High-maneuvering low-speed target take-off and landing process parameterization detection method based on common inductance signals

Abstract

The invention discloses a high-maneuvering low-speed target take-off and landing process parameterization detection method based on a common inductance signal. The method comprises the steps that a motion model of an aircraft in the take-off and landing stage is built, and a formula modeling result is obtained; obtaining distances and speeds of a plurality of suspicious targets; designing a band-pass filter in a Doppler domain to obtain an LVD result, calculating Doppler frequency information of the target, and obtaining accelerations of a plurality of suspicious targets; calculating retroreflection coefficients of the plurality of suspicious targets based on a radar antenna design principle; based on array antenna arrangement of a sensing system, obtaining space angles of a plurality of suspicious targets, and determining the space position of each suspicious target by combining the distances of the plurality of suspicious targets; and a detection module based on physical characteristics of the moving target is utilized to screen out an aircraft target according to respective threshold ranges of the speed, the acceleration, the retroreflection coefficient and the spatial position of the target. According to the invention, the detection capability of a low-speed and high-maneuverability moving target in a low signal-clutter-noise ratio environment is improved.

Description

Technical Field

[0001] This invention belongs to the field of sensor signal processing and moving target detection technology, specifically relating to a parameterized detection method for the take-off and landing process of highly maneuverable low-speed targets based on sensor signals. Background Technology

[0002] Like radar equipment, integrated sensing devices possess all-weather, 24 / 7 operational capabilities. Furthermore, compared to radar equipment, integrated sensing devices offer higher functional integration and lower deployment costs, making low-altitude moving target detection based on integrated sensing devices widely applicable in urban low-altitude traffic monitoring and public safety. The core idea of ​​using electromagnetic wave signals to detect moving targets is to distinguish them by leveraging the significant differences between the target's characteristics and the characteristics of scene clutter in the range-Doppler domain. Traditional constant false alarm rate (CFAR) algorithms in the range-Doppler domain rely solely on target energy information for detection. This leads to several shortcomings when detecting highly maneuverable, low-speed targets, including low detection rates for targets with low projected velocity and weak classification and recognition capabilities for targets with high projected acceleration. Therefore, research into designing and implementing accurate detection technologies for highly maneuverable, low-speed targets is of great significance.

[0003] Most existing moving target detection methods are based on coherent accumulation and error correction schemes under low signal-to-noise ratio. The basic idea is as follows: first, perform conventional imaging on the recorded echoes, infer the source of error by analyzing the imaging results, then estimate the parameters related to the source of error, correct the error by the estimated parameters, and finally perform moving target detection based on the focused image.

[0004] In the paper “RP Perry, RC Dipietro, RL Fante. SARimaging of moving targets[J]. IEEE Transactions on Aerospace and Electronic Systems, 1999, 35(1): 188-200”, RCDiPietr et al. proposed the Keystone Transform (KT), which eliminates the problem of linear increase in the relative distance between the target and the radar caused by the target's own velocity, i.e., the range movement problem. In the paper "Li X, Cui G, Yi W, et al. Fast coherent integration for maneuvering target with high-order range migration via TRT-SKT-LVD[J].IEEE Transactions on Aerospace and Electronic Systems, 2016, 52(6):2803-2814," Li Xiaolong et al. proposed a coherent accumulation method (TRT-SKT-LVD) based on Time-frequency modulation transform (TRT), wedge transform (SKT), and Lv transform (LVD). This method uses time-frequency modulation transform to perform multidimensional search and extraction of the signal's parameter space, and uses wedge transform to correct for distance curvature caused by acceleration. Finally, Lv transform is used to achieve coherent accumulation of the signal under low signal-to-noise ratio. To improve computational complexity, Li Xiaolong et al. also proposed an improved method based on the iterative adjacent cross-correlation function (ACCF) algorithm, namely, a coherent accumulation algorithm based on the adjacent cross-correlation function and LVD (ACCF-LVD). This algorithm is implemented using Fast Fourier Transform and Fast Inverse Fourier Transform, eliminating the need for any parameter search steps and significantly improving computational efficiency.

[0005] The above methods have two main drawbacks in achieving high-maneuverability target detection: First, during takeoff and landing, the aircraft's speed is low, resulting in a small Doppler frequency shift in its echo signal. In the traditional range-Doppler two-dimensional spectrum, this signal is easily overwhelmed by strong stationary or slow-moving clutter, causing conventional detection algorithms to fail completely. Second, even if the target signal is not completely overwhelmed, the presence of strong clutter severely affects the accuracy of parameter estimation (especially acceleration). Existing algorithms lack robustness in low signal-to-noise ratio (SNR) environments, leading to a significant deterioration in detection performance. These issues of high complexity and low robustness collectively constitute the main obstacle preventing current advanced algorithms from theoretical application to practical use. Summary of the Invention

[0006] To address the aforementioned problems in the existing technology, this invention provides a parameterized detection method and apparatus for the takeoff and landing process of highly maneuverable, low-speed targets based on sensing signals. The technical problem to be solved by this invention is achieved through the following technical solution: In a first aspect, embodiments of the present invention provide a parameterized detection method for the takeoff and landing process of a highly maneuverable, low-speed target based on sensor signals. For the interaction scenario between a sensor system and an aircraft target, the method includes: S1, Construct a motion model of the aircraft during takeoff and landing, and obtain the relevant formula modeling results; S2, Based on the modeling results of the formula, the distance to the moving target is estimated to obtain the distances to multiple suspicious targets; S3, based on the modeling results of the formula, estimate the velocity of the moving target to obtain the velocity of multiple suspicious targets; S4. Based on the modeling results of the formula, LVD results are obtained by designing a bandpass filter in the Doppler domain, and the Doppler modulation frequency information of the target is calculated to estimate the acceleration of the moving target and obtain the acceleration of multiple suspicious targets. S5, based on radar antenna design principles, calculates the back reflection coefficients of multiple suspicious targets; S6. Based on the array antenna configuration of the sensing system, the spatial position of the moving target is estimated to obtain the beam domain spatial angle measurement result spectrum, the spatial angles of multiple suspicious targets are obtained, and the spatial positions of multiple suspicious targets are determined by combining the distances of the multiple suspicious targets. S7 utilizes a detection module based on the physical characteristics of moving targets to filter out aircraft targets from multiple suspicious targets based on preset threshold ranges for the target's acceleration, back reflection coefficient, and spatial position.

