Star-ground bistatic radar target track-before-detect method based on greedy strategy

Abstract

The invention discloses a satellite-ground bistatic radar moving target tracking-before-detection method based on a greedy strategy, and the method comprises the steps: receiving a direct wave transmitted by a satellite and a moving target reflection echo, and generating a stacked distance-Doppler graph; constructing a candidate plot set and generating a storage vector for units on the stacked distance-Doppler maps exceeding a first-level threshold; a cumulative function and a track set of all trace point vectors in the first candidate trace point set are initialized; traversing all trace point vectors in each candidate trace point set, generating an association set through storage vectors and target kinematics constraint characteristics, and determining a trace point vector which is associated with the current trace point vector and is located in the next candidate trace point set from the association set by utilizing a greedy strategy, calculating an accumulation function and a track set of the associated trace point vectors; and traversing all trace point vectors in the last candidate trace point set, and if the cumulative function value of the current trace point vector exceeds a secondary threshold, judging that the moving target is detected.

Description

Technical Field

[0001] This invention belongs to the field of bistatic radar target detection technology, specifically, it relates to a satellite-to-ground bistatic radar target tracking method based on a greedy strategy. Background Technology

[0002] Satellite-transmitted signals are generally unrestricted by location or weather, offering wide coverage and enabling the construction of bistatic satellite-ground radar systems with separate transmitters and receivers for target detection. However, due to the limited transmission power of satellites, the target echo energy collected by the receiving station is very weak, making it difficult to detect directly from background noise, thus limiting the radar's effective range. Existing techniques typically employ strategies to increase observation time to enhance target echo energy, while simultaneously addressing range migration and Doppler frequency migration caused by target movement over long observation periods, thereby effectively improving radar target detection performance.

[0003] However, most existing satellite-to-ground bistatic radar target detection technologies consider simple detection scenarios with a single target moving at a constant speed. When faced with complex detection scenarios with multiple targets and maneuvering motion, the detection performance of existing technologies will be affected, or even rendered unusable. Summary of the Invention

[0004] To address the problems existing in the prior art, this invention provides a target detection and tracking method based on a greedy strategy for satellite-ground bistatic radar, which can realize the detection of weak moving targets in scenarios with multiple moving targets and maneuvering or non-maneuvering motion.

[0005] To achieve the above technical objectives, the present invention adopts the following technical solution: a target tracking method for satellite-to-ground bistatic radar based on a greedy strategy, comprising the following steps: Step S1: Select the satellite that constitutes the backscattering geometry as the signal source. The ground receiving station simultaneously receives the direct wave emitted by the satellite and the reflected echo from the moving target during the observation time. Divide the observation time into equal time periods. In each time period, use the direct wave to process the reflected echo from the moving target to generate a stacked distance-Doppler image. Step S2: Set a first-level threshold, and use the stacked distance-Doppler graph cells that exceed the first-level threshold to construct a candidate trace set and generate a storage vector; Step S3: Initialize the cumulative function and track set of all track vectors in the first candidate track set; Step S4: Traverse all point vectors in each candidate point set, generate an association set by storing vectors and target kinematic constraint characteristics, use a greedy strategy to determine the point vectors in the next candidate point set that are associated with the current point vector from the association set, and recursively calculate the cumulative function and track set of the associated point vectors. Step S5: Set a secondary threshold, traverse all point vectors in the last candidate point vector set, if the cumulative function value of the current point vector exceeds the secondary threshold, determine that the moving target has been detected, and obtain the motion parameter data of the moving target throughout the observation time through the track set of the current point vector.

[0006] Furthermore, the distance-Doppler plot of the stack is represented as... ,in, Indicates the distance between a satellite and a ground-based bibase. , This represents the equivalent pulse repetition period of the satellite signal. Represents the speed of light; Indicates the Doppler frequency. ; Indicates frequency modulation. , and These represent the lower and upper bounds of the frequency modulation sampling interval, respectively, and the sampling step size of the frequency modulation sampling interval. No more than , Indicates the duration of a single frame. Indicates the length of the time period.

[0007] Further, step S2 includes the following sub-steps: Step S2.1: Set a first-level threshold, traverse each cell in the distance-Doppler map of each stack, and if the signal amplitude in the cell exceeds the first-level threshold, put the satellite-to-ground bistatic distance, Doppler frequency, frequency modulation and signal amplitude contained in the cell into the candidate trace vector of the distance-Doppler map of the corresponding stack. Step S2.2: For each candidate point trace set, assign consecutive index numbers to all point trace vectors in the candidate point trace set; Step S2.3: Record the total number of point vectors with the same modulation frequency in the candidate point set into the storage vector.

