A distributed low-altitude target passive collaborative positioning and threat assessment system

Abstract

The present invention discloses a distributed low-altitude target passive collaborative positioning and threat assessment system, which relates to the field of target detection technology. The system comprises a spatial preliminary positioning subsystem, a deception feature identification subsystem, a swarm decomposition subsystem and a correction subsystem. The system captures all arrival signals in the low-altitude airspace and locks the three-dimensional spatial coordinates of the corresponding target aircraft, analyzes the degree of jump in frequency deviation within a short time window and identifies the power jump phenomenon of the corresponding target aircraft, so as to dynamically identify the credibility of the deception behavior of the corresponding target aircraft. The system also determines a number of group labels, analyzes the degree of operational difference of the target aircraft in the group and dynamically and in real time determines whether the corresponding group is in an organized, highly coordinated offensive convergence breakthrough stage, identifies high-risk key defense groups, combines a deception discrimination unit to re-judge the high-risk key defense groups, and performs dynamic scheduling operations in combination with the degree of attack risk.

Description

Technical Field

[0001] The present invention relates to the field of target detection technology, and in particular to a distributed low-altitude target passive collaborative positioning and threat assessment system. Background Art

[0002] The present invention relates to the technical field of target detection and threat assessment, and belongs to a distributed low-altitude target monitoring and dynamic threat management system. In particular, it relates to a distributed low-altitude target passive collaborative positioning and threat assessment system applied to low-altitude airspace, having passive detection capabilities, and oriented to complex swarm attack scenarios. The system can perform spatial positioning, deception identification, group collaborative behavior analysis, and dynamic threat level quantification output on new air threat targets such as high-speed small targets, drone clusters, and swarm-type penetration formations in a variety of complex low-altitude security defense environments, thereby realizing real-time dynamic defense resource adaptive scheduling and control.

[0003] Current low-altitude target detection and threat identification technologies, particularly for multi-target coordinated intrusion scenarios like drone swarms and micro-aircraft formations, primarily rely on single-station radars, traditional active scanning radars, or infrared imaging. However, these approaches have significant shortcomings. For example, in actual combat scenarios, attackers often disrupt the detection system's discrimination capabilities by deploying decoys (e.g., reflective clouds, false target simulators, and electronic frequency spoofing transmitters). This can easily mislead the detection system in identifying target numbers, determining movement patterns, and assessing threat levels. Summary of the Invention

[0004] In view of the above-mentioned problems existing in the prior art, the present application provides a distributed low-altitude target passive collaborative positioning and threat assessment system.

[0005] The present invention provides a distributed low-altitude target passive collaborative positioning and threat assessment system, comprising:

[0006] The space preliminary positioning subsystem deploys multiple passive detection nodes outside the target defense area to capture all incoming signals in the low-altitude airspace and lock the three-dimensional spatial coordinates of the corresponding target aircraft;

[0007] The deception feature analysis subsystem, which includes a frequency disturbance analysis unit, a radio frequency modulation analysis unit, and a deception discrimination unit, dynamically identifies the credibility of the target aircraft's deception behavior by analyzing the severity of frequency deviation jumps within a short time window and identifying power jumps in the target aircraft.

[0008] The swarm decomposition subsystem, including a clustering unit, a consistency analysis unit, a convergence analysis unit, and a coordinated penetration risk identification unit, determines several swarm labels, analyzes the operational differences of target aircraft within the swarm, and dynamically determines in real time whether the swarm is in an organized, highly coordinated offensive convergence penetration phase, thereby identifying high-risk, key surveillance swarms.

[0009] The correction subsystem, combined with the deception identification unit, re-evaluates the high-risk key surveillance groups and conducts dynamic scheduling operations based on the degree of attack risk.

[0010] Optionally, the spatial preliminary positioning subsystem includes an acquisition unit and a positioning unit;

[0011] The acquisition unit deploys multiple passive detection nodes outside the target defense zone to capture all arrival signals in the low-altitude airspace. Each detection node has a built-in clock module and achieves time synchronization through the Beidou timing system. The arrival signal includes the target reflection signal, arrival time difference, frequency drift and signal strength.

[0012] The positioning unit randomly selects a reference detection node and uses the time difference method to calculate the relative position of each signal arrival, obtains the time difference equation, and solves it to determine the preliminary spatial position and timestamp of the target aircraft. It also uses the Doppler effect generated by the high-speed motion of the target to detect the frequency offset value received by each station to infer the radial velocity of the target aircraft. Based on the radial velocity and the timestamp of the preliminary spatial position of the target aircraft, it determines the offset distance vector of the target aircraft in the direction of its velocity within the entire acquisition time window. Combined with the preliminary spatial position, the three-dimensional spatial coordinates of the corresponding target aircraft are locked. Specifically, ,in, is the three-dimensional space coordinate, is the initial spatial position, is the offset distance vector;

[0013] Among them, the specific expression of the time difference equation is: ,in, For the Station propagation path distance, is the reference detection node propagation path distance, is the signal propagation speed, For the The arrival time difference of the signal received by the station.

[0014] Optionally, the deception feature analysis subsystem includes a frequency disturbance analysis unit, a radio frequency modulation analysis unit, and a deception discrimination unit;

[0015] The frequency disturbance analysis unit extracts the frequency offset values ​​of multiple consecutive frames for each located target aircraft to analyze the severity of the frequency deviation jump within a short time window. It also calculates the differential frequency drift amplitude of adjacent frames to obtain the root mean square fluctuation rate within the entire sampling window.

[0016] During the continuous sampling process, the RF modulation analysis unit obtains a sequence curve of the instantaneous signal power measured in the time domain changing with time for each target aircraft, and makes a discrete first-order derivative in the sampling sequence to identify the amplitude of power increase or decrease between frames, and continuously integrates the modulus square value of the first-order derivative to identify the power jump phenomenon of the corresponding target aircraft and construct the modulation perturbation flicker energy index.

