching algorithm is an algorithm based on similarity calculation, used to judge the similarity between the to-be-identified signal features and the feature templates; the hardware feature library is a database integrating the feature templates, threshold parameters, and matching algorithms, supporting dynamic updating to adapt to new remote controller models.

[0097] In the embodiments of the present application, first, the signal amplitude, frequency offset, and modulation period, which are three core radio frequency features, are separated from the time-aligned base signal in step S1031, and then the specific numerical values of the three features are arranged in the order of "signal amplitude-frequency offset-modulation period" to form a numerical vector and store it. For example, in the A city drone supervision scene, for a certain road base signal captured earlier, the signal amplitude is 1.2V, the frequency offset is 50kHz, and the modulation period is 2ms, which are separated by a signal analysis tool. Then, the three numerical values are combined in order to generate a numerical vector [1.2, 50, 2] of the radio frequency features, and the vector is saved in the form of a document.

[0098] Secondly, in step S1032, a preset device code field is extracted from the multi-channel signal data. Each device code field is associated with the corresponding radio frequency feature value vector generated in step S1031 to ensure that each device code is bound to a unique value vector, thus forming a binding dataset of devices and features. For example, in the scenario of city A, the device code field corresponding to the basic signal is read as "D001" from the multi-channel signal data. "D001" is associated with the value vector [1.2,50,2] and recorded as "D001-[1.2,50,2]". At the same time, the same operation is performed on other basic signals, such as associating the device code "D002" with the value vector [1.16,49.6,2.04]. Finally, all association results are integrated to form a binding dataset containing dozens of sets of data.

[0099] Next, in step S1033, feature digitization processing is performed on all radio frequency feature numerical vectors in the bound dataset, converting each numerical vector into a fixed-dimensional numerical matrix with the same dimension as the number of radio frequency features (here, 3 rows and 1 column). Then, the K-means clustering algorithm is used to cluster all numerical matrices, grouping those with similar features into the same feature cluster. For example, in the scenario of city A, the numerical vectors... Convert to of A numerical matrix is ​​a vector of numbers. [2.04] converted to of The numerical matrices were analyzed. The Euclidean distance between the matrices was calculated using the K-means algorithm. It was found that the numerical matrices corresponding to 12 device codes, such as "D001" and "D002", had extremely high similarity. These matrices were classified into the feature cluster C01, while the matrices corresponding to other device codes were classified into different clusters such as C02 and C03.

[0100] Afterwards, by step S1034, first read the model information contained in each device code field from the binding data set, map each feature cluster to a unique drone remote control model according to the model information, form the mapping relationship between model and cluster; Then reposition the cluster associated with each model, calculate the arithmetic mean of all numerical vectors in the cluster to get the center vector, and finally standardize the center vector to generate a fixed format feature template. For example, in the A city market scenario, it is read that the model information corresponding to the device codes "D001", "D002" and the like is "M01", so the feature cluster C01 is mapped to the model "M01"; The arithmetic mean of the 12 numerical vectors in the C01 cluster is calculated, wherein the average value of the signal amplitude is (1.2+1.16+…+1.19) / 12=1.18V, the average value of the frequency offset is (50+49.6+…+50.2) / 12=49.8kHz, and the average value of the modulation period is (2+2.04+…+2.01) / 12=2.02ms, and the center vector [1.18, 49.8, 2.02] is obtained. The numerical accuracy of the vector is adjusted to two decimal places, and is packaged as a fixed format of "model M01-feature template [1.18, 49.8, 2.02]", and the generation of the "M01" model feature template is completed; The same operation is performed on "M02", "M03" and other models to generate corresponding feature templates.

[0101] Finally, by step S1035, based on the mapping relationship formed by step S1034, the boundary distance between different feature clusters is calculated, such as the boundary distance between C01 and C02 is 0.5, and the threshold value is set according to the boundary distance, such as setting the threshold value to 0.4, to ensure that similar features can be accurately matched; Design a matching algorithm based on cosine similarity to calculate the similarity between the to-be-identified signal feature and the feature template; Integrate all feature templates, threshold parameters and matching algorithms to build a dynamically updated hardware feature library, and if a new remote control model appears later, the steps S1031-S1034 can be repeated to add new feature templates. For example, in the A city market scenario, after calculating the boundary distance between C01 and C02, the threshold value is set to 0.4, and when the cosine similarity between the feature vector of the to-be-identified signal and the "M01" template is ≥0.4, it is determined to be the "M01" model; Integrate all model feature templates, threshold value 0.4 and cosine similarity algorithm into the hardware storage module to form a hardware feature library dedicated to A city drone supervision, and reserve a template adding interface to support dynamic updating.