[0007] Secondly, embodiments of the present invention provide a parameterized detection device for the take-off and landing process of a highly maneuverable, low-speed target based on sensor signals. For the interaction scenario between the sensor system and the aircraft target, the device includes: The motion model building module is used to build motion models of aircraft during takeoff and landing, and obtain relevant formula modeling results. The target distance estimation module is used to estimate the distance of moving targets based on the modeling results of the formula, and obtain the distances of multiple suspicious targets; The target velocity estimation module is used to estimate the velocity of moving targets based on the modeling results of the formula, thereby obtaining the velocities of multiple suspected targets. The target acceleration estimation module is used to estimate the acceleration of moving targets based on the modeling results of the formula, obtain LVD results by designing a bandpass filter in the Doppler domain, calculate the Doppler modulation frequency information of the target, and obtain the acceleration of multiple suspicious targets. The back reflection coefficient calculation module is used to calculate the back reflection coefficient of multiple suspicious targets based on the radar antenna design principle. The spatial position estimation module is used to estimate the spatial position of moving targets based on the array antenna settings of the sensing system, obtain the beam domain spatial angle measurement result spectrum, obtain the spatial angles of multiple suspicious targets, and determine the spatial positions of multiple suspicious targets by combining the distances of the multiple suspicious targets. The target screening module is used to filter out aircraft targets from multiple suspicious targets by using the detection module based on the physical characteristics of moving targets and according to preset threshold ranges for the target's speed, acceleration, back reflection coefficient and spatial position.

[0008] For the critical scenario of aircraft takeoff and landing, this invention provides a parameterized detection method for the takeoff and landing process of highly maneuverable low-speed targets based on sensing signals. First, a motion model of the aircraft's takeoff and landing phase is constructed, yielding relevant formula modeling results. Next, based on the formula modeling results, the distances of multiple suspected targets are estimated; the velocities of multiple suspected targets are estimated; and by designing a bandpass filter in the Doppler domain to obtain LVD results, the Doppler modulation information of the targets is calculated to estimate the acceleration of multiple suspected targets. Furthermore, based on radar antenna design principles, the back reflection coefficients of multiple suspected targets are calculated. Based on the array antenna settings of the sensing system, the spatial position of the moving targets is estimated, yielding a beam domain spatial angle measurement result spectrum, which provides the spatial angles of multiple suspected targets. Combined with the distances of the multiple suspected targets, the spatial positions of the multiple suspected targets are determined, thus obtaining the parameters of multiple suspected targets. Finally, using a detection module based on the physical characteristics of the moving targets, and according to preset threshold ranges for the target's velocity, acceleration, back reflection coefficient, and spatial position, the aircraft target is selected from the multiple suspected targets.

[0009] This method utilizes the parameterized physical characteristics of the target, employing a sensory system and parameterized detection algorithms for accurate detection during takeoff and landing, improving anti-interference capabilities in complex environments and enabling reliable detection of low-altitude, slow-moving, highly maneuvering targets. This invention addresses the problem of reduced time-frequency focusing performance caused by clutter masking by designing a bandpass filter in the Doppler domain to first separate the target signal and then perform time-frequency analysis. When multiple targets or interference sources exist in the detection airspace, the system needs to be able to distinguish them. This invention solves the problem of multi-target discrimination by combining range dimension resolution (moving target range estimation in step S2) and spatial angle resolution (BS-MUSIC algorithm in step S6). The detection range is optimized using an integrated sensory system, and the detection effect of low-speed, highly maneuvering targets in low signal-to-noise ratio environments is optimized using parameterized detection algorithms that utilize range, velocity, acceleration, back reflection system, and spatial position information in the echo. Attached Figure Description

[0010] Figure 1 This is a flowchart illustrating a parameterized detection method for the take-off and landing process of a highly maneuverable, low-speed target based on sensor signals, provided in an embodiment of the present invention. Figure 2 This is a schematic diagram of the aircraft motion model in an embodiment of the present invention; Figure 3 This is a schematic diagram of the target Doppler effect in an embodiment of the present invention; Figure 4 This is a schematic diagram of an ideal spatial array in an embodiment of the present invention; Figure 5 A schematic diagram of beamforming; Figure 6 This is a flow graph for target motion parameter estimation and detection in an embodiment of the present invention; Figure 7 This is a schematic diagram illustrating the principle of the parameterized detection method for the take-off and landing process of a highly maneuverable low-speed target based on sensor signals provided in an embodiment of the present invention. Figure 8 This is a range Doppler domain image from an experiment in an embodiment of the present invention; Figure 9 This is a graph showing the detection performance of the detector in an experiment according to an embodiment of the present invention. Figure 10 This is a schematic diagram of the structure of a parametric detection device for the take-off and landing process of a high-mobility, low-speed target based on sensing signals, provided in an embodiment of the present invention. Detailed Implementation

[0011] The present invention will be further described in detail below with reference to specific embodiments, but the implementation of the present invention is not limited thereto.

[0012] In a first aspect, embodiments of the present invention provide a parameterized detection method for the take-off and landing process of highly maneuverable low-speed targets based on sensor signals. Targeting the interaction scenario between the sensor system and the aircraft target, it mainly utilizes an integrated communication and sensing system (sensor system) combined with various parameterized features of the target to achieve more accurate detection of the take-off and landing phases of highly maneuverable targets. It can be used in urban low-altitude traffic supervision and public safety fields, such as low-altitude target detection in no-fly zones and urban areas with high population density.

[0013] Highly maneuverable low-speed targets primarily refer to highly maneuverable low-speed aircraft (such as drones). High maneuverability means the aircraft can change its flight direction within a certain timeframe (achieving fast turns and high acceleration). Low speed refers to speeds below a certain threshold, such as a speed range of -14 to 14. The acceleration range can be from -3 to 3. .

[0014] like Figure 1 As shown, the method may include the following steps S1 to S7: S1, Construct a motion model of the aircraft during takeoff and landing, and obtain the relevant formula modeling results; S1 includes the following steps: S11, establish the instantaneous expression for the distance between the aircraft target and the sensing system, expressed as: (1); Wherein, it is assumed that the station height of the sensor system is , The speed of light is [value], and the horizontal distance between the spacecraft and the sensing system is [value]. The aircraft uses speed as acceleration is Perform uniformly accelerated takeoff motion; For direction, slow time, The pitch angle of the aircraft target relative to the sensor system; a schematic diagram of the aircraft motion model is shown below. Figure 2 As shown.