[0008] Furthermore, the initialization process for the cumulative function of all trace vectors in the first candidate trace set and the track set is as follows:

[0009] in, This represents the index number within the first candidate point set. , This represents the total number of point vectors within the first candidate point set. Indicates that the index number of the first candidate point set is i The set of tracks corresponding to the point trace vectors, Indicates that the index number of the first candidate point set is i The signal amplitude corresponding to the trace vector. Indicates that the index number of the first candidate point set is i The cumulative function corresponding to the trace vector of the point.

[0010] Furthermore, step S4 includes the following sub-steps: Step S4.1: Traverse the first... n Given a set of candidate point vectors, determine whether the current point vector is a newly emerging moving target. Step S4.2: If so, estimate the moving target's position in the current point trace vector, using the current point trace vector's modulation frequency as the center. The frequency modulation range within the segment; otherwise, estimate the moving target's position in the current segment using the current point's track vector's track set. The range of frequency modulation variation within the segment; Step S4.3: Place the first In the candidate trace set, all trace vectors whose frequency modulation values ​​fall within the range of frequency modulation variation estimated in step S4.2 participate in the kinematic correlation calculation of the current trace vector and generate a correlation set using the stored vector; Step S4.4: Use a greedy strategy to determine the node located at the th position that is associated with the current point trace vector from the association set. The point vectors of the candidate point set are used to recursively calculate the cumulative function of the associated point vectors and the track set; Step S4.5: Repeat steps S4.1-S4.4 to complete the calculation of the cumulative function and the track set.

[0011] Furthermore, the specific process of determining whether the current point trace vector is a newly generated moving target in step S4.1 is as follows: If the first... n The index number within the candidate point set is i The trace vector Collection of flight tracks If the number of non-zero index numbers is greater than 1, the target is determined to be a non-newly emerging moving target; otherwise, it is determined to be a newly emerging moving target. , , and They represent the first n The index number within the candidate point set is The point trace vector contains the star-to-ground bistatic distance, Doppler frequency, frequency modulation, and signal amplitude.

[0012] Furthermore, in step S4.2, if it is a newly generated target, the current point trace vector is used. The frequency modulation is used to estimate the moving target in the first quarter. The frequency modulation range within the segment is ,in, Indicates the moving target in the 1st... Frequency modulation within the segment, This indicates that the setting is based on the frequency of the current point trace vector. The lower limit, , and These represent the lower and upper bounds of the frequency modulation sampling interval, respectively. This indicates that the setting is based on the frequency of the current point trace vector. The upper limit, .

[0013] Furthermore, in step S4.2, if the target is not a newborn target, it is determined by the current point trace vector. Collection of flight tracks Estimate the moving target in the 1st month The frequency modulation range within the segment is ,in, This represents the estimation of the set of tracks using the current point's track vector. The lower limit, , This indicates the offset of the frequency modulation sampling unit. This indicates the sampling step size of the frequency modulation sampling interval. Indicates the first The index number in the candidate point set is The difference in frequency between the current point trace vector and the current point trace vector; This represents the estimation of the set of tracks using the current point's track vector. The upper limit, .

[0014] Furthermore, the specific process of step S4.3 is as follows: i. Determine the first [step] through the storage vector. The index number of all point vectors in the candidate point set whose frequency modulation value falls within the estimated frequency modulation variation range. :

[0015] in, and They represent the first The minimum and maximum indexes of all point vectors in the candidate point set whose frequency modulation values ​​fall within the estimated frequency modulation variation range. Indicates the first The th storage vector The element stores the first element. Among the candidate point sets, the one with the frequency modulation sampling interval is the The total number of trace vectors for each tuning frequency value express The sampling index number in the frequency modulation sampling interval. ; express The sampling index number in the frequency modulation sampling interval. ; ii. According to In the Search for the vector that matches the current point from the set of candidate points. The trace vectors of the kinematic constraints form an associated set. :

[0016]

[0017]

[0018] in, , , They represent the first The index number in the candidate point set is The trace vector The satellite-to-ground bistatic distance, Doppler frequency, and frequency modulation. and These represent the bistatic distance and the Doppler frequency sampling cell offset, respectively. and These represent the sampling resolution of the bistatic distance and the Doppler frequency in the range-Doppler plot, respectively. express The expected center that satisfies the distance constraint. ; express The desired center that satisfies the Doppler frequency constraint. ; express and The maximum absolute value of the difference between them , Indicates the center wavelength of the satellite signal carrier. and These represent the upper limits of velocity and acceleration of a moving target, respectively. Indicates the length of a period of time.