[0102] In practical applications, in order to improve the identification accuracy of the unmanned aerial vehicle remote control signal, the air space management department of A city collects 15 types of unmanned aerial vehicle remote controllers on the market, denoted as M01 to M15, selects 20 different devices for each type of model to collect signals, extracts the radio frequency features of the basic signals of each device through the S1031 step and generates a numerical vector, and then associates the device code with the vector to form a binding data set through the S1032 step; then, the data set is digitized and clustered through the S1033 step, and 15 feature clustering clusters corresponding to M01 to M15 are obtained; the center vector of each cluster is calculated and a standardized feature template is generated through the S1034 step, and then a threshold parameter of 0.38 is set through the S1035 step, a cosine similarity matching algorithm is designed, and finally, a hardware feature library is integrated. When a new remote control signal is monitored subsequently, the radio frequency features thereof are only matched with the templates in the library, and the remote controller model can be quickly identified. If a new feature that is not matched is found, data can be collected and the feature library can be updated to ensure that the identification coverage is continuously improved.

[0103] In the overall scheme of the above step S103, the binding of the signal features and the specific remote controller is realized by extracting the radio frequency features and associating them with the device identifier; the radio frequency features can accurately correspond to the remote controller model through clustering analysis and mapping relationship establishment; and the hardware feature library containing the feature template, the threshold parameter and the matching algorithm is constructed to provide a standardized tool for subsequent signal identification. The whole process not only can effectively distinguish the remote control signals of unmanned aerial vehicles of different models and exclude interference signals such as civilian communication, but also can adapt to new remote controller models through dynamic updating function, which lays a foundation for accurate signal identification in unmanned aerial vehicle detection and reduces the error risk caused by signal misjudgment in the subsequent positioning process.

[0104] The embodiment of the present application provides a specific implementation flowchart of a comprehensive optoelectronic platform method for unmanned aerial vehicle detection, as shown in Figure 2 The embodiment of the present application provides a specific implementation flowchart of a comprehensive optoelectronic platform method for unmanned aerial vehicle detection, as shown in

[0105] S104, according to the identification result of the hardware feature library, combining the signal time difference in the basic signal of the plurality of nodes and the optical observation data in the space-time corresponding database, correcting and predicting the position of the operator at the next moment through the particle filtering technology, and then updating the predicted position by using the compensation result of the corrected multipath error, and iterating until convergence, realizing the comprehensive detection and positioning of the unmanned aerial vehicle combining the radio frequency detection and the optical observation.

[0106] Optionally, the step S104 can specifically include the following steps:

[0107] S1041, input the collected basic signal into the hardware feature library, output the identification result of the model of the remote controller of the unmanned aerial vehicle through a matching algorithm, calculate the time difference of arrival of the basic signal between different radio frequency receiving nodes, and combine the signal propagation speed to generate an initial position estimation area;

[0108] S1042, extract optical observation data corresponding to the identification result from the space-time correspondence database, the optical observation data including pixel coordinates of the unmanned aerial vehicle in a video frame, input the initial position estimation area and the optical observation data into a position prediction unit to generate an initial position prediction point of the operator;

[0109] S1043, simulate multi-path propagation paths based on a particle filtering model, and calculate the theoretical signal propagation delay of the initial position prediction point and each radio frequency receiving node;

[0110] Wherein, step S1043 can specifically include the following process: based on the initial position prediction point and the position of the radio frequency receiving node, a particle set representing different signal reflection paths is generated, each particle contains a path length parameter and a reflection point position parameter; calculate the matching degree of the path length parameter of each particle in the particle set and the initial position prediction point, and retain a preset number of particles with the highest matching degree to form a particle subset; randomly pair the particles in the particle subset to form a plurality of particle pairs, exchange the reflection point position parameters of each particle pair, generate new reflection paths based on the exchanged reflection point position parameters, and form a new particle set with the new reflection paths corresponding to the new particles; randomly select part of the particles in the new particle set, and randomly disturb the reflection point position parameters of the selected particles to form an optimized particle set, based on the optimized particle set, the path length of each propagation path from the initial position prediction point to each radio frequency receiving node is calculated; according to the electromagnetic wave propagation speed, the path length of each propagation path is converted into single path propagation time, and the single path propagation time is weighted and combined according to the particle weight to generate the theoretical signal propagation delay.