[0015] S12, based on the instantaneous expression for the distance between the aircraft target and the sensing system, the expression for the echo signal received by the sensing system after being reflected by the aircraft target is obtained, expressed as: (2); In the aforementioned sensing system, each pulse of the transmitted waveform consists of multiple complete OFDM (Orthogonal Frequency Division Multiplexing) symbols, and returns to the sensing system after being reflected from the target. The total time is... The above expression (2) can be obtained by recovering the echo signal after channel coding and channel estimation of a single OFDM symbol. The target's back reflection coefficient, For signal transmission bandwidth, Indicates distance in advance; Using distance to fast time The bandwidth of a signal transmission is a measure of its transmission bandwidth. The function; The wavelength of the signal electromagnetic wave; The imaginary unit; S13, by receiving the expression for the echo signal reflected by the target from the aircraft through the sensing system, the expression for the Doppler frequency of the target is obtained, which is expressed as: (3); in, Indicates the Doppler frequency of the target; echo signal The overall phase; The above expression shows that the target's velocity information is contained in the target's spatial position and Doppler.

[0016] S14, further differentiate the expression for the Doppler frequency of the target to obtain the expression for the Doppler modulation frequency information of the target, which is expressed as: (4); in, This indicates the Doppler modulation frequency information of the target.

[0017] The above expression shows that the target's acceleration information and spatial position information together determine the target's Doppler modulation frequency.

[0018] In step S1 of this embodiment, several relevant formulas (1) to (4) used in parameter estimation are derived through simple modeling, which serve as the result of formula modeling. S2, Based on the modeling results of the formula, the distance to the moving target is estimated to obtain the distances to multiple suspicious targets; Formula (2) in step S1 shows that the target energy is focused in the fast time domain at time t. (The focal point is not unique, and the obtained...) (The value is not unique). Therefore, S2 includes: Based on the expression for the echo signal received by the sensing system after reflection from the aircraft target (i.e., formula (2) in step S1), the focusing situation of the actual echo is analyzed to determine the time delay. Value, based on each determined delay The value is used to obtain the distance of the focal point relative to the synergistic system, and the distance of each suspicious target to the synergistic system is obtained. The calculation formula is as follows: (5); The distance from the suspected target to the sensing system; after the echo signal is discretized and sampled, each sampling point corresponds to a distance value. .

[0019] S3, based on the modeling results of the formula, estimate the velocity of the moving target to obtain the velocity of multiple suspicious targets; Specifically, S3 includes: S31, based on the expression of the Doppler frequency of the target (i.e. formula (3) in step S1), determine the target velocity information contained in the Doppler domain corresponding to the azimuth slow time, and determine the display form of the echo signal in the frequency domain based on the target Doppler effect; As shown in formula (3) in step S1, the signal can achieve coherence in the azimuth-slow time dimension, and the slant range variation information between the target and the sensing system is also contained in the azimuth-slow time dimension. Therefore, the target velocity information is contained in the spectrum (Doppler domain) corresponding to the azimuth-slow time dimension, and the target Doppler effect is as follows: Figure 3 As shown.

[0020] The echo signal is represented in the frequency domain by the Doppler frequency center at... Expanded to ; S32, based on the frequency domain representation of the echo signal and the expression for the Doppler frequency of the target, the velocity range of each suspected target is estimated, and the velocity of the suspected target is obtained by averaging. The calculation formula used is as follows: (6); Each suspected target can be calculated to have a target velocity, and a suspected target is represented by multiple sampling points. The number of grids to widen for a suspected target; For a certain suspicious target The Doppler frequency corresponding to the grid expansion; the number of grids for target expansion is determined based on the number of sampling points occupied by the suspected target during echo signal analysis, and the number of grids for expansion varies for different suspected targets.

[0021] S4. Based on the modeling results of the formula, LVD results are obtained by designing a bandpass filter in the Doppler domain, and the Doppler modulation frequency information of the target is calculated to estimate the acceleration of the moving target and obtain the acceleration of multiple suspicious targets. From formula (3) in step S1, it can be seen that the Doppler of the target contains velocity and acceleration information. From formula (4), it can be seen that the Doppler modulation frequency information can be obtained by further differentiating the Doppler. The velocity information, acceleration information, and spatial position information of the target together determine the Doppler modulation frequency of the target. The modulation frequency of the signal can be estimated using the traditional Lv's Distribution (LVD).

[0022] When estimating the frequency modulation of a signal using the Lv transform, this invention first considers the case where the range cell containing the target does not introduce clutter interference signals. In this case, the echo signal of formula (2) is... Representing the orientation in a slower time dimension yields: (7); in, This is the result of representing the azimuth of the echo signal in the slow time dimension; The number of signals in the echo; Indicates a suspicious target The target echo energy of the grid; For a certain suspicious target The Doppler frequency corresponding to the lattice broadening; For a suspicious target The Doppler modulation frequency corresponding to the lattice broadening; For formula (7) Do different lag times The instantaneous autocorrelation function is obtained as follows: (8); in, Indicates conjugate; Indicates the time delay transformation factor; Indicates cross-correlation; Introducing a scale conversion factor , making Equation (8) above becomes: (9); in, The azimuth timescale is the timescale after scaling. For formula (9) about , Performing Fourier transforms in both directions yields the LVD result: (10); in, This represents the LVD result at this point, where This represents the result after a two-dimensional Fourier transform; here Indicates the azimuth slow time after scale transformation Doppler frequencies after scaling via FFT transformation Indicates lag time The Doppler modulation frequency after scaling through FFT transformation; it can be seen that when there is no clutter interference, the LVD result is a two-dimensional focusing function. The above assumes that the range cell containing the target does not introduce clutter interference signals. However, actual situations are often more complex, so it is necessary to consider the case where the range cell containing the target introduces clutter interference signals. When the target and the interference characteristics are similar, or the interference energy is much stronger than the target energy (generally defined as an interference energy to target energy ratio exceeding 30dB), it will somewhat mask the time-frequency focusing results of the target. Simultaneously, since the observed signal is a time-finite signal, a time gate is used to limit the signal's validity. Based on the preceding text, S4 includes the following steps: S41 will transmit the echo signal Representing the orientation in a slower time dimension, we obtain: (11); in, The result is a slow-time dimension representation of the echo signal azimuth. On the middle molecule This is the time delay transformation factor. The pulse repetition interval is the time in the denominator. This represents the total observation time of the synsensory system. The number of signals in the echo; Indicates a suspicious target The target echo energy of the grid; For a certain suspicious target The Doppler frequency corresponding to the lattice broadening; For a suspicious target The Doppler modulation frequency corresponding to the lattice broadening; The azimuth Fourier transform of the echo signal yields a sinc function. Performing an azimuth Fourier transform yields a sum of multiple sinc functions, where different velocities cause targets of different frequencies to separate in the Doppler domain. In the range-time domain, the target is coupled with clutter signals, but in the range-Doppler domain, the target and clutter have a certain degree of distinction. This distinction can be used to represent the echo signal in the Doppler domain using a mid-pass filter; see S42.

