A remote identification method of a UAV applied to a shoulder lamp

Abstract

This invention provides a remote identification method for unmanned aerial vehicles (UAVs) applied to shoulder lights. The method includes: acquiring a target frequency band signal; preprocessing the target frequency band signal to determine message data; performing feature analysis on the message data to determine the protocol type and its corresponding parsing algorithm; parsing the message data according to the parsing algorithm to determine key data; filtering the key data using a preset filtering algorithm to determine filtered data; and determining structured information based on the filtered data and the parsing algorithm, and determining target detection data based on the structured information. This invention solves the problem in the prior art of lacking a method that can be integrated into a shoulder-worn device, operate reliably in complex electromagnetic environments, and directly output high-quality structured information.

Description

Technical Field

[0001] This invention relates to the field of remote identification technology for drones using shoulder lights, and particularly to a remote identification method for drones using shoulder lights. Background Technology

[0002] With the rapid popularization of consumer and commercial drone technology, their flight activities in unauthorized areas (such as around critical infrastructure, large event venues, and confidential areas) pose increasingly serious challenges to public safety, personal privacy, and airspace management. Against this backdrop, the development of efficient and reliable drone detection technology has become an urgent need in the security field.

[0003] Currently, drone detection primarily relies on large, fixed systems such as radar, radio frequency detection stations, acoustic arrays, and photoelectric tracking systems. While these solutions offer long detection ranges and high accuracy, they are bulky, power-hungry, inflexible in deployment, and expensive, failing to meet the high mobility requirements of applications such as individual soldier patrols, temporary deployments, and rapid responses. Existing portable detection devices, in order to achieve small size and power consumption constraints, often employ simple energy detection or wideband reception, resulting in poor selectivity in complex electromagnetic environments and an inability to effectively capture specific drone communication signals. Furthermore, their poor filtering performance for communication signals makes it difficult to accurately determine the drone's ID, precise latitude and longitude, and crucial structured information directly usable for analysis, such as the pilot's location, thus limiting the value of the output results. Summary of the Invention

[0004] Based on this, the purpose of this invention is to provide a remote identification method for unmanned aerial vehicles (UAVs) applied to shoulder lights, aiming to solve the problem that there is a lack of a method in the prior art that can be integrated into a shoulder-worn device, operate reliably in a complex electromagnetic environment, and directly output high-quality structured information.

[0005] A remote identification method for unmanned aerial vehicles (UAVs) applied to shoulder lights according to an embodiment of the present invention includes: The target frequency band signal is acquired, the target frequency band signal is preprocessed to determine the message data, and the message data is subjected to feature analysis to determine the protocol type and its corresponding parsing algorithm; The message data is parsed according to the parsing algorithm to determine key data, and the key data is filtered through a preset filtering algorithm to determine filtered data. Based on the filtered data, structured information is determined using the parsing algorithm, and target detection data is determined based on the structured information.

[0006] In addition, the UAV remote identification method applied to shoulder lights according to the above embodiments of the present invention may also have the following additional technical features: Furthermore, the step of preprocessing the target frequency band signal to determine the message data includes: The key data is processed through a preset narrowband filter to obtain the first signal; The first signal is subjected to adaptive threshold detection to dynamically determine the signal detection threshold in order to distinguish the second signal from background noise; The second signal is averaged using a moving average algorithm to smooth out random noise and determine the third signal. The third signal is digitally filtered through a preset filter to determine the message data.

[0007] Furthermore, the detection threshold is dynamically calculated based on the real-time channel noise level, and the calculation formula for the detection threshold is as follows:

[0008] Among them, The noise mean is... The standard deviation of noise. This is an empirical coefficient; The formula for the moving average algorithm is:

[0009] in, The number of samples in the sliding window. For the sample index within the sliding window, For the first The signal values ​​of the original sampling points.