[0017] Optionally, the deception identification unit performs a standardized ratio operation on the root mean square fluctuation rate in the entire sampling window and the modulation perturbation flicker energy index of the corresponding target aircraft, and forms a deception credibility coefficient after weighted summation. If the deception credibility coefficient exceeds a preset discrimination threshold, the corresponding target aircraft is marked as a decoy, otherwise no marking is performed.

[0018] Optionally, the swarm decomposition subsystem includes a clustering unit, a consistency analysis unit, a convergence analysis unit, and a collaborative penetration risk identification unit;

[0019] The clustering unit extracts the three-dimensional spatial coordinates and velocity vector of each target aircraft at the current sampling moment to obtain a joint feature vector, analyzes the similarity distance between any two target aircraft in the joint feature space, and determines several group labels based on the similarity distance values;

[0020] a consistency analysis unit, determining the average velocity vector within each group and the modulus of the speed deviation of each target aircraft within the group, and analyzing the degree of operational differences of the target aircraft within the group through the average velocity vector within each group and the modulus of the speed deviation of each target aircraft within the group to obtain an overall speed consistency coefficient;

[0021] The convergence analysis unit calculates the mean of all pairwise Euclidean distances between target aircraft in each group at each moment, and records it as the group average density value at each moment. Within the entire sampling window, the first-order difference method is used to approximate the group density convergence rate. The group density convergence rate is used to dynamically and in real time determine whether the corresponding group is in an organized, highly coordinated offensive convergence penetration stage.

[0022] Optionally, the collaborative penetration risk identification unit performs standard inversion processing on the overall speed consistency coefficient, converts the group density convergence rate into an absolute value, and then calculates the swarm attack risk factor by suppressing low-threat false alarms. If the swarm attack risk factor exceeds the preset risk threshold, the corresponding group will be marked as a high-risk key defense group.

[0023] Optionally, the correction subsystem includes a purification unit and a scheduling unit;

[0024] The purification unit extracts the high-risk key surveillance group containing decoys and removes the decoys to form a purified group. After removing the decoy components, the updated overall speed consistency coefficient and group density convergence rate of the purified group are recalculated, and the swarm attack risk factor is updated in real time. It is re-evaluated whether it is a high-risk key surveillance group. If so, it is included in the surveillance group set.

[0025] Optionally, the scheduling unit sorts the swarm attack risk factors corresponding to each high-risk key monitoring group in the monitoring group set in descending order to form a dynamic defense resource scheduling priority list, and divides them into three categories according to the dynamic defense resource scheduling priority list, namely high priority group, medium priority group and low priority group, and drives the system's multi-level dynamic defense actions according to the three categories.

[0026] Beneficial effects of the present invention:

[0027] 1. By organically integrating the modules of preliminary spatial positioning, decoy feature analysis, swarm decomposition, and corrected dynamic scheduling, the system achieves full-link intelligent perception, discrimination, and dynamic defense control in complex low-altitude multi-target environments. On the one hand, the system deploys multiple passive detection nodes outside the target defense zone, relying on Beidou timing technology to achieve high-precision time synchronization. This allows multi-station detection data to be combined in real time for spatial positioning based on the time difference method and Doppler frequency deviation velocity inversion analysis, accurately obtaining the three-dimensional spatial coordinates and motion state of the target aircraft. On the other hand, through dynamic feature extraction and group evolution behavior modeling, the system achieves dynamic identification and risk classification management of complex air threats such as high-speed small targets, drone swarms, and coordinated formations, providing reliable technical support and real-time defense strategy basis for the defense system when facing new swarm penetration combat threats.

[0028] 2. The decoy feature identification subsystem extracts the severity of target frequency deviation jumps in the frequency perturbation analysis unit and captures high-frequency frequency drift oscillation caused by unstable control of electronic decoy equipment by calculating the root mean square fluctuation rate within a short time window. In the RF modulation analysis unit, it monitors signal power flicker changes in real time, calculates the modulation perturbation flicker energy index, effectively identifies modulation anomalies caused by power supply ripple and circuit noise, and accurately eliminates decoys through the fusion of decoy credibility coefficients. The swarm decomposition subsystem implements dynamic swarm labeling through a joint feature space clustering algorithm. In the consistency analysis unit, it calculates the consistency coefficient of speed deviation within the swarm. In the convergence analysis unit, it continuously monitors the spatial density convergence rate, capturing in real time whether the swarm has entered a highly coordinated convergent penetration formation, effectively warning of the evolution of swarm attack organization, and overcoming the technical shortcomings of traditional systems that easily misjudge false swarm artifacts based on static snapshots.

[0029] 3. After removing deceptive elements at the swarm level in real time, the purification unit dynamically reconstructs the characteristics of the purified swarm, updating the overall speed consistency coefficient and density convergence rate in real time to form an updated swarm attack risk factor, ensuring that the risk assessment model is based on pure and valid target characteristics. The scheduling unit forms a dynamic defense resource scheduling priority list based on real-time risk ranking, dynamically classifying high-risk key defense swarms into three priority levels: high, medium, and low. This drives multi-level dynamic defense actions, effectively achieving the precise delivery and efficient utilization of limited defense resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the technical solutions in the present application or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application.

[0031] Figure 1 This is a subsystem diagram of the present invention. DETAILED DESCRIPTION

[0032] To make the objectives, technical solutions, and advantages of this application more clear, the technical solutions of this application will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only part of the embodiments of this application, not all of them. Based on the embodiments of this application, all other embodiments obtained by ordinary technicians in this field without making any creative efforts are within the scope of protection of this application.