[0111] S1044, compare the theoretical signal propagation delay with the actual time difference of arrival to generate a multi-path error compensation amount, correct the initial position prediction point based on the multi-path error compensation amount, and output the updated prediction position;

[0112] S1045, repeat the error compensation and position updating operation until the deviation of the prediction position is less than the convergence threshold for two consecutive times, and obtain the comprehensive detection and positioning result of the unmanned aerial vehicle combining radio frequency detection and optical observation.

[0113] In the above steps, the identification result of the hardware feature library refers to the model information of the unmanned aerial vehicle remote controller obtained by analyzing the basic signal input into the hardware feature library through the matching algorithm in the library; the signal time difference refers to the time difference of the arrival of the same basic signal received by different radio frequency receiving nodes; the optical observation data is the unmanned aerial vehicle related data extracted from the space-time corresponding database and matched in time with the identification result, which contains the pixel coordinates of the unmanned aerial vehicle in the video frame, which can be mapped to the actual space position reference; the particle filtering technology is an algorithm for simulating a particle set to represent the signal propagation path and realize error correction and position prediction; the multipath error is the deviation between the actual receiving time and the straight-line propagation time caused by the reflection of the signal on objects such as buildings to form multiple propagation paths; the convergence threshold is a standard for judging whether the position prediction result is stable, and when the deviation of the predicted positions in two consecutive times is less than the threshold, the positioning result is considered reliable; the initial position estimation area is the approximate space range where the operator may exist, which is calculated based on the signal time difference and the signal propagation speed; the initial position prediction point is the operator position point preliminarily determined by combining the initial position estimation area and the optical observation data; the theoretical signal propagation delay is the propagation time of the signal from the predicted position to each radio frequency receiving node simulated based on the particle filtering model; and the multipath error compensation amount is a value for correcting the position deviation calculated by comparing the theoretical delay with the actual signal time difference.

[0114] In the embodiment of the application, first, the basic signal obtained by previous processing is input into the hardware feature library through step S1041. The matching algorithm in the library compares the radio frequency characteristics of the basic signal with the feature templates, and outputs the corresponding model identification result of the unmanned aerial vehicle remote controller. At the same time, the time difference of the arrival of the basic signal received by different radio frequency receiving nodes is calculated, and the initial position estimation area where the operator may exist is outlined by using the relationship of "distance = speed x time difference" in combination with the signal propagation speed of electromagnetic waves. For example, in the unmanned aerial vehicle monitoring scene of the core business district of City A, the basic signals of the east, north and south radio frequency receiving nodes are input into the hardware feature library, and the B model unmanned aerial vehicle remote controller corresponding to the signals is matched. It is calculated that the north node receives the signal 0.002 seconds later than the east node, and the south node receives the signal 0.0015 seconds later than the east node. In combination with the propagation speed of 3x10 8 meters / second of electromagnetic waves, it is calculated that the signal propagation distance difference between the north and east nodes is 600 meters, and the distance difference between the south and east nodes is 450 meters, and then a fan-shaped initial position estimation area with the three nodes as the apexes and meeting the distance difference condition is outlined.

[0115] Secondly, the optical observation data corresponding to the B-type remote controller recognition result time is extracted from the space-time correspondence database through step S1042 to find the pixel coordinates of the UAV in the high-definition video frame in this time period, and then the pixel coordinates are converted into the actual spatial position of the UAV according to the deployment position of the photoelectric sensor and the lens parameters; subsequently, the initial position estimation area is spatially correlated with the actual position of the UAV - because there is a signal transmission distance constraint between the UAV and the operator, the estimation range is reduced with the UAV position as the reference, and finally the initial position prediction point of the operator is generated. For example, in the above scenario, the video frame matching the signal time is extracted from the space-time correspondence database, the pixel coordinates of the UAV in the frame are (X: 850, Y: 420), and combined with the parameters that the photoelectric sensor is deployed on a 100-meter-high building and the lens focal length is 100 millimeters, the actual position of the UAV is calculated to be 1.2 kilometers east of the commercial district; combined with the characteristics that the UAV remote controller is usually controllable within 1 kilometer, the part of the initial position estimation area beyond the 1-kilometer range of the UAV is removed, and the initial position prediction point of the operator is determined to be the coordinate point near a certain office building east of the commercial district.