[0010] Furthermore, the step of determining the filtered data based on the key data using a preset filtering algorithm includes: Based on the timestamp information in the key data, the transmission time difference between adjacent messages is calculated to filter out messages whose time difference is not within the normal period range. Based on the geographic location information in the key data, the Euclidean distance between consecutive message locations is calculated, and cluster analysis is performed on the location points based on a preset clustering distance threshold to filter out spatial outlier messages whose distance from the main cluster exceeds the clustering distance threshold. Based on the signal strength information in the key data, statistical characteristics of the signal strength are calculated to filter packets whose signal strength deviates from the statistical characteristics by more than a preset deviation threshold. The hash value obtained by hashing the target data in the key data is used to identify duplicate messages received within a preset time window and filter them. The target data is at least one or a combination of two of the message sequence number and message content.

[0011] Furthermore, the step of determining target detection data based on the structured information includes: The movement trajectory of the UAV is generated based on the continuously acquired structured information; Predict the future short-term trajectory of the drone based on the motion trajectory; The threat level is determined based on the relationship between the movement trajectory, the future short-term trajectory, and the preset geofence.

[0012] Furthermore, the step of determining the target detection data based on the structured information includes: Acquire multiple target detection data from different shoulder lights, wherein the target detection data includes at least the UAV's movement trajectory, future short-term trajectory, and single-node threat level; The target detection data from different detection nodes that correspond to the same UAV identifier are fused to generate collaborative perception data.

[0013] Furthermore, the step of fusing target detection data from different detection nodes that correspond to the same UAV identifier to generate collaborative perception data includes: Based on the time difference and angle difference of arrival of signals from different nodes, and combined with the UAV's motion trajectory and the future short-term trajectory, a fused high-precision trajectory is generated through motion model constraints. In the process of solving, confidence weights are assigned to the trajectory data of each node. The confidence weights are used to weight and fuse the threat levels of each node to generate a collaborative threat alert that has been confirmed by the network.

[0014] Another objective of this invention is to provide a remote identification system for drones using shoulder lights, for implementing the aforementioned remote identification method for drones using shoulder lights, the system comprising: The data acquisition module is used to acquire target frequency band signals, preprocess the target frequency band signals to determine message data, and perform feature analysis on the message data to determine the protocol type and its corresponding parsing algorithm; The data filtering module is used to parse the message data according to the parsing algorithm to determine key data, and to filter the key data through a preset filtering algorithm to determine filtered data. The data parsing module is used to determine structured information based on the filtered data and the parsing algorithm, so as to determine target detection data based on the structured information.

[0015] Another objective of this invention is to provide a storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described method for remote identification of unmanned aerial vehicles using shoulder lights.

[0016] Another objective of this invention is to provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the above-described method for remote identification of a drone using a shoulder light.

[0017] This invention employs professional preprocessing of target frequency band signals to filter out valid message data. Combined with feature analysis, it achieves a precise one-to-one match between protocol type and parsing algorithm, ensuring the professionalism and adaptability of message data parsing from the outset. Furthermore, a preset filtering algorithm specifically filters key data after parsing, effectively eliminating false, interference, and duplicate data. Finally, structured information is accurately extracted from the high-quality filtered data to determine target detection data. This invention effectively addresses the problems of low drone detection accuracy and insufficient data validity caused by poor protocol compatibility, numerous false interference signals, and limited hardware space for shoulder lights in practical applications. Moreover, the entire detection process is autonomously completed based on signal feature analysis and a dedicated parsing algorithm, without relying on any additional manual configuration of parsing strategies or complex external signal processing hardware. This significantly reduces the deployment and operating costs of the equipment, achieving lightweight and portable shoulder-light-based drone detection. Therefore, this invention solves the problem of the lack of a method in the prior art that can be integrated into a shoulder-worn device, operate reliably in complex electromagnetic environments, and directly output high-quality structured information. Attached Figure Description

[0018] Figure 1 This is a flowchart of a remote identification method for drones applied to shoulder lights according to the first embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of the drone remote identification system applied to shoulder lights according to the second embodiment of the present invention; Detailed Implementation To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Several embodiments of the invention are illustrated in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0020] Example 1 Please see Figure 1The figure shows a remote identification method for a drone applied to a shoulder light according to the first embodiment of the present invention. The method is applied to a device including a vehicle domain controller and multiple subordinate safety sub-controllers integrated therein. The method specifically includes steps S01-S03.