[0033] like Figure 1 As shown, the implementation of the present invention proposes a distributed low-altitude target passive collaborative positioning and threat assessment system, including:

[0034] The space preliminary positioning subsystem deploys multiple passive detection nodes outside the target defense area to capture all incoming signals in the low-altitude airspace and lock the three-dimensional spatial coordinates of the corresponding target aircraft;

[0035] Furthermore, the spatial preliminary positioning subsystem includes an acquisition unit and a positioning unit;

[0036] The acquisition unit deploys multiple passive detection nodes outside the target defense zone, relying on fixed or mobile support frames, to capture all arrival signals in the low-altitude airspace. Each detection node has a built-in clock module, which is synchronized through the Beidou timing system to ensure that the data collected by multiple nodes has a unified time reference standard, laying the foundation for subsequent time difference-based spatial positioning. The arrival signal includes the target reflection signal, arrival time difference, frequency drift, and signal strength.

[0037] The sources of target reflection signals include existing broadcast signals (such as civil radio and television, navigation signals) reflected or scattered by the surface of the flying target, communication link signals radiated by the flying target itself (such as drone control links, image return links), secondary reflection signals generated by the flying target after being irradiated by existing radar waves in the surrounding area, and nonlinear modulation scattering generated by the flying target (such as rotor blade modulation echo), etc.; in short, the target reflection signal is the electromagnetic wave reflection echo generated by the target due to its existence, movement, posture change, etc.

[0038] The positioning unit randomly selects a reference detection node and uses the time difference of arrival method (TDOA) to calculate the relative position of each signal, obtains the time difference equation, and solves it to determine the preliminary spatial position of the target aircraft and its timestamp;

[0039] The relative arrival positions of the signals refer to the multiple versions of the reflected signal from the same target that are received simultaneously by multiple detection nodes distributed at different locations (i.e., multi-station received signals).

[0040] The essence of the time difference of arrival (TDOA) is to calculate the spatial geometric position of the target relative to the detection nodes based on the time difference between the arrival times of the same signal received by different detection nodes. The specific calculation steps are as follows:

[0041] S1: Select a node as a reference station and calculate the arrival time difference between each other node and the reference station;

[0042] S2: Establish the time difference equations: Assume that the unknown coordinates of the target in three-dimensional space are (x, y, z), the detection stations know their respective coordinates are (x', y', z'), and the signal propagation speed is (close to the speed of light), according to the propagation path distance formula: ,in, is the propagation path distance;

[0043] Among them, the time difference equation is: ,in, For the target aircraft and The propagation path distance between stations, is the spatial propagation distance between the target aircraft and the reference detection station, is the signal propagation speed (close to the speed of light), For the The arrival time difference of the signal received by the station, To convert the spatial geometric distance difference into time difference; the logic of the time difference equation is distance difference ÷ propagation speed = time difference; so the time difference measured by multiple stations is actually to deduce the spatial distance difference between each detection station and the target.

[0044] S3: Solve the system of equations: At least 4 detection stations (3D positioning) are required to form the system of equations, and use nonlinear optimization methods such as least squares method to solve the preliminary spatial position (x, y, z) of the target aircraft.

[0045] During high-speed ground grazing, the target aircraft's position shifts within the time slices when multiple sampling nodes record signals. This drift in motion causes a systematic offset error in the spatial position calculated by the time difference method. Therefore, the initial spatial position of the target aircraft needs to be corrected.

[0046] By using the Doppler effect caused by the high-speed movement of the target and detecting the frequency offset value received by each station, the radial velocity of the target aircraft can be inferred. Specifically: ,Right now ,in, is the frequency offset value, For the target aircraft at the detection station The instantaneous radial velocity component in the direction (i.e. the radial velocity of the target aircraft inferred from the inverse), is the transmission carrier frequency of the signal;

[0047] Multiple detection stations obtained , combined with the spatial geometric relationship, the three-dimensional motion velocity vector of the target as a whole can be inferred, which is usually achieved by solving the velocity vector decomposition equation group: ,in, is the three-dimensional motion velocity vector, For the The unit vector of the direction of the station pointing to the current position of the target aircraft;

[0048] Since the target aircraft is assumed to be stationary when solving the time difference method, but the target aircraft is actually in constant motion, the target aircraft has shifted a small distance in the direction of its velocity (i.e., the offset distance vector) during the entire acquisition time window;

[0049] Based on the radial velocity and the timestamp of the initial spatial position of the target aircraft, the offset distance vector of the target aircraft in the direction of its velocity within the entire acquisition time window is determined, specifically: ,in, is the number of continuous sampling frames (i.e. the entire acquisition time window), is the offset distance vector;

[0050] The entire acquisition time window is the physical sampling window required by the system to complete a complete time difference method position solution, that is, the entire time period from signal transmission, to each station receiving the signal, recording data, and then the system synchronously aggregating the sampling data of each station for time difference method TDOA calculation.

[0051] Combined with the preliminary spatial position, the three-dimensional spatial coordinates of the corresponding target aircraft are locked, specifically: ,in, is the three-dimensional space coordinate, is the initial spatial position;

[0052] Spatial coordinates refer to the updated spatial position;

[0053] In this embodiment of the present invention, the acquisition unit first deploys multiple passive detection nodes around the perimeter of the target defense zone, enabling continuous signal coverage across a wide airspace. Each detection node is equipped with an independent built-in clock module, synchronized with the BeiDou timing system for high-precision synchronization. For example, if there are five passive detection nodes distributed at different locations around the defense zone, the BeiDou timing system can synchronize the time of each node with microsecond accuracy, ensuring that the signal data received simultaneously by multiple stations has a unified time reference. This time synchronization is the foundation for subsequent spatial positioning calculations using the Time Difference of Earth (TDOA).

[0054] Time synchronization ensures that the system has comparability and relative time reference for data on the same signal event between multiple stations, avoiding spatial positioning offsets caused by sampling time errors.

[0055] Secondly, during the signal acquisition process, the system can fully capture all incoming signals, including real target reflections, environmental background noise, and possible decoy interference signals. For example, when radar waves reflected from a target aircraft are received by multiple detection nodes, each node can simultaneously record the signal's arrival time, frequency drift (Doppler shift), and signal strength (power).