[0023] S42, In the Doppler domain, the azimuth slow-time dimension representation of the echo signal is filtered using a mid-pass filter and subjected to an inverse Fourier transform to obtain the transformed echo signal, expressed as: (12); in, The azimuth slow time after scale transformation Doppler frequencies after scaling transformation following FFT transformation; This indicates the center frequency location of the pass filter; This refers to the bandwidth of the mid-pass filter. Indicates Fourier transform; Indicates the inverse Fourier transform; This represents the total echo energy of the signal passing through the center-pass filter. It is the first The echo energy of a signal; This indicates the Doppler frequency of the target calculated in this step; This indicates the Doppler modulation information of the target; It should be noted that in S42, due to After Fourier transform, we can obtain Different goals It can be done The details are as follows, please refer to the relevant technical documentation, and will not be elaborated upon here. S43, Perform LVD analysis on the transformed echo signal to obtain the autocorrelation function, expressed as: (13); Among them, by passing a mid-pass filter and performing an inverse Fourier transform, the time-domain signal of the individual echo of the target of interest can be obtained, which relates to formula (12). The above formula (13) can be obtained by performing LVD analysis; As the scale transformation factor, using Direction to slower time Scale conversion factor Transformed into azimuth slow time after scale transformation With lag time The function; S44, the autocorrelation function along and Fourier transform is performed in two dimensions to achieve energy focusing and parameter estimation of the target, yielding the corresponding LVD result, represented as: (14); in, This represents the corresponding LVD result, where, The azimuth slow time after scale transformation Doppler frequencies after scaling via FFT transformation The Doppler modulation frequency after scaling; Let be the impulse function; S45, based on the obtained LVD results and the calculated... And the expression for the Doppler modulation information of the target (i.e., formula (4)) is used to calculate and utilize Calculate the acceleration of the corresponding suspicious target.

[0024] According to formula (4), the expression for acceleration is: (15); In this embodiment of the invention, the time-domain signal of a single target is obtained by using a sliding mid-pass filter, and the time-frequency signal of the single target is analyzed by time-frequency analysis to obtain the acceleration estimate of different targets.

[0025] S5, based on radar antenna design principles, calculates the back reflection coefficients of multiple suspicious targets; S5 includes the following steps: S51, based on radar principles, determines the distance to the sensing system. The power density irradiated by the target is expressed as: (16); in, The power of the transmitter in the sensing system. For the transmission gain of the sensing system, The distance between the transmitting antenna of the sensing system and the target point; S52, after being reflected by the target, the echo signal spreads omnidirectionally. Assume the target's rearward reflection coefficient is... The distance between the target and the receiving antenna is The power density of the target echo at the receiver of the sensing system is determined as follows: (17); S53, when the effective area of ​​the receiving antenna is The echo power received by the inductive system is determined and expressed as: (18); S54, using antenna theory, determines the receiving antenna gain. With the effective area of ​​the receiving antenna The following relationship exists: (19); in, The wavelength of the signal electromagnetic wave; S55, based on the derivation formulas in S51~S54, the expression for the received echo power of the inductive system is determined as follows: (20); S55, through echo signal analysis Then, based on the distance estimation of the moving target, the following is obtained: and And based on the fixed parameter , , , Using the expression for the received echo power of the aforementioned sensing system, the back reflection coefficient of each suspected target is calculated. . Specifically, the echo power is received by the sensing system. It can be seen that the target's echo power depends on the distance between the target and the sensing system, the target's back reflection coefficient, and the design of the sensing system's transmitter and receiver. After the sensing system design is completed, the system gain remains constant, and the target echo energy is affected by both the distance between the target and the sensing system and the back reflection coefficient. The estimated value of the back reflection coefficient can be further obtained from the estimated value of the distance between the target and the sensing system in step S2.

[0026] S6. Based on the array antenna configuration of the sensing system, the spatial position of the moving target is estimated to obtain the beam domain spatial angle measurement result spectrum, the spatial angles of multiple suspicious targets are obtained, and the spatial positions of multiple suspicious targets are determined by combining the distances of the multiple suspicious targets. Assume there exists a point target A in three-dimensional space, and points B and C are projections of point target A onto the plane. The azimuth angle between the point target and the sensory system is... The pitch angle is The pitch array spacing of the sensing system is The horizontal array spacing is Array antennas, such as Figure 4 As shown. Assume the sensor system has azimuth... The antenna, the sensing system has in the pitch direction If there is an antenna, then S6 may include the following steps: S61, based on the array antenna configuration of the sensing system, determine the steering vector caused by the target on each antenna in various directions, expressed as: (twenty one); in, The CCP Multiply the fractions together.

[0027] S62, assuming the sensing system has in the pitch direction The steering vector caused by the target on each elevation antenna is expressed as: (twenty two); in, The CCP Multiply the fractions together.

[0028] S63, considering the combined effects of elevation and azimuth angles, the echoes from each antenna are represented as follows: ; in, The echo vector is composed of the echo signals from each antenna in a single transmission. For the echo data of each antenna, the subscript is... Indicates the direction to the first Root antenna, Indicates pitch to the first Root antenna; Let be the guiding vector of the spatial array; where: (twenty three); in, , ; S64, by utilizing the phase difference of the target in the array antenna caused by the spatial position information, the spatial position of the target is calculated by the BS-MUSIC angle measurement algorithm in the beam domain space, and the beam domain space angle measurement result spectrum is obtained. from It can be seen that the target has a certain phase difference in the array antenna due to its spatial position information. This phase difference can be used to calculate the target's spatial position. Here, the BS-MUSIC angle measurement algorithm in the beam domain is used for the calculation.

[0029] BS-MUSIC is an improvement on the MUSIC algorithm, requiring integration with a beamforming algorithm to form a beam space angle estimation algorithm. The beamforming principle is as follows: Figure 5 As shown, beam domain data is generated by superimposing multiple antenna echo data after linear transformation, with each beam corresponding to a different beamforming weight. Among these, This indicates the pitch direction towards the first antenna. Indicates pitch to the first One antenna.

[0030] S64 includes the following steps: S641, based on the Discrete Fourier Transform (DFT) beamforming method, is a beamforming method with low computational cost, performing two-dimensional beamforming, wherein the formed two-dimensional beam is represented as: (twenty four); in, Represents the echo data of each antenna; the azimuth of the generated beam's main lobe. Looking up and down By changing , This allows the beam to be pointed at different main lobes to obtain the superposition gain between antennas.