[0116] Then, through step S1043, the signal propagation path simulation from the initial position prediction point to each radio frequency receiving node is constructed based on the particle filtering model: first, a set containing 1000 particles is generated, each particle representing a possible propagation path, containing two parameters of path length (total distance of signal propagation) and reflection point position (coordinates of signal reflection by buildings); the matching degree of the path length of each particle with the straight-line distance from the initial position prediction point to the corresponding node is calculated, and the top 300 particles with the highest matching degree are retained to form a particle subset; the particles in the particle subset are randomly paired, the reflection point position parameters of each pair of particles are exchanged, and 300 new particles are generated to form a new particle set; 100 particles are randomly selected from the new particle set, and the reflection point position parameters thereof are slightly adjusted to form an optimized particle set; finally, based on the optimized particle set, the actual path length of each path from the initial position prediction point to the three radio frequency receiving nodes is calculated, and then the path length is converted into single-path propagation time according to the electromagnetic wave propagation speed; the single-path propagation time is weighted and combined according to the matching degree of each particle to obtain the theoretical signal propagation delay. For example, in the above scenario, the straight-line distance from the initial position prediction point to the east node is 800 meters, and in the simulated particles, the path length of a certain particle after reflection by a nearby high-rise building is 850 meters, and the matching degree is high and is retained; after pairing and exchanging the reflection points and slight disturbance, the path length of the optimized particles is concentrated between 840-860 meters, and the weighted calculation obtains the theoretical signal propagation delay from the prediction point to the east node as 0.0028 milliseconds, and the theoretical delays to the north and south nodes are 0.0032 milliseconds and 0.0030 milliseconds, respectively.

[0117] The theoretical signal propagation time delay obtained in step S1043 is compared with the actual time difference of arrival calculated in step S1041 in step S1044, and the difference between the two is found, which is the deviation caused by multipath error; according to the deviation size and direction, the multipath error compensation amount is calculated, and the coordinate of the initial position prediction point is adjusted by the compensation amount, for example, if the theoretical time delay is larger than the actual time difference, it means that the simulation path is too long, and the prediction point needs to be adjusted towards the receiving node, and finally the updated prediction position is output. For example, in the above scenario, after comparing the theoretical time delay with the actual time difference of arrival, it is found that the theoretical time delay of the east node is 0.0003 milliseconds larger than the actual time difference, and the north node is 0.0002 milliseconds larger, and the multipath error compensation amount is calculated as moving the initial position prediction point 90 meters towards the east node and 60 meters towards the north node, and the new prediction position coordinate is obtained after adjustment.

[0118] Finally, through step S1045, the theoretical time delay calculation in step S1043 and the error compensation and position updating operation in step S1044 are repeatedly executed: after each update, the deviation between the new prediction position and the last prediction position is calculated, if the deviation is still greater than the convergence threshold, the particle path is simulated again, the time delay is calculated, and the position is corrected; until the deviation between the two consecutive prediction positions is less than the convergence threshold, the iteration is stopped, and the position obtained at this time is the comprehensive unmanned aerial vehicle detection positioning result of fusing the radio frequency detection and optical observation results. For example, in the above scenario, the deviation between the first updated prediction position and the initial prediction point is 120 meters, which is greater than the convergence threshold of 50 meters, and needs to be iterated again; after the second iteration, the deviation between the new prediction position and the last position is 35 meters, which is less than the convergence threshold, the calculation is stopped, and it is determined that the operator is located in a certain office building on the 15th floor near the window on the east side of the business district, and the comprehensive positioning is completed.

[0119] In practical application, the UAV detection system deployed by the airspace management department of City A in the core business district starts the S104 process in response to a monitored illegal UAV signal: first, input the basic signals of the three radio frequency receiving nodes into the hardware feature library, match the signal to the remote controller of a certain model of C brand UAV, and calculate the time difference between the nodes to determine the initial position estimation area; then extract the optical data of the same time period from the space-time correspondence database, find that the UAV is located above the square on the southeast side of the business district in the video frame, and after converting the actual position, eliminate the part of the initial area that exceeds the 1-kilometer control range of the UAV, determine the initial position prediction point as near a certain mall on the north side of the square; then start the particle filtering model, generate 1000 particle simulation signal propagation paths, after screening, pairing, and disturbance optimization, calculate the theoretical signal propagation delay, and after comparing with the actual time difference, generate a compensation amount, adjust the initial prediction point 110 meters northeast; after repeating iteration for 3 times, the deviation of the predicted position is reduced to 40 meters for two consecutive times, which is less than the convergence threshold of 50 meters, and finally the operator is located in a certain shop on the 7th floor of the mall; the staff quickly arrived at the scene according to the positioning result, timely stopped the illegal UAV operation, and verified the practicability of the process.