[0021] S01, acquire the target frequency band signal, preprocess the target frequency band signal to determine the message data, and perform feature analysis on the message data to determine the protocol type and its corresponding parsing algorithm.

[0022] Specifically, the key data is processed through a preset narrowband filter to obtain a first signal; the first signal undergoes adaptive threshold detection to dynamically determine the signal detection threshold, thus distinguishing the second signal from background noise; the second signal is averaged using a moving average algorithm to smooth random noise and determine a third signal; the third signal is then digitally filtered through a preset filter to determine the message data. Layered noise reduction achieves multi-dimensional interference suppression, significantly reducing random noise and greatly improving the identification of effective signals. Furthermore, the adaptive threshold detection dynamically adjusts the signal detection threshold according to the channel environment, adapting to the signal detection needs of different electromagnetic environments. This avoids the problem of missing effective signals and falsely detecting interference signals when noise fluctuates due to fixed thresholds, resulting in high purity and no redundant interference in the preprocessed message data. This significantly reduces the computational load of subsequent protocol feature analysis and adapts to the hardware limitations of low power consumption and low computing power of shoulder lights.

[0023] Furthermore, the detection threshold is dynamically calculated based on the real-time channel noise level, and the calculation formula for the detection threshold is as follows:

[0024] Among them, The noise mean is... The standard deviation of noise. This is an empirical coefficient; The formula for the moving average algorithm is:

[0025] in, The number of samples in the sliding window. For the sample index within the sliding window, For the first The signal values ​​of the original sampling points.

[0026] As an example, and not a limitation, in some alternative embodiments, noise reduction is performed in a coordinated manner using hardware and software. Hardware-level noise reduction may involve integrating a proprietary narrowband filter and low-noise amplifier (LNA) into each RF front-end to suppress adjacent channel interference and improve small-signal reception sensitivity. The narrowband filter parameters are: center frequency 2.412 GHz, bandwidth 5 MHz, and suppression ratio >40 dB. The LNA parameters are: noise figure <2 dB and gain >15 dB. Software-level noise reduction involves reducing random noise through adaptive threshold detection and signal averaging algorithms. The sliding window size during signal averaging can be [missing information]. Furthermore, interference signals at specific frequencies are suppressed using a digital filter bank. Specifically, an FIR filter is used, with the following filter coefficients: Filtering formula: .

[0027] In practical implementation, it is necessary to perform protocol feature analysis on the captured messages to accurately identify the communication protocol category used by the drone's Remote ID. First, check the frame control field to determine if it is a Beacon frame. The Beacon frame control field value is 0x0080 (binary: 10000000000000000). The detection logic is as follows ( Secondly, verify whether the manufacturer information field matches the Remote ID standard (FA:0B:BC); the OUI byte sequence is [0xFA, 0x0B, 0xBC]; the verification condition is... Next, check the protocol version field to confirm the Remote ID message format. The version field is located in the 5th byte, and the standard value is protocol version 2. Finally, analyze whether the payload contains a standard Remote ID message structure. Specifically, first check the message type, where (byte >> 4) & 0x0F is a valid Remote ID message type (0-5, 0x0F). Then, verify the message length; the IE length should be greater than or equal to 5 bytes (3 bytes for OUI + 1 byte for type + 1 byte for data).

[0028] In addition, the system automatically matches the corresponding dedicated parsing algorithm. First, it performs a fast match based on a preset protocol feature library, where WiFi features are... Bluetooth features are Then, processing priorities are set according to the probability of the protocol occurrence. For example, the priority for WiFi 2.4GHz is [priority value missing]. Bluetooth LE priority is Finally, multiple protocol parsers run concurrently to improve throughput; specifically, a multi-threaded pool technique is used, with the number of threads dynamically adjusted based on the number of CPU cores, and the thread pool size = number of CPU cores × 2 + 1.

[0029] S02, the message data is parsed according to the parsing algorithm to determine key data, and the key data is filtered according to the preset filtering algorithm to determine filtered data.