[0056] Frequency drift refers to the difference between the received signal frequency and the transmitted signal frequency caused by the Doppler effect generated by the high-speed movement of the target, which can reflect the target's movement speed in the radial direction;

[0057] The signal strength helps in analyzing the power stability characteristics in the subsequent deception feature identification.

[0058] Thirdly, the positioning unit uses the time difference method (TDOA) to perform spatial positioning calculations. After multiple synchronous detection nodes receive the same target signal, the system uses a randomly selected node as the reference node and establishes a time difference equation using the time difference between the signals received by different nodes. For example, station A is set as the reference node, and the arrival times of the signals received by stations B, C, and D are tB, tC, and tD, respectively. Based on the propagation speed (approximately the speed of light c), the distance difference between the respective propagation paths is formed, and the time difference equation is constructed. By solving this set of equations, the preliminary spatial position of the target aircraft and the corresponding timestamp are obtained.

[0059] The time difference equation converts the time difference information collected synchronously by multiple stations into a spatial geometry solution model, thereby locating the target's position in three-dimensional space. Furthermore, the positioning unit introduces Doppler frequency deviation analysis logic. This uses the frequency deviation value extracted by each node upon receiving the target signal, combined with the geometric relationship between the target and each station, to infer the radial velocity component of the target relative to each detection node. For example, if a target moves north at a speed of 100m / s, the frequency deviation values ​​measured by detection nodes at different locations will reflect differences related to the target's direction of motion. By integrating the frequency deviation observation data from multiple stations, the target's overall motion velocity vector can be accurately calculated.

[0060] Radial velocity represents the target's velocity component along the measurement direction of each station and is a key input to the dynamic spatial correction process. After obtaining the initial spatial position and radial velocity, the system further considers motion offset correction within the acquisition time window. For example, if the sampling window spans 20 milliseconds and the target aircraft is flying at 100 m / s, the target will have moved approximately 2 meters in the direction of velocity within this time window. By calculating the motion offset distance vector and combining it with the initial positioning results, the system corrects and outputs the 3D spatial coordinates with higher accuracy.

[0061] The offset distance vector is used to describe the actual spatial displacement of the target during the sampling integration period and is used to dynamically correct the spatial positioning accuracy.

[0062] In summary, through the overall design and synergy of the above-mentioned space preliminary positioning subsystem, a complete technical closed loop is achieved from raw signal acquisition, multi-station time synchronization, time difference spatial solution, Doppler frequency deviation correction to three-dimensional coordinate output. It can effectively improve the system's spatial detection accuracy for high-speed small aerial targets in complex low-altitude environments, and provide accurate basic data support for subsequent deception analysis and threat identification.

[0063] The deception feature analysis subsystem, which includes a frequency disturbance analysis unit, a radio frequency modulation analysis unit, and a deception discrimination unit, dynamically identifies the credibility of the target aircraft's deception behavior by analyzing the severity of frequency deviation jumps within a short time window and identifying power jumps in the target aircraft.

[0064] Furthermore, the deception feature analysis subsystem includes a frequency disturbance analysis unit, a radio frequency modulation analysis unit, and a deception discrimination unit;

[0065] Since real aircraft are usually subject to stable power and aerodynamic constraints, frequency deviation changes are usually smooth and continuous. However, due to hardware control instability or power supply fluctuations, decoy launchers may experience short-term frequency oscillations. Therefore, a frequency disturbance analysis unit is set up.

[0066] The frequency disturbance analysis unit extracts the frequency offset values ​​of multiple consecutive frames for each located target aircraft to analyze the severity of the frequency deviation jump within a short time window. It also calculates the differential frequency drift amplitude of adjacent frames to obtain the root mean square fluctuation rate within the entire sampling window.

[0067] The root mean square volatility within the entire sampling window is obtained by the following formula:

[0068] ,in, For the The root mean square fluctuation rate of the target aircraft in the entire sampling window, is the number of continuous sampling frames, is the frame number, For the The frequency offset value of the frame, For the The frequency offset value of the frame, Number the target aircraft. is the differential frequency drift amplitude of adjacent frames; The higher the value, the more unstable the target's frequency deviation changes in a short period of time. The more unstable it is, the higher the suspicion of deception, because deception sources are prone to high-value jump characteristics.

[0069] Since the signal strength fluctuations of the echo reflected by a real aircraft are relatively smooth unless it undergoes strong maneuvers, the RF modulation of artificially manufactured decoy transmitters is often prone to slight power supply ripple and power jumps. Therefore, an RF modulation analysis unit is set up.

[0070] During the continuous sampling process, the RF modulation analysis unit obtains a sequence curve of the instantaneous signal power measured in the time domain changing with time for each target aircraft, and makes a discrete first-order derivative in the sampling sequence to identify the amplitude of power increase or decrease between frames, and continuously integrates the modulus square value of the first-order derivative to identify the power jump phenomenon of the corresponding target aircraft and construct the modulation perturbation flicker energy index.

[0071] At each sampling moment, the detection system collects and receives complex baseband signal samples. The system directly performs modular square operations on the complex baseband signal samples to form a continuous power sequence and a continuous time-series power sample stream: ,in, 、 and is the signal power in different sampling frames;

[0072] The first-order derivative reflects the instantaneous slope change of the power curve. This is actually to observe how drastic the fluctuation speed of the sequence curve is.

[0073] The modulated perturbation scintillation energy index is obtained by the following formula: ,in, For the Modulated perturbation scintillation energy index of target aircraft, For the The target aircraft The first derivative of the frame, No. The target aircraft Signal power of the frame.

[0074] No. The target aircraft The first derivative of the frame is obtained by the following formula: ,in, For the The target aircraft Signal power of the frame;

[0075] The modulus square value of the first-order derivative of the continuous integration essentially calculates the total amount of energy of the overall power change. By summing up all the jumps, the overall quantified power curve within the entire sampling window is quantified. If the power changes smoothly over a period of time, the overall integral value is small. If the flicker jumps frequently, the integral value is large.