[0120] In the overall scheme of the above step S104, by fusing the accurate recognition results of the hardware feature library, the multi-node signal time difference, and the optical observation data, the multi-path error in the city dense environment is effectively corrected by means of particle filtering technology, and the problem of single radio frequency positioning difficult to deal with non-line-of-sight and multi-reflection scene is solved; by repeating iteration until the position prediction converges, the accuracy of operator tracing is greatly improved, meeting the demand of city management for meter-level positioning; at the same time, the signal analysis of radio frequency detection and the spatial reference of optical observation are combined to form a complete positioning closed loop of "recognition-prediction-correction-convergence", avoiding the misjudgment and omission problem easily appeared by single technical means, providing reliable technical support for illegal UAV control in city dense environment.

[0121] The following is a complete embodiment for steps S101 to S104:

[0122] As Figure 3As shown, A city deploys a comprehensive optoelectronic platform for unmanned aerial vehicle (UAV) detection in a core business district. Five radio frequency (RF) receiving nodes and two high-definition optoelectronic sensors are first deployed around the business district. When a UAV violates the flight regulations, each RF node captures the original signal of the remote control signal of the UAV, generates multi-channel signal data with microsecond-level time stamps, and triggers the optoelectronic sensor to shoot a high-definition video stream of the suspicious airspace. The system matches the signal time stamps with the video frame time codes, and establishes a time-space correspondence database. Subsequently, the platform takes a high-precision atomic clock as a reference, realizes time synchronization of the five RF nodes through a precise protocol, and completes frequency conversion and demodulation based on the frequency and modulation characteristics of the signal, to obtain time-aligned basic signals. Then, the amplitude, frequency offset and other RF characteristics of the basic signals are extracted, and the device identification information in the signal is associated. After digital processing and classification analysis, a mapping relationship between these characteristics and the remote controllers of B and C type UAVs is established, and a hardware feature library containing feature templates and matching algorithms is constructed. Finally, combined with the recognition results of the feature library, the B type remote controller, the multi-node signal time difference and the UAV position data in the video are determined. The error caused by building multi-path reflection is corrected through particle filtering technology, and the operator's position is repeatedly updated until the prediction deviation meets the convergence condition for two consecutive times, to finally realize accurate positioning of the UAV and the operator.

[0123] The comprehensive optoelectronic platform method for UAV detection provided in the present application realizes all-around optimization of UAV detection through multi-source data fusion and accurate processing. The time-space association of signals and videos provides a basis for data verification, high-precision time synchronization guarantees the consistency of signal analysis, and the hardware feature library improves the accuracy of remote control signal recognition, effectively distinguishing interference signals. Combined with error correction and iterative optimization of particle filtering, the positioning deviation problem caused by non-line-of-sight and multi-path reflection in complex urban environments is solved. The advantages of RF detection and optical observation are combined to form a complete detection closed loop, significantly improving the reliability and accuracy of UAV and operator positioning, and providing strong technical support for urban airspace supervision.

[0124] Figure 4 A specific implementation structure diagram of a comprehensive optoelectronic platform system for UAV detection provided in an embodiment of the present application is shown in Figure 4 The system can include:

[0125] The acquisition module 41 is configured to acquire the original signal of the UAV remote control signal, generate multi-channel signal data with accurate time stamps based on the original signal, synchronously trigger the optoelectronic platform to capture a high-definition video stream of the suspicious airspace, and establish a time-space correspondence database of the multi-channel signal data and the high-definition video stream.

[0126] The conversion module 42 is configured to configure a high-precision atomic clock as a unified time reference, synchronize each radio frequency receiving node in time through a precision time synchronization protocol, perform frequency conversion and signal demodulation processing on the signals based on frequency characteristics and modulation characteristics in the multi-channel signal data, and obtain time-aligned basic signals.

[0127] The construction module 43 is configured to extract radio frequency characteristics in the basic signals, associate the radio frequency characteristics with device identification information in the multi-channel signal data, establish a mapping relationship between the radio frequency characteristics and the model of the unmanned aerial vehicle remote controller through feature digitization processing and classification analysis, and construct a hardware feature library containing a feature template, a threshold parameter, and a matching algorithm according to the mapping relationship.