[0019] Further, step S4.4 includes the following sub-steps: Step S4.4.1: If the associative set Not empty, follow the greedy strategy from The index number of the trace vector with the largest signal amplitude value is selected internally. , will the The index number in the candidate point set is The trace vector With the The index number in the candidate point set is i The trace vector Establish a connection; Step S4.4.2: Traverse the first... The set of associated vectors of all candidate point traces; Step S4.4.3: Traverse the first... If all point vectors in the candidate point vector set are associated with the current point vector and are located at the th , then... If the number of point vectors in a candidate point set exceeds 0, a greedy strategy is used to select the point vector with the largest cumulative function value to recursively calculate the cumulative function and track set of the current point vector; otherwise, the cumulative function and track set of the current point vector are directly initialized.

[0020] Compared with existing technologies, the present invention has the following advantages: The target tracking method for satellite-to-ground bistatic radar based on a greedy strategy achieves effective accumulation of energy for multiple targets through stacked range-Doppler maps and allows the actual motion of targets to deviate from the assumed motion model, enabling satellite-to-ground bistatic radar systems to effectively detect multiple maneuvering and non-maneuvering weak targets. Compared with existing methods, the motion parameter estimation error of the present invention is the same. However, the present invention makes full use of the kinematic constraints of targets in adjacent time periods and the hierarchical characteristics of stacked range-Doppler maps to significantly reduce the computational load of the association set, thereby effectively improving computational efficiency. Attached Figure Description

[0021] Figure 1 This is a flowchart of the target tracking method for satellite-to-ground bistatic radar based on a greedy strategy according to the present invention; Figure 2 This is a schematic diagram of the bistatic geometric configuration for target detection according to the present invention; Figure 3 A schematic diagram of the stacking distance-Doppler plot; Figure 4 This is a schematic top view of the simulation experiment scene of the present invention; Figure 5 This is a schematic diagram of the cumulative function during the simulation experiment of this invention, wherein, Figure 5 (a) in the figure is a graph of the cumulative function values ​​of all point vectors in the first candidate point set during the initialization of the method of the present invention; Figure 5(b) in the figure is a graph of the cumulative function values ​​of all point vectors in the fifth candidate point set after the fourth recursive accumulation of the method of the present invention; Figure 5 (c) in the figure is a graph of the cumulative function values ​​of all point vectors in the tenth candidate point set after the ninth recursive accumulation of the method of the present invention; Figure 6 The target motion parameters output by the simulation experiment of this invention are as follows, Figure 6 (a) in the figure is a bistatic distance-Doppler frequency trace plot of three moving targets in the simulation experiment. Figure 6 (b) in the figure is the estimated frequency modulation curve of the three moving targets in the simulation experiment. Detailed Implementation

[0022] The technical solution of the present invention will be further explained and described below with reference to the accompanying drawings.

[0023] like Figure 1 This is a flowchart of the target tracking method for satellite-to-ground bistatic radar based on a greedy strategy according to the present invention. The target tracking method for satellite-to-ground bistatic radar includes the following steps: Step S1: As Figure 2 As shown, a satellite constituting the backscattering geometry is selected as the signal source, and the observation time at the ground receiving station is... Simultaneously receiving direct waves transmitted by satellites and reflected echoes from moving targets, a stacked range-Doppler map is generated; including the following sub-steps: Step S1.1: Observation time Divided into N A time interval of equal length, the length of the time interval Referring to the mobile target detection method of a GNSS-R passive radar moving target detection method and system, which was filed on September 1, 2023, with publication number CN117214852A, the direct wave is purified in each time period, and then the purified direct wave and the reflected echo of the moving target are used for range compression. Step S1.2: Divide the time period into several frames to overcome the decorrelation problem of target echo scattering characteristics within the segment, and correct the range migration and Doppler frequency migration of moving target reflected echoes within and between frames by searching for frequency modulation. Step S1.3: Coherent accumulation is performed within a frame using Fast Fourier Transform, while incoherent accumulation is performed between frames. Finally, Generate time period The stacked distance-Doppler plot, such as Figure 3 The stacked distance-Doppler plot is represented as ,in, Indicates the distance between a satellite and a ground-based bibase. , This represents the equivalent pulse repetition period of the satellite signal. Represents the speed of light; Indicates the Doppler frequency. ; Indicates frequency modulation. , and These represent the lower and upper bounds of the frequency modulation sampling interval, respectively, and the sampling step size of the frequency modulation sampling interval. No more than , Indicates the duration of a single frame. Indicates the length of the time period.