[0030] Specifically, based on the timestamp information in the key data, the transmission time difference between adjacent messages is calculated to filter messages whose time difference is not within the normal period range; based on the geographical location information in the key data, the Euclidean distance between consecutive message locations is calculated, and cluster analysis is performed on the locations based on a preset clustering distance threshold to filter out spatial outlier messages whose distance from the main cluster exceeds the clustering distance threshold; based on the signal strength information in the key data, the statistical characteristics of the signal strength are calculated to filter messages whose signal strength deviates from the statistical characteristics by more than a preset deviation threshold; based on the hash value obtained by hashing the target data in the key data, duplicate messages received within a preset time window are identified and filtered, wherein the target data is at least one or a combination of two of the message sequence number and message content. By employing multiple filtering strategies, invalid data was accurately removed, significantly improving the filtering efficiency of false signals, interference signals, and duplicate messages, and significantly increasing the proportion of valid data. Furthermore, dynamic channel switching ensured that the shoulder light always detected on the channel with the optimal signal-to-noise ratio, continuously optimizing signal strength and stability. This avoided data distortion caused by the degradation of communication quality on a single channel. In addition, the high-quality filtered data significantly reduced the computational redundancy in subsequent structured information extraction, further adapting to the hardware computing power limitations of the shoulder light and improving the accuracy of the final information output.

[0031] In practical implementation, to effectively filter false and interference signals, firstly, verification can be performed based on the message sending periodicity. This involves calculating the time difference between adjacent messages and determining whether the time difference falls within the normal periodic range, thus filtering the signals. The normal periodic range is typically 0.1s to 10s. Secondly, signals from geographically close locations can be clustered for further filtering; the Euclidean distance calculation formula is as follows: In the formula, R is the radius of the region. , These are the latitude values ​​of the next and previous position points; , These are the longitude values ​​of the previous and next location points. Finally, power anomaly signals can be identified through power analysis. Specifically, the moving average of the RSSI is calculated using the following formula: ,in, This represents the original signal strength value at the i-th sampling point. This represents the number of signal strength sampling windows. Furthermore, to identify and filter duplicate messages, a message counter can be used to determine if a message is a duplicate; the formula for duplicate detection is... The process involves several steps: First, a time window is set to allow the re-reception of messages with the same sequence number within 3 seconds. Second, a hash value for the message content is generated using the SHA-256 algorithm. A hash table is used to store the hash values ​​of the N most recent messages, enabling rapid filtering and comparison of duplicate messages based on the hash values. Finally, messages from the same source are merged within a specified time window, which can be 1 second in size. The deduplication algorithm can be a two-dimensional index based on MAC address and sequence number. Additionally, to optimize signal strength, the SNR (Signal Notation) of each channel is calculated using the formula: The channel quality scoring formula is: ,in The channel stability factor is set to (0-1). Next, the channel with the best signal-to-noise ratio is selected for continuous monitoring. The channel switching condition can be that a switch is performed when the channel quality score drops by more than 20% for 5 seconds. The channel with the highest current quality score is selected.

[0032] S03, determine structured information based on the filtered data and the parsing algorithm, and determine target detection data based on the structured information.

[0033] In practical implementation, based on the identified protocol format, the filtered data stream is deeply parsed to extract structured information from the messages. For example, the complete 802.11 frame is first separated by field. The frame control field (2 bytes) contains information such as type and subtype; the duration field (2 bytes) is the frame duration; the address field (3×6 bytes) contains the destination address, source address, and BSSID; the sequence control field (2 bytes) contains the sequence number and fragment number; the marker field (2 bytes) contains the timestamp, interval, and function marker; the beacon interval field (2 bytes) is the beacon transmission interval; the capability field (2 bytes) contains device capability information; and the information element (variable length) contains the Remote ID data. Next, the Remote ID-related IE (Information Element) is located and extracted. The IE format is [Type (1 byte), Length (1 byte), Data (N bytes)]; the Vendor-Specific IE type is... Remote ID OUI is The check logic is as follows: Bit field resolution is performed according to the Remote ID standard to interpret the meaning of each field; message type resolution is performed as follows. The status bit is resolved as follows: Direction encoding is Based on the above analysis results, the required structured information is extracted in a targeted manner, such as drone type, ID type, ID string, drone location information, motion status information, operator-related information, and other structured information.