[0128] The update module 44 is configured to correct and predict the position of the operator at the next moment through a particle filtering technique according to a recognition result of the hardware feature library, a signal time difference in the basic signals of the multiple nodes, and optical observation data in the space-time corresponding database, update the predicted position by using a corrected multipath error compensation result, and iteratively loop until convergence, so as to realize comprehensive unmanned aerial vehicle detection and positioning combining radio frequency detection and optical observation.

[0129] The comprehensive optoelectronic platform system for unmanned aerial vehicle detection of the embodiments of the present application is used to implement the comprehensive optoelectronic platform method for unmanned aerial vehicle detection described above, and therefore the specific embodiments of the comprehensive optoelectronic platform system for unmanned aerial vehicle detection can be seen from the foregoing embodiment part of the comprehensive optoelectronic platform method for unmanned aerial vehicle detection, and the specific embodiments can be referred to the description of the corresponding embodiment part, which will not be described herein again.

[0130] The present application further provides an electronic device, comprising a memory for storing a computer program and a processor for executing the computer program to implement the steps of the comprehensive optoelectronic platform method for unmanned aerial vehicle detection described above.

[0131] The present application further provides a computer readable storage medium having a computer program stored thereon, and the computer program is executed by a processor to implement the steps of the comprehensive optoelectronic platform method for unmanned aerial vehicle detection described above.

[0132] In an exemplary embodiment, the computer readable storage medium described above can include but is not limited to a U disk, a read-only memory, a random access memory, a mobile hard disk, a magnetic disk or an optical disk, and various media that can store computer programs.

[0133] The embodiment of the present application further provides a computer program product, the computer program product comprising a computer program, the computer program being executed by a processor to implement the steps in any of the above-mentioned comprehensive optoelectronic platform methods for detecting a UAV.

[0134] Those skilled in the art will further appreciate that the functions of the examples described herein, including any related steps of a method, can be implemented using electronic hardware, computer software, or any combination thereof. To clearly illustrate this interchangeability of hardware and software, various examples have been described herein generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends on the particular application and design constraints imposed on the overall system. Skilled artisans can implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.

[0135] The above provides a comprehensive optoelectronic platform method, system, electronic device and storage medium for detecting a UAV. The principles and implementation modes of the present application are described herein by applying specific examples, and the above description of the examples is only used to help understand the method of the present application and its core idea. It should be pointed out that, for those skilled in the art, some improvements and modifications can be made to the present application without departing from the principles of the present application, and these improvements and modifications also fall within the protection scope of the present application.