[0024] Step S2: To reduce subsequent computation, a first-level threshold is set, and cells on the stacked distance-Doppler map exceeding the first-level threshold are used to construct a candidate trace set and generate a storage vector; this includes the following sub-steps: Step S2.1: Set a lower first-level threshold, traverse each cell in the distance-Doppler map of each stack, and if the signal amplitude in the cell exceeds the first-level threshold, put the satellite-to-ground bistatic distance, Doppler frequency, frequency modulation and signal amplitude contained in the cell into the candidate trace vector of the distance-Doppler map of the corresponding stack.

[0025] Step S2.2: For each candidate point track set, assign consecutive index numbers to all point track vectors in the candidate point track set; specifically, assign index numbers to all point track vectors in the candidate point track set in ascending order of their modulation frequency values; for point track vectors with the same modulation frequency value, the index numbers can be assigned according to the principle of prioritizing bistatic distance value or Doppler frequency value. After index number assignment, the... The set of candidate points is represented as ,in, It is an integer. Indicates the first The index number within the candidate point set is The trace vector of the point, , , and These represent index numbers respectively. The bistatic distance, Doppler frequency, frequency modulation rate, and signal amplitude of the trace vector. Indicates the first The total number of point vectors in the candidate point set.

[0026] Step S2.3: Record the total number of point vectors with the same modulation frequency in the candidate point set into the storage vector, then the first... Each storage vector is represented as ,in, Indicates the first Among the candidate point sets, the one with the frequency modulation sampling interval is the The total number of trace vectors for each tuning frequency value This represents the total number of samples within the frequency modulation sampling interval. Ultimately, Distance-Doppler map generation of stacked units A set of candidate points and One storage vector, Step S3: Initialize the cumulative function and track set for all track vectors in the first candidate track set:

[0027] in, This represents the index number within the first candidate point set. , This represents the total number of point vectors within the first candidate point set. Indicates that the index number of the first candidate point set is i The set of tracks corresponding to the point trace vectors, Indicates that the index number of the first candidate point set is i The signal amplitude corresponding to the trace vector. Indicates that the index number of the first candidate point set is i The cumulative function corresponding to the trace vector of the point.

[0028] Step S4: Traverse all point vectors in each candidate point set, generate an association set by storing vectors and target kinematic constraints, and use a greedy strategy to determine the point vectors in the next candidate point set that are associated with the current point vector from the association set. Recursively calculate the cumulative function and track set of the associated point vectors, including the following sub-steps: Step S4.1: Traverse the first... n A set of candidate points For all point trace vectors within the range, determine whether the current point trace vector is a newly generated target. Specifically, if the first... n The index number within the candidate point set is i The trace vector Collection of flight tracks If the number of non-zero index numbers is greater than 1, it is determined to be a non-new target; otherwise, it is determined to be a new target.

[0029] Step S4.2: If it is a newly generated target, estimate the moving target's position in the current trace vector using the current frequency of the target's trajectory vector as the center. The range of frequency modulation (FM) changes within a segment is used to roughly estimate the range of FM changes for the target in the next time segment; otherwise, the range of FM changes for the moving target in the next time segment is estimated using the track set of the current point's track vector. The frequency modulation range within a segment can be estimated relatively accurately for non-new targets in the next time segment based on historical information. In one technical solution of the present invention, since the maneuverability of targets on land or sea is relatively slow, if it is a newly created target, the current point vector is used. The potential target represented by the first Move the segment to the first The frequency modulation value will not change much after the segment, therefore, the current point trace vector The frequency modulation is used to estimate the moving target in the first quarter. The frequency modulation range within the segment is ,in, Indicates the moving target in the 1st... Frequency modulation within the segment, This indicates estimation centered on the frequency modulation of the current point trace vector. The lower limit, ; This indicates estimation centered on the frequency modulation of the current point trace vector. The upper limit, .