[0076] Deceiving the discriminant unit, due to Describes short-term instabilities on the frequency channel, The short-term flicker energy on the power channel is described. The two belong to different dimensional physical quantities with large differences in numerical dimensions and absolute value ranges. Direct addition or direct ratio is not appropriate. Therefore, it is necessary to perform a standardized ratio operation on the root mean square fluctuation rate within the entire sampling window and the modulation perturbation flicker energy index of the corresponding target aircraft to make them comparable under the same discrimination scale. After weighted summation, the deception credibility coefficient is formed.

[0077] The deception credibility coefficient is obtained by the following formula: ,in, For the The target aircraft's deception credibility coefficient, and are weight values, and They are the reference values ​​of the root mean square fluctuation rate and the modulation perturbation scintillation energy index obtained by statistics in the historical period, which can be determined according to the mean value of the historical samples of the aircraft; is the frequency perturbation normalization, Normalize for perturbation energy;

[0078] If the decoy credibility coefficient exceeds the preset discrimination threshold, the corresponding target aircraft will be marked as a decoy, otherwise no marking will be performed.

[0079] In this embodiment of the present invention, the system first incorporates a frequency disturbance analysis unit. For each successfully positioned target aircraft, the unit extracts its frequency offset values ​​over multiple consecutive sampling frames to analyze the severity of frequency offset variations within a short time window. For example, in a practical application, the frequency offset values ​​of a drone swarm member in 10 consecutive sampling frames are Δf1, Δf2, ..., Δf10. The system calculates the frequency drift differences between adjacent frames, obtaining a frequency drift amplitude sequence. The RMS value of this sequence is then calculated to form the RMS fluctuation rate of the target within the entire sampling window.

[0080] The root mean square fluctuation rate is used to quantify the overall fluctuation intensity of the frequency deviation change process and is the core indicator for capturing the characteristics of frequency disturbances. It should be noted that the frequency deviation curve of a real target usually changes smoothly due to the influence of aerodynamic stability control, while the decoy launcher often exhibits short-term frequency deviation jumps due to unstable internal oscillation control.

[0081] Secondly, the system further features an RF modulation analysis unit that assists in identifying power modulation perturbations through time-domain power curve analysis. During continuous sampling, the system extracts the signal power value sequence for each frame in real time and calculates the first-order discrete derivative within the frame sequence, representing the magnitude of the power increase or decrease between adjacent frames. For example, if the power in one frame is 10dBm and the next is 13dBm, the difference in the first-order derivative is 3dBm. The system then performs a square integral of the first-order derivatives across all frames to generate an overall modulation perturbation flicker energy index.

[0082] The modulation perturbation flicker energy index describes whether the power variation curve exhibits dramatic jumps and flickering within a short period of time. Higher values ​​indicate more unstable target signal power fluctuations. It's important to note that while the echo power reflected by a normal aircraft is generally stable, decoys are prone to exhibiting slight power flickering characteristics during the RF modulation process due to factors such as power supply ripple and modulation index control instability.

[0083] Next, the system incorporates a deception detection unit to integrate the characteristic indicators of the two independent channels. First, the system normalizes the root mean square fluctuation rate and the modulated perturbation scintillation energy indicators. For example, by dividing them by a baseline reference value obtained from historical sample statistics (e.g., the typical mean and standard deviation of a real aircraft group under normal flight), the indicators of different scales are brought into a unified dimensional space. The deception credibility coefficient is then calculated by combining the normalized ratios and weighted summation.

[0084] The deception credibility coefficient is an overall deception probability discriminant indicator formed by fusing the characteristics of two different physical channels. A larger value indicates a higher degree of deception suspicion. Ultimately, the system automatically screens out suspicious targets that exceed the threshold. Through the aforementioned multi-dimensional feature extraction, multi-channel indicator fusion, and dynamic discrimination mechanism, the deception feature analysis subsystem of the present invention possesses highly adaptive intelligent recognition capabilities. It can effectively filter out misleading interference caused by counterfeit signals released by electronic deception devices to the detection system, ensuring the purity of the basic data for subsequent group organization identification and threat assessment, and improving the overall system's robustness and anti-deception capabilities in complex low-altitude swarm combat environments.

[0085] The swarm decomposition subsystem, including a clustering unit, a consistency analysis unit, a convergence analysis unit, and a coordinated penetration risk identification unit, determines several swarm labels, analyzes the operational differences of target aircraft within the swarm, and dynamically determines in real time whether the swarm is in an organized, highly coordinated offensive convergence penetration phase, thereby identifying high-risk, key surveillance swarms.

[0086] Furthermore, the swarm decomposition subsystem includes a clustering unit, a consistency analysis unit, a convergence analysis unit, and a collaborative penetration risk identification unit;

[0087] The clustering unit extracts the three-dimensional spatial coordinates and velocity vector of each target aircraft at the current sampling moment to obtain a joint feature vector, analyzes the similarity distance between any two target aircraft in the joint feature space, and determines several group labels based on the similarity distance values;

[0088] Use Euclidean distance to measure the similarity distance between two target aircraft in the joint feature space:

[0089] ,in, For the Target aircraft and Similarity distance between target aircraft, and These are the target aircraft numbers. is the axis coordinate number, For the The target aircraft The coordinate components on the axis coordinates, For the The target aircraft The coordinate components on the axis coordinates, For the The target aircraft The velocity component in the axis coordinate, For the The target aircraft Velocity components in axis coordinates;

[0090] When they are 1, 2 and 3 respectively, they are represented as X-axis coordinate, Y-axis coordinate and Z-axis coordinate;

[0091] After the pairwise distance matrix between all target aircraft is calculated, the hierarchical clustering process is started:

[0092] Set the merge distance threshold;

[0093] Iteratively merge adjacent target aircraft, specifically:

[0094] Each time, the two objects or clusters closest to each other are merged into a new cluster;

[0095] And update the new inter-cluster distance matrix;

[0096] Repeat the iteration until all inter-cluster distances are greater than the merging threshold;

[0097] Finally, several group labels are output, each of which represents a group subset with high similarity in spatial position and velocity characteristics.