Claims

1. A comprehensive optoelectronic platform method for UAV detection, characterized in that, The application relates to a method for comprehensive detection and positioning of unmanned aerial vehicles (UAVs), which comprises the following steps: collecting original signals of UAV remote control signals, generating multi-channel signal data with accurate time stamps based on the original signals, synchronously triggering a photoelectric platform to capture high-definition video streams of suspicious airspace, and establishing a space-time corresponding database of the multi-channel signal data and the high-definition video streams; configuring a high-precision atomic clock as a unified time reference, achieving time synchronization of each radio frequency receiving node through a precise time synchronization protocol, performing frequency conversion and signal demodulation processing on the signals based on frequency characteristics and modulation characteristics in the multi-channel signal data to obtain time-aligned basic signals; extracting radio frequency characteristics in the basic signals, associating the radio frequency characteristics with device identification information in the multi-channel signal data, establishing a mapping relationship between the radio frequency characteristics and UAV remote controller models through feature digitization processing and classification analysis, and constructing a hardware feature library containing feature templates, threshold parameters and matching algorithms according to the mapping relationship; according to the recognition result of the hardware feature library, combining signal time differences in the basic signals of multiple nodes and optical observation data in the space-time corresponding database, correcting and predicting the next time position of the operator through particle filtering technology, updating the predicted position by using the compensation result of the corrected multipath error, and iteratively updating until convergence, so as to realize comprehensive detection and positioning of UAVs by combining radio frequency detection and optical observation; wherein the extraction of the radio frequency characteristics in the basic signals, the association of the radio frequency characteristics with the device identification information in the multi-channel signal data, the establishment of the mapping relationship between the radio frequency characteristics and the UAV remote controller models through the feature digitization processing and the classification analysis, and the construction of the hardware feature library containing the feature templates, the threshold parameters and the matching algorithms according to the mapping relationship comprise the following steps: separating signal amplitude, frequency offset and modulation period as radio frequency characteristics from the basic signals, and storing the radio frequency characteristics in the form of numerical vectors; associating a preset device code field in the multi-channel signal data with the numerical vectors of the radio frequency characteristics to form a binding data set of devices and characteristics; performing feature digitization processing on the binding data set, quantizing the numerical vectors of the radio frequency characteristics into digital matrices of fixed dimensions, clustering and dividing the digital matrices, and generating feature clustering clusters corresponding to different UAV remote controller models; establishing a mapping relationship between the feature clustering clusters and the UAV remote controller models, extracting center vectors of the feature clustering clusters corresponding to each model based on the mapping relationship, and generating feature templates; setting threshold parameters based on the mapping relationship and the clustering boundary distance, designing a matching algorithm based on similarity calculation, integrating the feature templates, the threshold parameters and the matching algorithm, and constructing a dynamically updated hardware feature library. ​ ​ ​ ​ ​ ​ 2. The method of claim 1, wherein, According to the identification result of the hardware feature library, combined with the signal time difference in the basic signal of multiple nodes and the optical observation data in the space-time corresponding database, the multipath error is corrected and the next time position of the operator is predicted through the particle filtering technology, and then the predicted position is updated by using the compensation result of the corrected multipath error, and the cycle iteration is performed until convergence, so as to realize the unmanned aerial vehicle comprehensive detection positioning combined with radio frequency detection and optical observation, including: The collected basic signal is input into the hardware feature library, and the identification result of the unmanned aerial vehicle remote controller model is output through the matching algorithm, the arrival time difference of the basic signal between different radio frequency receiving nodes is calculated, and the initial position estimation area is generated combined with the signal propagation speed; The optical observation data corresponding to the identification result is extracted from the space-time corresponding database, the optical observation data includes the pixel coordinates of the unmanned aerial vehicle in the video frame, the initial position estimation area and the optical observation data are input into the position prediction unit, and the initial position prediction point of the operator is generated; Based on the particle filtering model, the multipath propagation path is simulated, and the theoretical signal propagation time delay of the initial position prediction point and each radio frequency receiving node is calculated; The theoretical signal propagation time delay is compared with the actual arrival time difference, and the multipath error compensation amount is generated, the initial position prediction point is corrected based on the multipath error compensation amount, and the updated predicted position is output; The error compensation and position updating operation is repeatedly performed until the deviation of the predicted position is less than the convergence threshold value for two times in succession, and the unmanned aerial vehicle comprehensive detection positioning result combined with radio frequency detection and optical observation is obtained.

3. The method of claim 2, wherein, Based on the particle filtering model, the multipath propagation path is simulated, and the theoretical signal propagation time delay of the initial position prediction point and each radio frequency receiving node is calculated, including: Based on the initial position prediction point and the radio frequency receiving node position, a particle set representing different signal reflection paths is generated, each particle contains a path length parameter and a reflection point position parameter; The matching degree of the path length parameter of each particle in the particle set and the initial position prediction point is calculated, and a particle subset is formed by retaining a preset number of particles with the highest matching degree; The particles in the particle subset are randomly paired to form a plurality of particle pairs, the reflection point position parameters of each particle pair are exchanged, a new reflection path is generated based on the exchanged reflection point position parameters, and new particles corresponding to the new reflection path form a new particle set; In the new particle set, a part of particles are randomly selected, the reflection point position parameters of the selected particles are randomly disturbed, and an optimized particle set is formed, and based on the optimized particle set, the path length of each propagation path from the initial position prediction point to each radio frequency receiving node is calculated; According to the electromagnetic wave propagation speed, the path length of each propagation path is converted into single path propagation time, the single path propagation time is weighted and combined according to the particle weight, and the theoretical signal propagation time delay is generated.

4. The method of claim 1, wherein, The mapping relationship between the feature clustering cluster and the unmanned aerial vehicle remote controller model is established, and the center vector of the feature clustering cluster corresponding to each model is extracted based on the mapping relationship, and a feature template is generated, including: reading a device code field from the binding data set, mapping each of the feature clustering clusters to a unique drone remote controller model according to model information contained in the device code field, forming a mapping relationship between the model and the feature clustering cluster; based on the mapping relationship, positioning the feature clustering cluster associated with each drone remote controller model, calculating the arithmetic mean of all numerical vectors in the feature clustering cluster to obtain a center vector representing the characteristics of each model; standardizing the center vector of each drone remote controller model, adjusting the numerical precision of all vector elements to a unified standard, encapsulating the standardized center vector into a fixed format data structure, and generating a feature template with uniform dimensions and encoding format.