[0034] Specifically, the drone's trajectory is generated based on the continuously acquired structured information; the drone's future short-term trajectory is predicted based on the trajectory; and the threat level is determined based on the relationship between the trajectory, the future short-term trajectory, and a preset geofence. By deeply analyzing and extracting standardized, high-value structured information from high-quality filtered data, and combining this with drone trajectory generation, short-term trajectory prediction, and threat level determination, the raw data is transformed into target detection data that can be directly used for security assessment. This avoids the problem that traditional portable devices only output simple signals, lack structured information, and cannot perform threat assessment.

[0035] In practice, the location information of the drone at continuous intervals is serialized into a motion trajectory. Using a Kalman filter algorithm, with current speed and position as state variables, a short-term prediction of the drone's trajectory within the next 3 seconds is made. The real-time trajectory and the predicted trajectory are compared with a pre-defined electronic geofence. A threat behavior pattern library is defined. Based on the behavior matching results, minimum distance to sensitive areas, speed, and other factors, a quantified single-node threat level is calculated.

[0036] Furthermore, after determining the target detection data based on the structured information, the process includes: acquiring multiple target detection data from different shoulder lights, wherein the target detection data includes at least the UAV's movement trajectory, future short-term trajectory, and single-node threat level; and performing fusion processing on the target detection data from different detection nodes that correspond to the same UAV identifier to generate collaborative perception data.

[0037] Furthermore, the step of fusing target detection data from different detection nodes corresponding to the same UAV identifier to generate collaborative perception data includes: based on the signal arrival time difference and angle difference from different nodes, and combined with the UAV's motion trajectory and the future short-term trajectory, a high-precision fused trajectory is generated through motion model constraints. During the calculation process, confidence weights are assigned to the trajectory data of each node. These confidence weights are used to weight and fuse the threat levels reported by each node to generate a network-confirmed collaborative threat alert. By sharing target detection data among multiple nodes, centralized fusion of the same UAV data is achieved, solving the problems of low single-node detection accuracy, small coverage area, and poor threat judgment reliability, thus improving the overall accuracy and reliability of UAV detection. Moreover, the weighted fused threat level incorporates the judgment results of multiple nodes, avoiding false or missed threats due to local environmental factors at a single node, making threat judgment more objective and reliable.

[0038] In practical implementation, when multiple shoulder light devices are present, collaborative sensing is performed. Each shoulder light device broadcasts and shares its acquired target detection data through an inter-device ad hoc network. The collaborative processing unit receives data from multiple nodes. For data from the same UAV ID, a collaborative solution is performed using a time difference of arrival (TDOA) or angle of arrival (ADO) algorithm, and motion model constraints are applied in conjunction with the predicted trajectories of each node to generate a more accurate fused trajectory. During this process, a confidence weight is assigned to each data point based on indicators such as signal quality, historical reliability, and consistency of the fused trajectory. The threat levels of each reported node are weighted and fused according to these confidence weights to generate the final collaborative threat alert.

[0039] More specifically, firstly, the system synchronizes and correlates two types of key information reported by multiple shoulder light nodes for the same UAV: ​​one is the real-time signal observation, such as the time difference of the signal arriving at different nodes; the other is the short-term predicted trajectory of the UAV independently calculated by each node based on its own historical data. Next, the system constructs a unified mathematical optimization problem. The core objective function of this optimization problem aims to minimize two types of errors simultaneously: the first is the observation fitting error, which requires that the final calculated trajectory should geometrically match the actual signal observations reported by each node as closely as possible; the second is the motion model constraint error, which includes two parts: firstly, the calculated trajectory itself should be smooth and continuous, conforming to the general physical motion laws of UAV flight (e.g., uniform or uniformly accelerated motion); secondly, the calculated trajectory should be as consistent as possible with the predicted trajectories reported by each node, but not simply averaged; instead, a dynamic confidence weight is assigned to the predicted data of each node. This weight value is dynamically calculated based on the real-time quality of the node's data. For example, nodes with higher signal-to-noise ratios and more stable and reliable historical prediction results receive greater weights, meaning their predicted data has a higher influence in the final fusion process. The system then uses an iterative optimization algorithm to solve the constructed objective that incorporates multiple constraints. Finally, an optimal trajectory is output. This trajectory achieves a balance between multiple objectives, conforms geometrically to multi-node observations, is kinematically smooth and reasonable, and comprehensively incorporates reliable prediction information from each node. Furthermore, by organically embedding the locally predicted trajectories of each node as prior motion knowledge into the global optimization framework of collaborative localization, and by introducing a dynamic confidence weight mechanism, the system can automatically identify and prioritize the data from high-quality nodes. This achieves an adaptive and interference-resistant trajectory fusion, improving the positioning accuracy, smoothness, and robustness to erroneous or deceptive data in complex environments.