[0098] Real swarm attacks are usually accompanied by highly coordinated, consistent movements with low speed deviation. If the speed dispersion of swarm members is too large, it may indicate a false induced assembly or the infiltration of interference signals. Therefore, a consistency analysis unit is introduced.

[0099] The consistency analysis unit determines the average velocity vector within each group and the modulus of the speed deviation of each target aircraft within the group. Based on the average velocity vector within each group and the modulus of the speed deviation of each target aircraft within the group, the operating differences of the target aircraft within the group are analyzed to obtain the overall speed consistency coefficient, which is specifically expressed as:

[0100] ;in, For the The group's overall speed consistency coefficient, that is, the normalized magnitude of the individual speed deviation from the mean within the group, Number the group, is the number of target aircraft in the corresponding group, Number the target aircraft in the corresponding group, For the corresponding group The three-dimensional motion velocity vector of the target aircraft, is the mean of the three-dimensional motion velocity vectors of all target aircraft in the corresponding group, is the mean modulus of velocity deviation of each target aircraft, is the modulus of the average velocity vector within each group;

[0101] The closer the overall speed consistency coefficient is to 0, the higher the speed consistency within the group is. Conversely, the more obvious the individual movement differences within the group are.

[0102] The convergence analysis unit calculates the mean of all pairwise Euclidean distances between target aircraft in each group at each moment, and records it as the group average density value at each moment. Within the entire sampling window, the first-order difference method is used to approximate the group density convergence rate. The group density convergence rate is used to dynamically and in real time determine whether the corresponding group is in an organized, highly coordinated offensive convergence penetration stage.

[0103] Among them, pairwise Euclidean distance refers to the similarity distance between two target aircraft in the joint feature space measured by Euclidean distance in the clustering unit;

[0104] Specifically, if the average group density value is less than 0, it means that the average distance between group members is shrinking, indicating that the group is gathering and converging towards the center. If it is greater than 0, it means that the average group distance is increasing, indicating that the group is dispersing, disbanding, or roaming. If it is approximately equal to 0, it means that the group maintains a balanced distance, indicating a stable formation or cruising state.

[0105] The collaborative penetration risk identification unit performs standard inversion processing on the overall speed consistency coefficient, converts the group density convergence rate into an absolute value, and then calculates the swarm attack risk factor by suppressing low-threat false alarms. If the swarm attack risk factor exceeds the preset risk threshold, the corresponding group will be marked as a high-risk key defense group.

[0106] The specific content of converting the population density convergence rate into an absolute value is as follows: ,in, For the The group density convergence rate after the absolute value processing, For the The rate of convergence of the population density of the population, To take the minimum value; only take the absolute value when converging (negative), and the diffusion (positive value) is automatically regarded as 0;

[0107] The swarm attack risk factor is obtained by the following formula: ,in, For the The swarm attack risk factor of the group, is the convergence rate safety threshold, To prevent small positive numbers from dividing by zero, we usually take the minimum value (e.g. ), avoid the denominator to be zero and ensure numerical stability, This is the result of standard inversion processing of the overall speed consistency coefficient. It is a lower limit suppression function, which ensures that when the convergence rate is lower than the safety threshold, the risk factor output is zero to avoid false alarms;

[0108] In the embodiment of the present invention, by setting up a swarm decomposition subsystem, the present invention can effectively perform real-time identification and hierarchical evaluation of the organizational structure and coordinated attack intentions of low-altitude target groups during their dynamic evolution, and has the following beneficial technical effects:

[0109] First, the system incorporates a clustering unit that extracts the 3D spatial coordinates and velocity vector of each target aircraft at the current sampling moment to form a joint feature vector. For example, if there are 20 aircraft at a given moment, some of them are close together with the same velocity direction, while some are flying independently. The system calculates the joint feature space distance between any two targets and uses a hierarchical clustering algorithm to automatically classify them into several group labels.

[0110] The joint eigenvector normalizes the position and velocity characteristics, unifies different physical quantities into the clustering model, improves the organizational accuracy of group division, and avoids the pseudo-clustering phenomenon caused by spatial proximity but unrelated speed.

[0111] Secondly, the consistency analysis unit quantitatively analyzes the operational consistency of each swarm's members. The system first calculates the average velocity vector within the swarm. For example, within a certain swarm, the average velocity vector is [unit:s]. It then calculates the deviation of each member from the average velocity, i.e., the velocity deviation modulus. If the swarm's members have highly consistent speeds, the deviation is small, and the overall speed consistency coefficient is low. For example, in a real swarm attack formation, the speed difference between members is typically less than 2 m / s. However, in decoy formations or temporary gathering and dispersal, the deviation can rapidly increase.

[0112] The speed consistency coefficient is used to quantify the degree of coordination within the group. Highly coordinated attacks often correspond to low consistency coefficients.

[0113] Third, the convergence analysis unit models the dynamic convergence trend of the group's spatial distribution in real time. At each moment, the system calculates the mean pairwise Euclidean distance between all members within the group, forming the group's average density value at that moment. For example, if the average distance of a group gradually shrinks from 80 meters to 50 meters within 10 seconds, the system calculates the density convergence rate using the first-order difference method. If the density convergence rate is negative and its absolute value gradually increases, it indicates that the group is rapidly converging towards the center and may be entering a highly coordinated attack formation. If the density convergence rate is close to zero or positive, it indicates that the group as a whole is in a dispersed cruising state.

[0114] The density convergence rate is used to capture the dynamic convergence and gathering trends of a swarm, a key dynamic characteristic for early identification of coordinated penetration. Finally, the coordinated penetration risk identification unit integrates consistency analysis with convergence trends to form a unified attack risk factor. Specifically, the velocity consistency coefficient is subjected to a standard inversion, and the density convergence rate is taken as the absolute value. The fusion model then calculates the swarm attack risk factor. For example, if a swarm has extremely high consistency and a high convergence rate, the risk factor is amplified, and the system sets a warning threshold. For example, if the swarm attack risk factor exceeds 0.7, the swarm is automatically marked as a high-risk, key surveillance group.

