Geosynchronous orbit synthetic aperture radar (RD) marine moving target RD positioning method based on time sequence information

Abstract

The invention belongs to the technical field of synthetic aperture radars, and particularly relates to a geosynchronous orbit synthetic aperture radar sea moving target RD positioning method based on time sequence information, which comprises the following steps of: firstly, establishing a radar-target motion model based on position vectors and velocity vectors of a satellite and a target; utilizing the radar-target motion model to calculate azimuth position offset caused by target motion speed; secondly, establishing a time sequence RD object image relation model according to the azimuth position offset, and obtaining a time sequence RD equation based on the relation model; and finally, aiming at the time sequence RD equation, calculating the position speed of the target by adopting weighted least square estimation.

Description

Technical Field

[0001] The present invention belongs to the technical field of synthetic aperture radars, and in particular relates to a geosynchronous orbit synthetic aperture radar (RD) positioning method for marine moving targets based on time series information. Background Art

[0002] Target positioning, also known as radar photography, is the foundation of synthetic aperture radar (SAR) imaging measurements and a key branch of radar photography. It plays a crucial role in military reconnaissance, mapping, and high-value target location. Its basic principle is to accurately calculate the target's position coordinates using pixel coordinates in SAR images and information such as the attitude and geometry of radar transmission and reception, based on established positioning models and solution methods. There are generally three SAR target positioning methods: stereo positioning, interferometry positioning, and range-Doppler (RD) positioning. Stereo positioning and interferometry have higher requirements for the measurement system, making RD positioning more universal than the first two methods.

[0003] Traditional RD positioning methods use the time delay and Doppler information of the echo contained in SAR images to locate the target, but they are mostly limited to locating stationary targets on the ground. It is impossible to effectively extract the position and velocity information of moving targets in a single image, so it is difficult to locate moving targets.

[0004] Spaceborne synthetic aperture radars (SAR) can be categorized by orbital altitude into low Earth Orbit (LEO SAR), medium Earth Orbit (MEO SAR), and geosynchronous SAR (GEO SAR). High-Earth Orbit (GEO SAR), operating at an altitude of approximately 36,000 km, offers short revisit periods, slow flight speeds, large slant ranges, wide beam footprints, and the ability to maintain a long-term fixed-point observation of targets. Therefore, GEO SAR's exceptionally long observation times can be leveraged to build upon traditional RD methods by introducing the time dimension and combining it with existing range-Doppler information to calculate the position and velocity parameters of moving targets, ultimately enabling GEO SAR to locate moving targets.

[0005] However, due to the change in the Doppler center frequency of a moving target, the target becomes defocused in the image domain, resulting in degraded resolution. This also results in azimuth offsets in each frame, affecting the accuracy of RD positioning. Therefore, by utilizing two-frame and multi-frame target RD sequences, the timing information is decoupled from the target position and velocity information, allowing the position and velocity parameters of the moving target to be jointly estimated.

[0006] To locate moving targets by introducing temporal information, it is necessary to first establish an object-image relationship model for a multi-frame RD sequence. Furthermore, the range-Doppler relationship of the target signal must be analyzed based on the orbital motion characteristics of the GEO SAR and the target's trajectory. Factors affecting positioning accuracy primarily include image resolution and the accuracy of the target's radial velocity estimation. Furthermore, for moving targets at sea, the constraints imposed by background sea clutter on subframe accumulation time must also be considered. Therefore, joint optimization of resolution and frame sequence based on positioning accuracy is essential. Summary of the Invention

[0007] To address these issues, the present invention provides a time-series-based geostationary orbit synthetic aperture radar (SAR) positioning method for moving targets at sea. This method establishes a time-series RD object-image relationship model by incorporating GEO SAR data, a target motion model, and image domain parameters from a multi-frame sequence. This model then uses the RD equation to calculate the target's actual velocity parameters. This method enables precise positioning of moving targets and expands the applicability and performance of RD positioning methods.

[0008] The technical solutions for implementing the present invention are as follows:

[0009] A method for locating moving targets at sea using a geosynchronous orbit synthetic aperture radar (SAR) based on time series information includes the following steps:

[0010] First, a radar-target motion model is established based on the position vectors and velocity vectors of the satellite and the target, and the azimuth position offset caused by the target motion velocity is calculated using the radar-target motion model;

[0011] Secondly, a time series RD object-image relationship model is established according to the azimuth position offset, and a time series RD equation is obtained based on the relationship model;

[0012] Finally, the position and velocity of the target are solved by using weighted least squares estimation for the time series RD equation.

[0013] Optionally, the radar-target motion model of the present invention is:

[0014]

[0015] Among them, S0 represents the GEO SAR position vector at the initial moment, v S represents the GEO SAR velocity vector, a S represents the GEOSAR acceleration vector, T0 represents the target position vector, v T represents the velocity vector of the target, a T represents the acceleration vector of the target, and t represents the azimuth time.

[0016] Optionally, the azimuth position offset caused by the target motion speed of the present invention is:

[0017]

[0018] in, Indicates the target movement speed at t a The radial projection size of the azimuth time, R0 represents the reference slant distance of pulse compression, v S represents the GEO SAR velocity vector, v T Represents the velocity vector of the target.