[0030] In one technical solution of the present invention, if the target is not a newly generated target, the current point trace vector is used. The trajectory set estimates the moving target in the first... The specific process for varying the frequency modulation range within a segment is as follows: extract... The last two non-zero index numbers, i.e., the first two non-zero index numbers. The and the first The index number, because the first index number, The index number is index number Therefore, only the first one is recorded. The index number is Then, find The index number is The trace vector of the point is denoted as ,calculate and The frequency modulation difference, Because surface or sea targets are relatively slow to maneuver, it can be estimated that... The potential target represented in the first Frequency variation range within the segment ,in, This represents the estimation of the set of tracks using the current point's track vector. The lower limit, , This represents the offset of the frequency modulation sampling unit, thus allowing the motion of the real target to deviate from the assumed motion model. This indicates the sampling step size of the frequency modulation sampling interval. This represents the estimation of the set of tracks using the current point's track vector. The upper limit, .

[0031] Step S4.3: Place the first All point vectors whose frequency modulation values ​​fall within the estimated frequency modulation variation range in the candidate point vector set participate in the current point vector. The kinematic correlations are calculated and correlation sets are generated; specifically: i. By storing vectors Determine the first The index number of all point vectors in the candidate point set whose frequency modulation value falls within the estimated frequency modulation variation range. The index range of all point trace vectors within the estimated frequency modulation range can be quickly determined through simple calculation using the following formula:

[0032] in, and They represent the first The minimum and maximum indexes of all point vectors in the candidate point set whose frequency modulation values ​​fall within the estimated frequency modulation variation range. Indicates the first The th storage vector The element stores the first element. In the set of candidate points, the frequency modulation value is the first frequency modulation sampling interval. The total number of all trace vectors of a given frequency. express The sampling index number in the frequency modulation sampling interval. ; express The sampling index number in the frequency modulation sampling interval. ; ii. According to In the Searching for matching candidate points in a set of candidate points The trace vectors of the kinematic constraints form an associated set. :

[0033]

[0034]

[0035] in, , , They represent the first The index number in the candidate point set is The trace vector The satellite-to-ground bistatic distance, Doppler frequency, and frequency modulation. and These represent the bistatic distance and the Doppler frequency sampling cell offset, respectively, thus allowing the motion of the real target to deviate from the assumed motion model. and Let represent the sampling resolution of the bistatic distance and Doppler frequency in the range-Doppler image, respectively. Since the maneuverability of targets on land or sea is relatively slow, the target motion within a segment can be assumed using a second-order polynomial motion model, the mathematical expression of which is: , Indicates the time within a segment. and These represent the bistatic distance and Doppler frequency of the target at the reference time within the segment, respectively, with the midpoint of the segment used as the reference time. express The expected center that satisfies the distance constraint. ; express The desired center that satisfies the Doppler frequency constraint. ; express and The maximum absolute value of the difference between them , Indicates the center wavelength of the satellite signal carrier. and These represent the upper limits of velocity and acceleration of a moving target, respectively.

[0036] Step S4.4: Use a greedy strategy to determine the node located at the th position that is associated with the current point trace vector from the association set. The process involves recursively calculating the cumulative function of the associated point vectors and the track set from the set of candidate point vectors, including the following sub-steps: Step S4.4.1: If the associative set Not empty, follow the greedy strategy from The index number of the trace vector with the largest signal amplitude value is selected internally. , will the The index number in the candidate point set is The trace vector With the The index number in the candidate point set is i The trace vector Related; however, Possibly with Multiple point trace vectors within a given area are associated; therefore, a set is used. Record with Related The vector index of the point within is represented as: ; Step S4.4.2: Traverse the first... A set of candidate points The set of associations of all trace vectors; Step S4.4.3: Traverse the first... Let all point vectors within the candidate point set be... express The index number within, if it is related to the current point trace vector Related to the first If the number of point vectors in a candidate point set exceeds 0, a greedy strategy is used to select from the index set. The current point vector is recursively calculated by selecting the point vector with the largest cumulative function value from the associated point vectors. The cumulative function and the set of tracks are represented as follows: in, Indicates that the record is in Internal matters The index number; otherwise, directly initialize the cumulative function and track set of the current point's track vector, represented as .

[0037] Step S4.5: Repeat steps S4.1-S4.4 until all point vectors in the last candidate point track set have completed the calculation of the cumulative function and track set.