[0115] The swarm attack risk factor, a comprehensive indicator that integrates the degree of organizational coordination and dynamic convergence trends, reflects the likelihood that the swarm will enter the attack and penetration phase. Through real-time dynamic analysis of the swarm decomposition subsystem described above, the system effectively avoids the false alarms and misjudgments often associated with traditional static identification methods due to the occasional convergence of single distributions. Through continuous dynamic behavior modeling and highly coordinated clustering process extraction, the system achieves accurate and early identification of the evolution of swarm attack organizations. This effectively extends the lead time for defensive decisions, improves overall dynamic monitoring accuracy, and enhances the intelligent scheduling of defense resources. This makes it particularly well-suited for complex combat scenarios involving multiple swarms operating in parallel formations and multi-directional infiltration and intrusion.

[0116] The correction subsystem, combined with the deception identification unit, re-evaluates the high-risk key surveillance groups and conducts dynamic scheduling operations based on the degree of attack risk.

[0117] Furthermore, the correction subsystem includes a purification unit and a scheduling unit;

[0118] The purification unit extracts the high-risk key surveillance group containing the decoys and removes the decoys to form a purified group. After removing the decoy components (decoys), the updated overall speed consistency coefficient and group density convergence rate of the purified group are recalculated, and the swarm attack risk factor is updated in real time. It is re-evaluated whether it is a high-risk key surveillance group. If so, it is included in the surveillance group set.

[0119] The scheduling unit sorts the swarm attack risk factors corresponding to each high-risk key monitoring group in the monitoring group set in descending order to form a dynamic defense resource scheduling priority list. According to the dynamic defense resource scheduling priority list, it is evenly divided into three levels, namely high priority group, medium priority group and low priority group, and according to the three levels, it drives the system's multi-level dynamic defense actions.

[0120] Drive the system's multi-level dynamic defense actions: high-priority groups are focused on surveillance, allocated multi-frequency and high-refresh rate detection resources, activated advanced interference source identification channels and physical countermeasure platforms (such as high-frequency interception interference, directed energy weapons, directional communication suppression, etc.); medium-priority groups maintain medium refresh frequency tracking, continuously observe dynamic changes, and prevent sudden evolution; low-priority groups only retain low-power background cruise monitoring, do not consume high-level detection resources, and save system load.

[0121] In this embodiment of the present invention, the system first features a purification unit. Its core function is to further cleanse and update the data of key target groups that are still identified as high-risk after deception feature analysis. The system first extracts the labels of high-risk groups that include members marked as decoys. For example, if a high-risk group originally contains 20 members, and the system detects that the deception credibility coefficients of three members exceed a threshold and are therefore marked as decoys, the system automatically removes these three decoy individuals, resulting in a purified true target subgroup (i.e., 17 true members remaining).

[0122] The purification unit is used to complete decoy removal at the group level, ensuring that subsequent risk assessment is based on a real and effective set of target members, and avoiding decoy contamination affecting the accuracy of group characteristic modeling.

[0123] Secondly, after removing decoys, the purification unit immediately performs a secondary feature reconstruction analysis on the purified swarm, recalculating the overall velocity consistency coefficient and the swarm density convergence rate. For example, after removing decoys, the swarm's internal velocity deviation may decrease, reflecting stronger coordinated consistency, or the spatial convergence rate may be further amplified. Based on these two updated indicators, the system recalculates the swarm's attack risk factor. If the updated risk factor still exceeds the preset high-risk threshold, the system reclassifies the swarm as an effective high-risk, prioritized group.

[0124] The real-time update of the swarm attack risk factor reflects the dynamic correction characteristics of the group status after the deception is eliminated, ensuring that the risk assessment results are highly timely and adaptable in real time.

[0125] Next, the dispatch unit ranks the identified high-risk key groups within the current set of monitored groups by their swarm attack risk factor values, forming a dynamic priority list for defense resource scheduling. For example, within the current detection cycle, there are three high-risk monitoring groups with calculated swarm attack risk factors of 0.92, 0.85, and 0.76, respectively. The system ranks them as the first high-risk key group, the second high-risk key group, and the third high-risk key group.

[0126] The dynamic defense resource scheduling priority list intelligently and adaptively orchestrates resource allocation priorities based on real-time quantified risk, ensuring the defense system consistently focuses on the groups with the greatest potential threat. Finally, the scheduling unit categorizes the targeted groups into three levels based on the rankings: high-priority, medium-priority, and low-priority. This then drives system defense resources to implement multi-level dynamic defense actions based on these levels.

[0127] For example, high-priority groups can be immediately allocated more high-refresh rate radar beams, key optoelectronic observation windows and interference suppression beams, medium-priority groups maintain high-frequency monitoring and tracking, and low-priority groups enter low-frequency cruise mode to save the overall system load.

[0128] Multi-level dynamic defense action is to intelligently allocate defense means based on real-time risk levels to optimize defense effectiveness and avoid waste and excessive dispersion of defense resources. Through the purification, stripping, dynamic reconstruction and intelligent scheduling closed-loop design of the above-mentioned correction subsystem, the system of the present invention can achieve continuous adaptive evolution of defense resource optimization configuration under the background of swarm-type highly coordinated dynamic penetration attack, improve dynamic monitoring sensitivity, shorten response delay, and effectively improve the overall decision-making efficiency and defense robustness of the overall defense system in the face of complex deceptive mixed swarm invasion situations.

[0129] All the above parameters are scaled using dimensionless processing technology to eliminate the dimension differences between different data sources.