[0031] S642, with The main lobe of each beam points to , The main lobe of each beam points to Beamformer, definition Beamforming matrix, yes dimensional vector, containing The main lobe of each beam points to The beam formed by the beamformer yes dimensional vector, containing The main lobe of each beam points to The beam formed by the beamformer; one dimension corresponds to the beam pointing, and the other dimension corresponds to the beam weights of different antennas; where the beamforming matrix is ​​represented as: (25); in, for The form of expression; ; S643, the beamspace expression after beamspace transformation is determined as follows: (26); in, It is a beam domain spatial signal, which is a fast time representation of the range axis of a two-dimensional beam; S644, For the beamspace expression after beamspace transformation, calculate the covariance matrix, which is expressed as: (27); The MUSIC algorithm performs eigenvalue decomposition on the covariance matrix of the signal, decomposing it into a signal subspace and a noise subspace. The covariance matrix of equation (26) is given by equation (27). This is the autocorrelation matrix in the beam space; This is the autocorrelation matrix after spatial transformation; for The corresponding eigenvector matrix; Let be the covariance matrix of the beam domain spatial signal; This is the sum of squares of the mean values ​​of the row vectors of beam spatial information; This is the spatial transformation matrix after eigenvalue decomposition; S645 decomposes the beam space signal into a noise subspace. The expression for the beam domain spatial angle measurement result spectrum is as follows: ; in, This is the spatial transformation vector after eigenvalue decomposition. The computational steps of the BS-MUSIC algorithm can be summarized as follows: (1) Beamforming, and retaining the right to beamform.

[0032] (2) Select the multi-shot data to be located in the beam space.

[0033] (3) Use the data matrix in the beam space to perform MUSIC spectrum estimation, search for the maximum point, and find the corresponding signal direction.

[0034] For details on the specific processing procedure, please refer to the relevant technical explanations; further details will not be provided here.

[0035] S65, based on the beam domain spatial angle measurement result spectrum, obtain the spatial angles of multiple suspicious targets, and combine the distances of the multiple suspicious targets to determine the spatial positions of the multiple suspicious targets. Those skilled in the art will understand that the beam domain spatial angle measurement result spectrum is information composed of spatial angles corresponding to different sampling points of the echo signal. It can be used to determine the spatial angle of a suspicious moving target and, together with the estimated distance information, determine the spatial location of the suspicious moving target. For specific determination methods, please refer to existing calculation methods, which will not be described here.

[0036] S7 utilizes a detection module based on the physical characteristics of moving targets to filter out aircraft targets from multiple suspicious targets based on preset threshold ranges for the target's speed, acceleration, back reflection coefficient, and spatial position.

[0037] During takeoff and landing, the target's speed is low, making it difficult to distinguish it from strong ground clutter in the range-Doppler domain. The target exhibits significant projected acceleration during takeoff; however, static clutter, due to phase noise, spreads as a random signal and lacks stable acceleration. This specific situation can be used as a criterion for target detection during takeoff and landing.

[0038] The detection module based on the physical characteristics of moving targets is an algorithm-based module that can exist as software. This embodiment of the invention, based on the target detection requirements during takeoff and landing, pre-sets threshold ranges for target velocity, acceleration, backscatter coefficient, and spatial position for low-speed, highly maneuverable moving targets in low signal-to-noise ratio environments.

[0039] S7 includes: For any suspicious target, the data set obtained includes its velocity, acceleration, backscatter coefficient, and spatial position. It is determined whether all of these fall within the corresponding threshold range. If so, the suspicious target is identified as an aircraft target. It is understood that the spatial position of the suspicious target includes its distance and spatial angle.

[0040] In other words, a suspected target can only be identified as an aircraft target if its velocity, acceleration, backscatter coefficient, and spatial position all fall within their respective threshold ranges.

[0041] In practical processing, a feasible implementation method is to perform threshold determination according to a certain order. For example, first, the entire echo signal is analyzed, and signal-to-noise ratio (SNR) filtering (i.e., energy detection) is performed to filter out strong clutter interference and noise signals. Specifically, the SNR is calculated using the energy of the echo signal and the back reflection coefficient. Only when the SNR falls within the corresponding threshold range is it identified as a suspicious target area (the larger the back reflection coefficient, the stronger the energy), thus implicitly utilizing the back reflection coefficient for threshold determination. Then, the projection velocity, acceleration, and spatial position (including distance and spatial angle) of the suspicious target area are compared with the threshold range. If all fall within the corresponding threshold range, it is determined to be a suspicious low-altitude flying target (aircraft target). When determining the suspicious target area, the filtering speed is first controlled to reduce the amount of calculation, then the signal acceleration is estimated and determined to be a moving target, and finally the target's true position is determined by the spatial position information. The specific detection flowchart is as follows. Figure 6 As shown.

[0042] The detection range of the sensing system during target takeoff and landing can be determined by the density of sensing system stations in the detection environment, with multiple stations working together to avoid detection blind spots. The target energy range should be determined by the target's backscatter coefficient and the detection distance. The detection velocity range can be determined by clutter diffusion; traditional constant false alarm rate (CFAR) detection can be performed in weak clutter areas. Target acceleration is the key criterion for target detection; the target's acceleration information is ultimately reflected in parameters estimated through time-frequency analysis. In the middle. When facing The threshold is too small, which will generate a large number of false alarms; A large threshold will affect the detection rate of parameterized detection.

[0043] For the processing procedures of embodiments S1 to S7 of this invention, please refer to [link / reference]. Figure 7 understand.

[0044] The effectiveness of the method of the present invention can be further illustrated by the following simulation experiments.

[0045] 1. Simulation conditions: This invention utilizes a Xeon Silver 4110 CPU, an NVIDIA RTX 4090 GPU, and MATLAB 2020a developed by Mathworks for simulation. The target of interest in this invention's experiments is an aircraft.

[0046] The methods compared in the experiment are as follows: Traditional Constant False-Alarm Rate (CFAR) algorithm; 2. Simulation content: The algorithm was simulated, and the simulation parameters of the signal are shown in Table 1. Complex Gaussian white noise was added to the echo signal to make the signal-to-noise ratio of the echo -20dB; ground clutter was added to the echo signal with a ground clutter noise ratio of 20dB.

[0047] Table 1 Simulation parameters for low-speed moving targets

[0048] The range Doppler image after suppressing static clutter using the mean cancellation method is shown below. Figure 8 As shown. Figure 8 The results show that the spread signal of ground clutter obstructs the signal generated by a real low-speed moving aircraft. Traditional constant false alarm rate (CFAR) detection algorithms rely on the energy difference between the signal and surrounding signals for signal detection. In the current scenario, since the energy of the spread signal of the ground clutter is comparable to or differs from the echo signal of the small aircraft by no more than one order of magnitude, energy-based CFAR detection fails in this situation.