[0038] Step S5: Set a higher secondary threshold, traverse all point vectors in the last candidate point vector set, if the cumulative function value of the current point vector exceeds the secondary threshold, determine that the moving target has been detected, and obtain the motion parameter data of the moving target throughout the observation time through the track set of the current point vector; otherwise, the existence of the moving target has not been detected.

[0039] The following simulation experiment verifies the target detection and tracking method for satellite-ground bistatic radar based on a greedy strategy of the present invention. The simulation scenario is as follows: Figure 4As shown, the relevant parameters of the simulation experiment are described as follows: the total observation time is 120s, the duration of each segment is 12s, the total number of segments is 10, the duration of a single frame within a segment is 3s, and the total number of frames within a segment is 4; the L1 signal transmitted by the GPS satellite is selected as the signal source, and the satellite trajectory data is derived from real satellite ephemeris data. During the entire observation time, the satellite's elevation angle and azimuth angle vary from 71.97° to 70.97° and 147.38° to 148.55°, respectively. The receiving station is deployed at a location [0, -1000, 50]m; Figure 4 There are three moving targets in the XY plane. Target 1 is performing a maneuver, with an initial position of [-400, 500] m, a velocity vector of [8.66, 5] m / s, and an acceleration vector of [0, -0.1] m / s². 2 Targets 2 and 3 move at uniform speeds, with initial positions of [400, 500] m and [400, 200] m respectively, and velocity vectors of [-10.39, -6] m / s and [-8, 0] m / s respectively. The signal-to-noise ratio of the echo signals from all three targets reaching the receiving station is -65 dB. The receiver's sampling frequency is 16.368 MHz, the frequency modulation sampling range is [-1, 1] Hz / s, and the total number of sampling points is 72. The state transition numbers for bistatic distance, Doppler frequency, and frequency modulation are 4, 4, and 10 respectively. The upper limits of target velocity and acceleration are 15 m / s and 0.5 m / s² respectively. 2 .

[0040] Figure 5 (a)-(c) illustrate the cumulative function values ​​of all trace vectors within the candidate trace set during the recursive process of the satellite-to-ground bistatic radar target detection pre-tracking method of the present invention. Figure 5 As shown in (a), during the initialization phase, the cumulative function values ​​of all trace vectors in the first candidate trace set are very similar. At this point, it is difficult to distinguish the three target trace vectors from a large number of noisy trace vectors. As the recursive accumulation proceeds, from Figure 5 In (b) and (c), three target point trace vectors, distinct from the noise point trace vectors, can be clearly seen. (Comparison) Figure 5 From (b) and (c), we can see that Figure 5 In (c), the cumulative function values ​​of the three target point trace vectors differ more from the cumulative function values ​​of other noise point trace vectors, which is more conducive to detecting the presence of targets and reducing the false alarm rate. Once the target is detected, the motion parameter data during the observation time can be obtained from the track set of the target point trace vectors.

[0041] from Figure 6 As can be seen in (a) and (b), the 10 sets of measured bistatic distance-Doppler frequency traces and estimated frequency values ​​for the three targets are very close to their true values. Figure 5 and Figure 6 The experimental results demonstrate the effectiveness of the method of the present invention in detecting multiple maneuvering and non-maneuvering targets and the accuracy of estimating motion parameters.

[0042] To further verify the superiority of the target tracking method for satellite-ground bistatic radar detection in this invention, Table 1 compares the root mean square error and computation time of the target motion parameter estimation method of this invention with that of the prior art "Santi F, Pastina D, Bucciarelli M. Experimental demonstration of ship target detection in GNSS-based passive radar combining target motion compensation and track-before-detect strategies[J]. Sensors,2020, 20(3): 599". As can be seen from Table 1, the target motion parameter estimation error of the method of this invention is the same as that of the prior art. However, the computation time of the method of this invention is only about 54% of that of the prior art.

[0043] Table 1: Comparison of the target tracking method before detection by the present invention with existing technologies in the satellite-ground bistatic radar.

[0044] In one technical solution of the present invention, a computer-readable storage medium is also provided, storing a computer program that enables a computer to execute the satellite-to-ground bistatic radar target detection and tracking method provided by the present invention.

[0045] In one technical solution of the present invention, an electronic device is also provided, including: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the satellite-to-ground bistatic radar target detection pre-tracking method provided by the present invention.