[0040] In summary, this invention employs professional preprocessing of target frequency band signals to filter out valid message data. Combined with feature analysis, it achieves a precise one-to-one match between protocol type and parsing algorithm, fundamentally ensuring the professionalism and adaptability of message data parsing. Furthermore, a pre-defined filtering algorithm specifically filters key data after parsing, effectively eliminating false, interference, and duplicate data. Finally, structured information is accurately extracted from the high-quality filtered data to determine target detection data. This effectively addresses the problems of low drone detection accuracy and insufficient data validity caused by poor protocol compatibility, numerous false interference signals, and limited hardware space for shoulder lights in practical applications. Moreover, the entire detection process is autonomously completed based on signal feature analysis and a dedicated parsing algorithm, without relying on any additional manual configuration of parsing strategies or complex external signal processing hardware. This significantly reduces the deployment and operating costs of the equipment, achieving lightweight and portable shoulder-light-based drone detection. Therefore, this invention solves the problem in the prior art of lacking a method that can be integrated into a shoulder-worn device, operate reliably in complex electromagnetic environments, and directly output high-quality structured information.

[0041] Example 2 Please see Figure 2 The diagram shown is a structural block diagram of a remote identification system for unmanned aerial vehicles (UAVs) applied to shoulder lights, as proposed in the second embodiment of the present invention. This remote identification system 200 for UAVs applied to shoulder lights includes: a feature data determination module 21, a feature data transmission module 22, and a security judgment and control module 23, wherein: The data acquisition module 21 is used to acquire the target frequency band signal, preprocess the target frequency band signal to determine the message data, and perform feature analysis on the message data to determine the protocol type and its corresponding parsing algorithm; Data filtering module 22 is used to parse the message data according to the parsing algorithm to determine key data, and to filter the key data through a preset filtering algorithm to determine filtered data. The data parsing module 23 is used to determine structured information based on the filtered data and the parsing algorithm, so as to determine target detection data based on the structured information.

[0042] Example 3 In another aspect, the present invention also proposes a shoulder light device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program 30, it implements the above-described method for remote identification of unmanned aerial vehicles applied to a shoulder light.

[0043] In some embodiments, the processor may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chip, used to run program code stored in memory or process data, such as executing access restriction programs.

[0044] The memory includes at least one type of readable storage medium, such as flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory can be an internal storage unit of an electronic device, such as the hard disk of the electronic device. In other embodiments, the memory can be an external storage device of the electronic device, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc. Furthermore, the memory can include both internal storage units and external storage devices of the electronic device. The memory can be used not only to store application software and various types of data of the electronic device, but also to temporarily store data that has been output or will be output.

[0045] This invention also proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned method for remote identification of unmanned aerial vehicles (UAVs) using shoulder lights.

[0046] Those skilled in the art will understand that the logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can mean any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0047] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0048] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0049] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0050] The above embodiments merely illustrate several implementation methods of the present invention, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this patent should be determined by the appended claims.

Claims

1. A remote identification method for unmanned aerial vehicles (UAVs) applied to shoulder lights, characterized in that, The method includes: The target frequency band signal is acquired, the target frequency band signal is preprocessed to determine the message data, and the message data is subjected to feature analysis to determine the protocol type and its corresponding parsing algorithm; The message data is parsed according to the parsing algorithm to determine key data, and the key data is filtered through a preset filtering algorithm to determine filtered data. Based on the filtered data, structured information is determined using the parsing algorithm, and target detection data is determined based on the structured information.