[0049] During the takeoff phase, the aircraft is relatively short, slow, and accelerates significantly. It is difficult to find a clutter suppression method that can effectively suppress static clutter and its spread while preserving the aircraft's echo signal. The aircraft's long-term motion is continuous, and target identification can be determined by correlating the distance, velocity, acceleration, spatial position, and energy of multiple signal frames. However, clutter and its spread are randomly distributed over long periods, and the parameters between multiple signal frames are not correlated. Therefore, parametric detection of the aircraft, replacing constant false alarm rate (CFAR) detection after clutter suppression, can accomplish the detection task for low-speed aircraft.

[0050] Using the parameters in Table 1 as an example, simulations of aircraft and ground clutter echo signals were performed. Constant false alarm rate (CFAR) detection and parameterized detection were performed on the simulated echo signals, and some detection results are shown in Tables 2 and 3. CFAR detection only provides information in three dimensions: signal-to-noise ratio (SNR), projection velocity, and distance; the parameterized detection method can add the Doppler frequency modulation (FCM) dimension, expanding the information to four dimensions: SNR, projection velocity, distance, and FCM.

[0051] Table 2 Result of Constant False Alarm Rate (CFAR) Detection

[0052] Table 3 Parameterized Detection Results

[0053] Table 2 shows that constant false alarm rate (CFAR) detection is based on signal-to-clutter ratio (SNR) to complete signal detection, so the detection results are mostly ground clutter spread signals. Table 3 shows that parameterized detection is based on the physical characteristics of the signal and can complete the aircraft detection task under low SNR.

[0054] Assuming the target is 150m away from the sensing system, the platform height is 30m, and the aircraft takes off with an initial velocity uniformly distributed within 0~1m / s, and the acceleration follows a uniform distribution of 1m / s²~3m / s², the energy of the static clutter after two-dimensional accumulation is 100dB. The aircraft echo signal energy is gradually increased in steps of 2dB with a signal-to-noise ratio improvement. 500 rounds of Monte Carlo experiments are conducted for each signal-to-noise ratio scenario, and the detection rate is statistically analyzed using CFAR and parametric detection methods. The detection performance verified by the Monte Carlo experiments is as follows: Figure 9 As shown. From Figure 9 It can be seen that in low-altitude detection scenarios, when the energy of static clutter is much greater than that of the target, the aircraft is in a low-speed flight state, and CFAR detection is almost ineffective.

[0055] from Figure 9 As can be seen, the parametric detection method used in this invention fully explores the physical characteristics of the signal, which can achieve higher detection results and verify the advanced nature of this invention.

[0056] This invention can perform high-accuracy detection of the takeoff / landing phase of an aircraft, and has better detection performance in low signal-to-noise ratio environments compared to traditional constant false alarm rate (CFAR) detection.

[0057] In summary, for the critical scenario of aircraft takeoff and landing, this invention provides a parameterized detection method for the takeoff and landing process of highly maneuverable low-speed targets based on sensing signals. First, a motion model of the aircraft's takeoff and landing phase is constructed to obtain relevant formula modeling results. Next, based on the formula modeling results, the distances of multiple suspected targets are estimated; the velocities of multiple suspected targets are estimated; and by designing a bandpass filter in the Doppler domain to obtain LVD results, the Doppler modulation frequency information of the target is calculated to estimate the acceleration of the moving target, obtaining the acceleration of multiple suspected targets. Furthermore, based on radar antenna design principles, the back reflection coefficients of multiple suspected targets are calculated; and based on the array antenna settings of the sensing system, the spatial position of the moving target is estimated, obtaining a beam domain spatial angle measurement result spectrum to obtain the spatial angles of multiple suspected targets. Combined with the distances of the multiple suspected targets, the spatial positions of the multiple suspected targets are determined, thus obtaining the parameters of multiple suspected targets. Finally, using a detection module based on the physical characteristics of the moving target, the aircraft target is selected from the multiple suspected targets according to preset threshold ranges for the target's velocity, acceleration, back reflection coefficient, and spatial position.

[0058] This method utilizes the parameterized physical characteristics of the target, employing a sensory system and parameterized detection algorithms for accurate detection during takeoff and landing, improving anti-interference capabilities in complex environments and enabling reliable detection of low-altitude, slow-moving, highly maneuvering targets. This invention addresses the problem of reduced time-frequency focusing performance caused by clutter masking by designing a bandpass filter in the Doppler domain to first separate the target signal and then perform time-frequency analysis. When multiple targets or interference sources exist in the detection airspace, the system needs to be able to distinguish them. This invention solves the problem of multi-target discrimination by combining range dimension resolution (moving target range estimation in step S2) and spatial angle resolution (BS-MUSIC algorithm in step S6). The detection range is optimized using an integrated sensory system, and the detection effect of low-speed, highly maneuvering targets in low signal-to-noise ratio environments is optimized using parameterized detection algorithms that utilize range, velocity, acceleration, back reflection system, and spatial position information in the echo.

[0059] When making target decisions based on acceleration, setting the acceleration threshold too low will generate a large number of false alarms (mistaking clutter for targets), while setting it too high will reduce the detection rate (missing real targets). This invention clarifies this contradiction and, through experimental testing, adjusts the appropriate acceleration threshold, ultimately providing a framework for controlling false alarms and the detection rate using this threshold, thus solving the trade-off problem of detector performance in practical applications.

[0060] Secondly, corresponding to the above method embodiments, this invention also provides a parameterized detection device for the take-off and landing process of a highly maneuverable, low-speed target based on sensor signals, targeting the interaction scenario between the sensor system and the aircraft target, such as... Figure 10 As shown, the device includes: The motion model building module is used to build motion models of aircraft during takeoff and landing, and obtain relevant formula modeling results. The target distance estimation module is used to estimate the distance of moving targets based on the modeling results of the formula, and obtain the distances of multiple suspicious targets; The target velocity estimation module is used to estimate the velocity of moving targets based on the modeling results of the formula, thereby obtaining the velocities of multiple suspected targets. The target acceleration estimation module is used to estimate the acceleration of moving targets based on the modeling results of the formula, obtain LVD results by designing a bandpass filter in the Doppler domain, calculate the Doppler modulation frequency information of the target, and obtain the acceleration of multiple suspicious targets. The back reflection coefficient calculation module is used to calculate the back reflection coefficient of multiple suspicious targets based on the radar antenna design principle. The spatial position estimation module is used to estimate the spatial position of moving targets based on the array antenna settings of the sensing system, obtain the beam domain spatial angle measurement result spectrum, obtain the spatial angles of multiple suspicious targets, and determine the spatial positions of multiple suspicious targets by combining the distances of the multiple suspicious targets. The target screening module is used to filter out aircraft targets from multiple suspicious targets by using the detection module based on the physical characteristics of moving targets and according to preset threshold ranges for the target's speed, acceleration, back reflection coefficient and spatial position.