[0046] In the embodiments disclosed in this application, a computer storage medium may be a tangible medium that may contain or store programs for use by or in conjunction with an instruction execution system, apparatus, or device. The computer storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of computer storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0047] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed in this application can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0048] The above are merely preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should be considered within the scope of protection of the present invention.

Claims

1. A target tracking method for satellite-to-ground bistatic radar based on a greedy strategy, characterized in that, Includes the following steps: Step S1: Select the satellite that constitutes the backscattering geometry as the signal source. The ground receiving station simultaneously receives the direct wave emitted by the satellite and the reflected echo from the moving target during the observation time. Divide the observation time into equal time periods. In each time period, use the direct wave to process the reflected echo from the moving target to generate a stacked distance-Doppler image. Step S2: Set a first-level threshold, and use the stacked distance-Doppler graph cells that exceed the first-level threshold to construct a candidate point trace set and generate a storage vector; Step S3: Initialize the cumulative function and track set of all track vectors in the first candidate track set; Step S4: Traverse all point vectors in each candidate point set, generate an association set by storing vectors and target kinematic constraint characteristics, use a greedy strategy to determine the point vectors in the next candidate point set that are associated with the current point vector from the association set, and recursively calculate the cumulative function and track set of the associated point vectors. Step S5: Set a secondary threshold, traverse all point vectors in the last candidate point vector set, if the cumulative function value of the current point vector exceeds the secondary threshold, determine that the moving target has been detected, and obtain the motion parameter data of the moving target throughout the observation time through the track set of the current point vector.

2. The target detection and tracking method based on a greedy strategy for satellite-to-ground bistatic radar according to claim 1, characterized in that, The stacked distance-Doppler plot is represented as ,in, Indicates the distance between a satellite and a ground-based bibase. , This represents the equivalent pulse repetition period of the satellite signal. Represents the speed of light; Indicates the Doppler frequency. ; Indicates frequency modulation. , and These represent the lower and upper bounds of the frequency modulation sampling interval, respectively, and the sampling step size of the frequency modulation sampling interval. No more than , Indicates the duration of a single frame. Indicates the length of the time period.

3. The target detection and tracking method based on a greedy strategy for satellite-to-ground bistatic radar according to claim 1, characterized in that, Step S2 includes the following sub-steps: Step S2.1: Set a first-level threshold, traverse each cell in the distance-Doppler map of each stack, and if the signal amplitude in the cell exceeds the first-level threshold, put the satellite-to-ground bistatic distance, Doppler frequency, frequency modulation and signal amplitude contained in the cell into the candidate trace vector of the distance-Doppler map of the corresponding stack. Step S2.2: For each candidate point trace set, assign consecutive index numbers to all point trace vectors in the candidate point trace set; Step S2.3: Record the total number of point vectors with the same modulation frequency in the candidate point set into the storage vector.

4. The target detection and tracking method based on a greedy strategy for satellite-to-ground bistatic radar according to claim 1, characterized in that, The initialization process for the cumulative function of all trace vectors in the first candidate trace set and the track set is as follows: in, This represents the index number within the first candidate point set. , This represents the total number of point vectors within the first candidate point set. Indicates that the index number of the first candidate point set is i The set of tracks corresponding to the point trace vectors, Indicates that the index number of the first candidate point set is i The signal amplitude corresponding to the trace vector. Indicates that the index number of the first candidate point set is i The cumulative function corresponding to the trace vector.

5. The target detection and tracking method based on a greedy strategy for satellite-to-ground bistatic radar according to claim 1, characterized in that, Step S4 includes the following sub-steps: Step S4.1: Traverse the first... n Given a set of candidate point vectors, determine whether the current point vector is a newly emerging moving target. Step S4.2: If so, estimate the moving target's position in the current point trace vector, using the current point trace vector's modulation frequency as the center. The frequency modulation range within the segment; otherwise, estimate the moving target's position in the current segment using the current point's track vector's track set. The range of frequency modulation variation within the segment; Step S4.3: Place the first In the candidate trace set, all trace vectors whose frequency modulation values ​​fall within the range of frequency modulation variation estimated in step S4.2 participate in the kinematic correlation calculation of the current trace vector and generate a correlation set using the stored vector; Step S4.4: Use a greedy strategy to determine the node located at the th position that is associated with the current point trace vector from the association set. The point vectors of the candidate point set are used to recursively calculate the cumulative function of the associated point vectors and the track set; Step S4.5: Repeat steps S4.1-S4.4 to complete the calculation of the cumulative function and the track set.