5. The method of claim 1, wherein, Configure a high-precision atomic clock as a unified time reference, synchronize each radio frequency receiving node through a precise time synchronization protocol, and process the signal through frequency conversion and signal demodulation based on the frequency characteristics and modulation characteristics in the multi-channel signal data to obtain a time-aligned basic signal, including: Deploy a high-stability atomic clock as a unified time reference, the atomic clock outputs a standard clock signal to all radio frequency receiving nodes, and distributes the standard clock signal through a wired transmission link to make the deviation of the local clock of each radio frequency receiving node from the unified time reference less than a microsecond threshold, to complete the time synchronization between nodes; Separate the frequency characteristic parameters and modulation waveform characteristic parameters from the multi-channel signal data, and use a frequency conversion unit to convert the original signals of different wave bands into unified intermediate frequency signals; Perform demodulation operation on the unified intermediate frequency signal, strip off the carrier component and extract the baseband signal to obtain a basic signal containing timing information; According to the time synchronization result between nodes, the timing of the basic signal is calibrated to eliminate the phase offset caused by the transmission delay between nodes, and a time-aligned basic signal is generated.

6. The method of claim 1, wherein, Collect the original signals of the drone remote control signals, generate multi-channel signal data with accurate time stamps based on the original signals, synchronously trigger the optoelectronic platform to capture the high-definition video stream of the suspicious airspace, and establish a space-time correspondence database of the multi-channel signal data and the high-definition video stream, including: Capture the original signals transmitted by the drone remote controller at multiple radio frequency receiving nodes, the original signals contain frequency bands and modulation waveforms, and record the initial time information of the original signals through a signal capture unit; Input the initial time information into a time marking unit, add a unique time identifier to each of the original signals in combination with a global clock reference, and generate multi-channel signal data with accurate time stamps; Based on the time identifier sequence in the multi-channel signal data, control the optoelectronic sensor to perform directional scanning on the suspicious airspace, and capture the high-definition video stream containing the drone image; Map the time identifier of the multi-channel signal data with the frame time code of the high-definition video stream to form a space-time correspondence database containing signal characteristic fields and video frame coordinates.

7. An integrated optoelectronic platform system for UAV detection, configured to perform the integrated optoelectronic platform method for UAV detection according to any one of claims 1-6. including: The collection module is used for collecting original signals of the unmanned aerial vehicle remote control signals, generating multi-path signal data with accurate time stamps based on the original signals, synchronously triggering the optoelectronic platform to capture high-definition video streams of suspicious airspace, and establishing a space-time corresponding database of the multi-path signal data and the high-definition video streams. The conversion module is used for configuring a high-precision atomic clock as a unified time reference, achieving time synchronization of each radio frequency receiving node through a precise time synchronization protocol, performing frequency conversion and signal demodulation processing on the signals based on frequency characteristics and modulation characteristics in the multi-path signal data, and obtaining time-aligned basic signals. The construction module is used for extracting radio frequency characteristics in the basic signals, associating the radio frequency characteristics with device identification information in the multi-path signal data, establishing a mapping relationship between the radio frequency characteristics and unmanned aerial vehicle remote controller models through feature digitization processing and classification analysis, and constructing a hardware feature library containing feature templates, threshold parameters and matching algorithms according to the mapping relationship. The update module is used for correcting and predicting the next time position of the operator through particle filtering technology according to the recognition result of the hardware feature library, combining signal time differences in the basic signals of multiple nodes and optical observation data in the space-time corresponding database, updating the predicted position by using the corrected multipath error compensation result, and iteratively updating until convergence, so as to realize comprehensive unmanned aerial vehicle detection positioning combining radio frequency detection and optical observation.

8. An electronic device, comprising: The computer readable storage medium stores a computer program, and the computer program is executed by the processor to realize the steps of the comprehensive optoelectronic platform method for unmanned aerial vehicle detection. The computer readable storage medium stores a computer program, and the computer program is executed by the processor to realize the steps of the comprehensive optoelectronic platform method for unmanned aerial vehicle detection. The computer readable storage medium stores a computer program, and the computer program is executed by the processor to realize the steps of the comprehensive optoelectronic platform method for unmanned aerial vehicle detection.

9. A computer-readable storage medium, characterized in that, ​

Images