[0130] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A distributed low-altitude target passive collaborative positioning and threat assessment system, characterized by: include, The space preliminary positioning subsystem deploys multiple passive detection nodes outside the target defense area to capture all incoming signals in the low-altitude airspace and lock the three-dimensional spatial coordinates of the corresponding target aircraft; The deception feature analysis subsystem, which includes a frequency disturbance analysis unit, a radio frequency modulation analysis unit, and a deception discrimination unit, dynamically identifies the credibility of the target aircraft's deception behavior by analyzing the severity of frequency deviation jumps within a short time window and identifying power jumps in the target aircraft. The swarm decomposition subsystem, including a clustering unit, a consistency analysis unit, a convergence analysis unit, and a coordinated penetration risk identification unit, determines several swarm labels, analyzes the operational differences of target aircraft within the swarm, and dynamically determines in real time whether the swarm is in an organized, highly coordinated offensive convergence penetration phase, thereby identifying high-risk, key surveillance swarms. The collaborative penetration risk identification unit performs standard inversion processing on the overall speed consistency coefficient, converts the swarm density convergence rate into an absolute value, and calculates the swarm attack risk factor by suppressing low-threat false alarms. If the swarm attack risk factor exceeds the preset risk threshold, the corresponding swarm is marked as a high-risk key surveillance swarm. The correction subsystem, including the purification unit and the scheduling unit, combines with the deception identification unit to re-evaluate high-risk key surveillance groups and dynamically schedule operations based on the degree of attack risk; The purification unit extracts the high-risk key surveillance group containing decoys and removes the decoys to form a purified group. After removing the decoy components, the updated overall speed consistency coefficient and group density convergence rate of the purified group are recalculated, and the swarm attack risk factor is updated in real time. It is re-evaluated whether it is a high-risk key surveillance group. If so, it is included in the surveillance group set.

2. The distributed low-altitude target passive collaborative positioning and threat assessment system according to claim 1, characterized in that: The spatial preliminary positioning subsystem includes an acquisition unit and a positioning unit; The acquisition unit deploys multiple passive detection nodes outside the target defense zone to capture all arrival signals in the low-altitude airspace. Each detection node has a built-in clock module and achieves time synchronization through the Beidou timing system. The arrival signal includes the target reflection signal, arrival time difference, frequency drift and signal strength. The positioning unit randomly selects a reference detection node and uses the time difference method to calculate the relative position of each signal arrival, obtains the time difference equation, and solves it to determine the preliminary spatial position and timestamp of the target aircraft. It also uses the Doppler effect generated by the high-speed motion of the target to detect the frequency offset value received by each station to infer the radial velocity of the target aircraft. Based on the radial velocity and the timestamp of the preliminary spatial position of the target aircraft, it determines the offset distance vector of the target aircraft in the direction of its velocity within the entire acquisition time window. Combined with the preliminary spatial position, the three-dimensional spatial coordinates of the corresponding target aircraft are locked. Specifically, ,in, is the three-dimensional space coordinate, is the initial spatial position, is the offset distance vector; Among them, the specific expression of the time difference equation is: ,in, For the Station propagation path distance, is the reference detection node propagation path distance, is the signal propagation speed, For the The arrival time difference of the signal received by the station.

3. The distributed low-altitude target passive collaborative positioning and threat assessment system according to claim 2, characterized in that: The deception feature identification subsystem includes a frequency disturbance analysis unit, a radio frequency modulation analysis unit, and a deception discrimination unit; The frequency disturbance analysis unit extracts the frequency offset values ​​of multiple consecutive frames for each located target aircraft to analyze the severity of the frequency deviation jump within a short time window. It also calculates the differential frequency drift amplitude of adjacent frames to obtain the root mean square fluctuation rate within the entire sampling window. During the continuous sampling process, the RF modulation analysis unit obtains a sequence curve of the instantaneous signal power measured in the time domain changing with time for each target aircraft, and makes a discrete first-order derivative in the sampling sequence to identify the amplitude of power increase or decrease between frames, and continuously integrates the modulus square value of the first-order derivative to identify the power jump phenomenon of the corresponding target aircraft and construct the modulation perturbation flicker energy index.

4. The distributed low-altitude target passive collaborative positioning and threat assessment system according to claim 3, characterized in that: The decoy identification unit performs a standardized ratio operation on the root mean square fluctuation rate in the entire sampling window and the modulation perturbation scintillation energy index of the corresponding target aircraft, and forms a decoy credibility coefficient after weighted summation. If the decoy credibility coefficient exceeds the preset discrimination threshold, the corresponding target aircraft will be marked as a decoy, otherwise no marking will be performed.

5. The distributed low-altitude target passive collaborative positioning and threat assessment system according to claim 4, characterized in that: The swarm decomposition subsystem includes clustering unit, consistency analysis unit, convergence analysis unit and collaborative penetration risk identification unit; The clustering unit extracts the three-dimensional spatial coordinates and velocity vector of each target aircraft at the current sampling moment to obtain a joint feature vector, analyzes the similarity distance between any two target aircraft in the joint feature space, and determines several group labels based on the similarity distance values; a consistency analysis unit, determining the average velocity vector within each group and the modulus of the speed deviation of each target aircraft within the group, and analyzing the degree of operational differences of the target aircraft within the group through the average velocity vector within each group and the modulus of the speed deviation of each target aircraft within the group to obtain an overall speed consistency coefficient; The convergence analysis unit calculates the mean of all pairwise Euclidean distances between target aircraft in each group at each moment, and records it as the group average density value at each moment. Within the entire sampling window, the first-order difference method is used to approximate the group density convergence rate. The group density convergence rate is used to dynamically and in real time determine whether the corresponding group is in an organized, highly coordinated offensive convergence penetration stage.

6. The distributed low-altitude target passive collaborative positioning and threat assessment system according to claim 5, characterized in that: The scheduling unit sorts the swarm attack risk factors corresponding to each high-risk key monitoring group in the monitoring group set in descending order to form a dynamic defense resource scheduling priority list. According to the dynamic defense resource scheduling priority list, it is evenly divided into three levels, namely high priority group, medium priority group and low priority group, and according to the three levels, it drives the system's multi-level dynamic defense actions.

Images