6. The target detection and tracking method based on a greedy strategy for satellite-to-ground bistatic radar according to claim 5, characterized in that, The specific process for determining whether the current point trace vector is a newly generated moving target in step S4.1 is as follows: If the first... n The index number within the candidate point set is i The trace vector Collection of flight tracks If the number of non-zero index numbers is greater than 1, the target is determined to be a non-newly emerging moving target; otherwise, it is determined to be a newly emerging moving target. , , and They represent the first n The index number within the candidate point set is The point trace vector contains the star-to-ground bistatic distance, Doppler frequency, frequency modulation, and signal amplitude.

7. The satellite-to-ground bistatic radar target detection pre-tracking method based on a greedy strategy according to claim 6, characterized in that, If it is a newly generated target in step S4.2, use the current point trace vector. The frequency modulation is used to estimate the moving target in the first quarter. The frequency modulation range within the segment is ,in, Indicates the moving target in the 1st... Frequency modulation within the segment, This indicates that the setting is based on the frequency of the current point trace vector. The lower limit, , and These represent the lower and upper bounds of the frequency modulation sampling interval, respectively. This indicates that the setting is based on the frequency of the current point trace vector. The upper limit, .

8. A target detection and tracking method based on a greedy strategy for satellite-to-ground bistatic radar according to claim 6, characterized in that, In step S4.2, if the target is not a newborn target, it is determined by the current point trace vector. Collection of flight tracks Estimate the moving target in the 1st month The frequency modulation range within the segment is ,in, This represents the estimation of the set of tracks using the current point's track vector. The lower limit, , This indicates the offset of the frequency modulation sampling unit. This indicates the sampling step size of the frequency modulation sampling interval. Indicates the first The index number in the candidate point set is The difference in frequency between the current point trace vector and the current point trace vector; This represents the estimation of the set of tracks using the current point's track vector. The upper limit, .

9. A target tracking method for satellite-to-ground bistatic radar based on a greedy strategy according to claim 7 or 8, characterized in that, The specific process of step S4.3 is as follows: i. Determine the first [unit] through the storage vector. The index number of all point vectors in the candidate point set whose frequency modulation value falls within the estimated frequency modulation variation range. : in, and They represent the first The minimum and maximum indexes of all point vectors in the candidate point set whose frequency modulation values ​​fall within the estimated frequency modulation variation range. Indicates the first The th storage vector The element stores the first element. Among the candidate point sets, the one with the frequency modulation sampling interval is the The total number of trace vectors for each tuning frequency value express The sampling index number in the frequency modulation sampling interval. ; express The sampling index number in the frequency modulation sampling interval. ; ii. According to In the Search for the vector that matches the current point from the set of candidate points. The trace vectors of the kinematic constraints form an associated set. : in, , , They represent the first The index number in the candidate point set is The trace vector The satellite-to-ground bistatic distance, Doppler frequency, and frequency modulation. and These represent the bistatic distance and the Doppler frequency sampling cell offset, respectively. and These represent the sampling resolution of the bistatic distance and the Doppler frequency in the range-Doppler plot, respectively. express The expected center that satisfies the distance constraint. ; express The desired center that satisfies the Doppler frequency constraint. ; express and The maximum absolute value of the difference between them , Indicates the center wavelength of the satellite signal carrier. and These represent the upper limits of velocity and acceleration of a moving target, respectively. Indicates the length of time within a segment.

10. A target detection and tracking method based on a greedy strategy for satellite-to-ground bistatic radar according to claim 9, characterized in that, Step S4.4 includes the following sub-steps: Step S4.4.1: If the associative set Not empty, follow the greedy strategy from The index number of the trace vector with the largest signal amplitude value is selected internally. , will the The index number in the candidate point set is The trace vector With the The index number in the candidate point set is i The trace vector Establish a connection; Step S4.4.2: Traverse the first... The set of associated vectors of all candidate point traces; Step S4.4.3: Traverse the first... If all point vectors in the candidate point vector set are associated with the current point vector and are located at the th , then... If the number of point vectors in a candidate point set exceeds 0, a greedy strategy is used to select the point vector with the largest cumulative function value to recursively calculate the cumulative function and track set of the current point vector; otherwise, the cumulative function and track set of the current point vector are directly initialized.

Images