2. The method for remote identification of unmanned aerial vehicles (UAVs) applied to shoulder lights according to claim 1, characterized in that, The steps for preprocessing the target frequency band signal to determine the message data include: The key data is processed through a preset narrowband filter to obtain the first signal; The first signal is subjected to adaptive threshold detection to dynamically determine the signal detection threshold in order to distinguish the second signal from background noise; The second signal is averaged using a moving average algorithm to smooth out random noise and determine the third signal. The third signal is digitally filtered through a preset filter to determine the message data.

3. The method for remote identification of unmanned aerial vehicles (UAVs) applied to shoulder lights according to claim 1, characterized in that, The detection threshold is dynamically calculated based on the real-time channel noise level, and the calculation formula for the detection threshold is as follows: Among them, The noise mean is... The standard deviation of noise. This is an empirical coefficient; The formula for the moving average algorithm is: in, The number of samples in the sliding window. For the sample index within the sliding window, For the first The signal values ​​of the original sampling points.

4. The method for remote identification of unmanned aerial vehicles (UAVs) applied to shoulder lights according to claim 2, characterized in that, The steps for determining the filtered data based on the key data using a preset filtering algorithm include: Based on the timestamp information in the key data, the transmission time difference between adjacent messages is calculated to filter out messages whose time difference is not within the normal period range. Based on the geographic location information in the key data, the Euclidean distance between consecutive message locations is calculated, and cluster analysis is performed on the location points based on a preset clustering distance threshold to filter out spatial outlier messages whose distance from the main cluster exceeds the clustering distance threshold. Based on the signal strength information in the key data, statistical characteristics of the signal strength are calculated to filter packets whose signal strength deviates from the statistical characteristics by more than a preset deviation threshold. The hash value obtained by hashing the target data in the key data is used to identify duplicate messages received within a preset time window and filter them. The target data is at least one or a combination of two of the message sequence number and message content.

5. The method for remote identification of unmanned aerial vehicles (UAVs) applied to shoulder lights according to claim 1, characterized in that, The steps for determining target detection data based on the structured information include: The movement trajectory of the UAV is generated based on the continuously acquired structured information; Predict the future short-term trajectory of the drone based on the motion trajectory; The threat level is determined based on the relationship between the movement trajectory, the future short-term trajectory, and the preset geofence.

6. The method for remote identification of unmanned aerial vehicles (UAVs) applied to shoulder lights according to claim 1, characterized in that, The step of determining the target detection data based on the structured information includes: Acquire multiple target detection data from different shoulder lights, wherein the target detection data includes at least the UAV's movement trajectory, future short-term trajectory, and single-node threat level; The target detection data from different detection nodes that correspond to the same UAV identifier are fused to generate collaborative perception data.

7. The method for remote identification of unmanned aerial vehicles (UAVs) applied to shoulder lights according to claim 1, characterized in that, The steps of fusing target detection data from different detection nodes that correspond to the same UAV identifier to generate collaborative sensing data include: Based on the time difference and angle difference of arrival of signals from different nodes, and combined with the UAV's motion trajectory and the future short-term trajectory, a fused high-precision trajectory is generated through motion model constraints. In the process of solving, confidence weights are assigned to the trajectory data of each node. The confidence weights are used to weight and fuse the threat levels of each node to generate a collaborative threat alert that has been confirmed by the network.

8. A remote identification system for unmanned aerial vehicles (UAVs) applied to shoulder lights, characterized in that, The system for implementing the UAV remote identification method applied to shoulder lights as described in any one of claims 1 to 7, the system comprising: The data acquisition module is used to acquire target frequency band signals, preprocess the target frequency band signals to determine message data, and perform feature analysis on the message data to determine the protocol type and its corresponding parsing algorithm; The data filtering module is used to parse the message data according to the parsing algorithm to determine key data, and to filter the key data through a preset filtering algorithm to determine filtered data. The data parsing module is used to determine structured information based on the filtered data and the parsing algorithm, so as to determine target detection data based on the structured information.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps of the drone remote identification method applied to shoulder lights as described in any one of claims 1 to 7.

10. A shoulder light device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the remote identification method for a drone applied to a shoulder light as described in any one of claims 1-7.

Images