[0061] For details on the specific processing procedures of each module of the device, please refer to the relevant content in the first section, which will not be repeated here.

[0062] This invention employs an integrated sensing system and a parameterized detection algorithm to detect the takeoff and landing phases, resulting in stronger detection capabilities compared to traditional constant false alarm rate (CFAR) detection methods. Furthermore, this invention utilizes a high-density sensing system deployed in urban areas and describes signal motion parameters, avoiding detection blind spots in traditional methods and overcoming the shortcomings of existing methods in utilizing signal motion parameter information. This enhances the detection capability of low-speed, highly maneuverable moving targets in low signal-to-noise ratio environments.

[0063] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention are included within the scope of protection of the present invention.

Claims

1. A parameterized detection method for the takeoff and landing process of a highly maneuverable, low-speed target based on sensor signals, characterized in that, For interaction scenarios between a sensing system and an aircraft target, the method includes: S1, Construct a motion model of the aircraft during takeoff and landing, and obtain the relevant formula modeling results; S2, Based on the modeling results of the formula, the distance to the moving target is estimated to obtain the distances to multiple suspicious targets; S3, based on the modeling results of the formula, estimate the velocity of the moving target to obtain the velocity of multiple suspicious targets; S4. Based on the modeling results of the formula, LVD results are obtained by designing a bandpass filter in the Doppler domain, and the Doppler modulation frequency information of the target is calculated to estimate the acceleration of the moving target and obtain the acceleration of multiple suspicious targets. S5, based on radar antenna design principles, calculates the back reflection coefficients of multiple suspicious targets; S6. Based on the array antenna configuration of the sensing system, the spatial position of the moving target is estimated to obtain the beam domain spatial angle measurement result spectrum, the spatial angles of multiple suspicious targets are obtained, and the spatial positions of multiple suspicious targets are determined by combining the distances of the multiple suspicious targets. S7 utilizes a detection module based on the physical characteristics of moving targets to filter out aircraft targets from multiple suspicious targets based on preset threshold ranges for the target's speed, acceleration, back reflection coefficient, and spatial position.

2. The method according to claim 1, characterized in that, S1 includes: S11, establish the instantaneous expression for the distance between the aircraft target and the sensing system, expressed as: ； Wherein, it is assumed that the station height of the sensor system is , The speed of light is [value], and the horizontal distance between the spacecraft and the sensing system is [value]. The aircraft uses speed as acceleration is Perform uniformly accelerated takeoff motion; For direction, slow time, The pitch angle of the aircraft target relative to the sensor system; S12, based on the instantaneous expression for the distance between the aircraft target and the sensing system, the expression for the echo signal received by the sensing system after being reflected by the aircraft target is obtained, expressed as: ； In the aforementioned sensing system, each pulse of the transmitted waveform consists of multiple complete OFDM symbols, and returns to the sensing system after being reflected from the target. The total time is... ; The target's back reflection coefficient, For signal transmission bandwidth, Indicates distance in advance; Using distance to fast time The bandwidth of a signal transmission is used to represent the signal's transmission bandwidth, which is a property of... The function; The wavelength of the signal electromagnetic wave; The imaginary unit; S13, by receiving the expression for the echo signal reflected by the target from the aircraft through the sensing system, the expression for the Doppler frequency of the target is obtained, which is expressed as: ； in, Indicates the Doppler frequency of the target; echo signal The overall phase; S14, further differentiate the expression for the Doppler frequency of the target to obtain the expression for the Doppler modulation frequency information of the target, which is expressed as: ； in, This indicates the Doppler modulation frequency information of the target.

3. The method according to claim 2, characterized in that, S2 includes: Based on the expression for the echo signal received by the sensing system after reflection from the aircraft target, the focusing of the actual echo is analyzed to determine the time delay. Value, based on each determined delay The value is used to obtain the distance of the focal point relative to the sensor system, and the distance of each suspicious target to the sensor system; wherein, the calculation formula is: ； in, The distance from the suspected target to the sensing system; after the echo signal is discretized and sampled, each sampling point corresponds to a distance value. .

4. The method according to claim 3, characterized in that, S3 includes: S31, based on the expression of the Doppler frequency of the target, determine the target velocity information contained in the Doppler domain corresponding to the azimuth slow time, and determine the frequency domain display form of the echo signal based on the target Doppler effect; In the frequency domain, the echo signal is represented as follows: the Doppler frequency center is at... Expanded to ; S32, based on the frequency domain representation of the echo signal and the expression for the Doppler frequency of the target, the velocity range of each suspected target is estimated, and the velocity of the suspected target is obtained by averaging. The calculation formula used is as follows: ； in, The number of grids to widen for a suspected target; For a certain suspicious target The Doppler frequency corresponding to the grid expansion; the number of grids for target expansion is determined based on the number of sampling points occupied by the suspected target during echo signal analysis, and the number of grids for expansion varies for different suspected targets.

5. The method according to claim 4, characterized in that, S4 includes: S41 will transmit the echo signal Representing the orientation in a slower time dimension, we obtain: ； in, The result is a slow-time dimension representation of the echo signal azimuth. On the middle molecule This is the time delay transformation factor. The pulse repetition interval is the time in the denominator. This represents the total observation time of the synsensory system. The number of signals in the echo; Indicates a suspicious target The target echo energy of the grid; For a certain suspicious target The Doppler frequency corresponding to the lattice broadening; For a suspicious target The Doppler modulation frequency corresponding to the lattice broadening; S42, In the Doppler domain, the azimuth slow-time dimension representation of the echo signal is filtered using a mid-pass filter and subjected to an inverse Fourier transform to obtain the transformed echo signal, represented as: ； in, The azimuth slow time after scale transformation Doppler frequencies after scaling transformation following FFT transformation; This indicates the center frequency location of the pass filter; This refers to the bandwidth of the mid-pass filter. Indicates Fourier transform; Indicates the inverse Fourier transform; This represents the total echo energy of the signal passing through the center-pass filter; This represents the calculated Doppler frequency of the target; This indicates the Doppler modulation information of the target; S43, Perform LVD analysis on the transformed echo signal to obtain the autocorrelation function, expressed as: ； in, As the scale transformation factor, using Direction to slower time Scale conversion factor Transformed into azimuth slow time after scale transformation With lag time The function; S44, the autocorrelation function along and Fourier transform is performed in two dimensions to achieve energy focusing and parameter estimation of the target, yielding the corresponding LVD result, represented as: ； in, This represents the corresponding LVD result, where, The azimuth slow time after scale transformation Doppler frequencies after scaling via FFT transformation The Doppler modulation frequency after scaling; Let be the impulse function; S45, based on the obtained LVD results and the calculated... And the expression for the Doppler modulation frequency information of the target, calculate and utilize Calculate the acceleration of the corresponding suspicious target.

6. The method according to claim 5, characterized in that, S5 includes: S51, based on radar principles, determines the distance to the sensing system. The power density irradiated by the target is expressed as: ； in, The power of the transmitter in the sensing system. For the transmission gain of the sensing system, The distance between the transmitting antenna of the sensing system and the target point; S52, after being reflected by the target, the echo signal spreads omnidirectionally. Assume the target's rearward reflection coefficient is... The distance between the target and the receiving antenna is The power density of the target echo at the receiver of the sensing system is determined as follows: ； S53, when the effective area of ​​the receiving antenna is The echo power received by the inductive system is determined and expressed as: ； S54, using antenna theory, determines the receiving antenna gain. With the effective area of ​​the receiving antenna The following relationship exists: ； in, The wavelength of the signal electromagnetic wave; S55, based on the derivation formulas in S51~S54, the expression for the received echo power of the inductive system is determined as follows: ； S55, through echo signal analysis Then, based on the distance estimation of the moving target, the following is obtained: and And based on the fixed parameter , , , Using the expression for the received echo power of the aforementioned sensing system, the back reflection coefficient of each suspected target is calculated. .

7. The method according to claim 6, characterized in that, S6 includes: S61, based on the array antenna configuration of the sensing system, determine the steering vector caused by the target on each antenna in various directions, expressed as: ； Wherein, the azimuth angle between the point target and the sensing system is The pitch angle is The pitch array spacing of the sensing system is The horizontal array spacing is Assuming the sensor system has azimuth capability... Root antenna, The CCP Multiply the formulas together; S62, assuming the sensing system has in the pitch direction The steering vector caused by the target on each elevation antenna is expressed as: ； Wherein, it is assumed that the sensing system has in the pitch direction Root antenna, The CCP Multiply the formulas together; S63, considering the combined effects of elevation and azimuth angles, the echoes from each antenna are represented as follows: ； in, The echo vector is composed of the echo signals from each antenna in a single transmission. For the echo data of each antenna, the subscript is... Indicates the direction to the first Root antenna, Indicates pitch to the first Root antenna; Let be the guiding vector of the spatial array; where: ； in, , ; S64, by utilizing the phase difference of the target in the array antenna caused by the spatial position information, the spatial position of the target is calculated by the BS-MUSIC angle measurement algorithm in the beam domain space, and the beam domain space angle measurement result spectrum is obtained. S65, based on the beam domain spatial angle measurement result spectrum, obtain the spatial angles of multiple suspicious targets, and combine the distances of the multiple suspicious targets to determine the spatial positions of the multiple suspicious targets.

8. The method according to claim 7, characterized in that, S64 includes: S641, based on the Discrete Fourier Transform beamforming method, is a beamforming method with low computational cost, performing two-dimensional beamforming, wherein the formed two-dimensional beam is represented as: ； in, Echo data representing each antenna; S642, with The main lobe of each beam points to , The main lobe of each beam points to Beamformer, definition Beamforming matrix, yes dimensional vector, containing The main lobe of each beam points to The beam formed by the beamformer yes dimensional vector, containing The main lobe of each beam points to The beam formed by the beamformer; one dimension corresponds to the beam pointing, and the other dimension corresponds to the beam weights of different antennas; where the beamforming matrix is ​​represented as: ； in, for The form of expression; ; S643, the beamspace expression after beamspace transformation is determined as follows: ； in, It is a beam domain spatial signal, which is a fast time representation of the range axis of a two-dimensional beam; S644, For the beamspace expression after beamspace transformation, calculate the covariance matrix, which is expressed as: ； in, This is the autocorrelation matrix in the beam space; This is the autocorrelation matrix after spatial transformation; for The corresponding eigenvector matrix; Let be the covariance matrix of the beam domain spatial signal; This is the sum of squares of the mean values ​​of the row vectors of beam spatial information; This is the spatial transformation matrix after eigenvalue decomposition; S645 decomposes the beam space signal into a noise subspace. The expression for the beam domain spatial angle measurement result spectrum is as follows: ； in, This is the spatial transformation vector after eigenvalue decomposition.

9. The method according to any one of claims 1-7, characterized in that, S7 includes: For any suspicious target, the data set obtained includes its distance, speed, acceleration, back reflection coefficient, and spatial position. It is determined whether all of these fall within the corresponding threshold range. If so, the suspicious target is identified as an aircraft target.

10. A parameterized detection device for the takeoff and landing process of a highly maneuverable, low-speed target based on sensor signals, characterized in that, For interaction scenarios between a sensing system and an aircraft target, the device includes: The motion model building module is used to build motion models of aircraft during takeoff and landing, and obtain relevant formula modeling results. The target distance estimation module is used to estimate the distance of moving targets based on the modeling results of the formula, and obtain the distances of multiple suspicious targets; The target velocity estimation module is used to estimate the velocity of moving targets based on the modeling results of the formula, thereby obtaining the velocities of multiple suspected targets. The target acceleration estimation module is used to estimate the acceleration of moving targets based on the modeling results of the formula, obtain LVD results by designing a bandpass filter in the Doppler domain, calculate the Doppler modulation frequency information of the target, and obtain the acceleration of multiple suspicious targets. The back reflection coefficient calculation module is used to calculate the back reflection coefficient of multiple suspicious targets based on the radar antenna design principle. The spatial position estimation module is used to estimate the spatial position of moving targets based on the array antenna settings of the sensing system, obtain the beam domain spatial angle measurement result spectrum, obtain the spatial angles of multiple suspicious targets, and determine the spatial positions of multiple suspicious targets by combining the distances of the multiple suspicious targets. The target screening module is used to filter out aircraft targets from multiple suspicious targets by using the detection module based on the physical characteristics of moving targets and according to preset threshold ranges for the target's speed, acceleration, back reflection coefficient and spatial position.

Images