[0019] Optionally, the azimuth position offset in the present invention is used to establish a time-series RD object-image relationship model:

[0020]

[0021] Among them, θ(·,·) means finding the projection of the first vector on the second vector, x k Indicates the azimuth direction vector of the k-th frame sequence, x indicates the azimuth coordinate axis direction in the positioning coordinate system, and R indicates the distance coordinate axis direction in the positioning coordinate system. Indicates projecting the offset of the k-th frame sequence to the final positioning coordinate system direction, Indicates the azimuth position offset of the k-th frame sequence, k = 0, 1, 2...N, where N is the number of frames involved in the calculation.

[0022] Optionally, the time series RD equation of the present invention is:

[0023]

[0024] in, Indicates the azimuth coordinates and range coordinates of the target in the k-th frame image, t k represents the azimuth time of the center moment of the kth frame, Represents the distance equation of the target at the center of the k-th frame image.

[0025] Optionally, the position velocity parameter β to be estimated in the present invention is:

[0026] β=(H T C L -1 H) -1 H T C L -1 L

[0027] Among them, C L is the covariance matrix of the image positioning error, L is the observation quantity, and H is the observation matrix.

[0028] Optionally, the present invention further comprises the following steps:

[0029] With comprehensive positioning accuracy σ β The resolution level of the subframe image is set as the target of the signal-to-clutter ratio (SCR) of ship target imaging detection. The number of frames N involved in the calculation is the optimization parameter, and the multi-objective optimization genetic algorithm is used to achieve joint optimization.

[0030] Optionally, the comprehensive positioning accuracy σ of the present invention β for:

[0031] σ β =diag(H T C L -1 H) -1 .

[0032] Optionally, the signal-to-noise ratio expression of the present invention is:

[0033]

[0034] Among them, σ t represents the target RCS within the unit resolution unit, σ c represents the clutter RCS within the unit resolution unit, and C represents the clutter coherence factor.

[0035] Optionally, in the present invention, if the clutter is fully coherent, then C=1; if the clutter is non-coherent, then C=1 / N, where N represents the number of accumulated pulses.

[0036] Beneficial effects:

[0037] The present invention introduces time series information (i.e., multi-frame data) into the traditional RD positioning model, proposes a series positioning model and estimation method, and realizes accurate estimation of velocity parameters of moving targets and image position correction by GEO SAR. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0039] Figure 1 This is a schematic diagram of the overall process of the method proposed in the present invention;

[0040] Figure 2 This is a schematic diagram of the GEO SAR system positioning the ship;

[0041] Figure 3Flowchart for positioning accuracy optimization based on multi-objective optimization genetic algorithm.

[0042] Figure 4 Schematic diagram of the multi-frame RD sequence positioning model;

[0043] Figure 5 Point target positioning result. DETAILED DESCRIPTION

[0044] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0045] It should be noted that, in the absence of conflict, the following embodiments and features in the embodiments may be combined with each other; and, based on the embodiments in this disclosure, all other embodiments obtained by persons of ordinary skill in the art without creative work are within the scope of protection of this disclosure.

[0046] It should be noted that various aspects of the embodiments within the scope of the appended claims are described below. It should be apparent that the aspects described herein can be embodied in a wide variety of forms, and any specific structure and / or function described herein is merely illustrative. Based on this disclosure, it should be understood by those skilled in the art that an aspect described herein can be implemented independently of any other aspect, and two or more of these aspects can be combined in various ways. For example, any number of aspects described herein can be used to implement an apparatus and / or practice a method. In addition, other structures and / or functionalities other than one or more of the aspects described herein can be used to implement this apparatus and / or practice this method.

[0047] The specific processing flow of the present invention is as follows Figure 1 As shown. Based on the radar-target motion model, the RD object-image relationship model is derived, and then the time-series RD equation is solved. A multi-frame time-series RD positioning method is proposed, and finally the comprehensive positioning accuracy is improved through a multi-objective optimization algorithm. This method solves the positioning parameters of the moving target through a time-series multi-frame RD sequence. The traditional RD equation cannot solve the positioning problem of moving targets. This method estimates the motion parameters of the target by introducing time dimension information, and finally obtains the estimated results of the positioning parameters. The factors affecting positioning accuracy are mainly the accumulation time and the subframe sequence. The accumulation time further affects the signal-to-noise ratio in the target detection domain and the resolution in the image domain. Therefore, it can be modeled as a multi-objective optimization problem to further optimize the comprehensive positioning accuracy of the target.

[0048] The simulation input of this method is Figure 1 The GEO SAR orbital parameters S and motion parameters v shown in S As well as the target position parameter T and motion parameter v T, the output is the optimized subframe division and accumulation time, which meets the target detection signal-to-noise ratio requirements and the optimal positioning accuracy σ β The main parameters in the verification process are: GEO SAR orbit parameters and radar system parameters, a hypothetical moving target, and the output is the estimated target motion and position parameters, i.e. the positioning result. The specific steps are as follows:

[0049] Step 1: Radar-target motion model establishment: Based on the GEO SAR and target motion imaging relationship, establish the GEOSAR-motion model and target motion model. The specific models include: the position vectors S, T of the satellite and the target, the velocity vector v of the satellite and the target S ,v T ;

[0050] The radar-target motion model can be expressed as:

[0051]

[0052] Among them, S0 represents the GEO SAR position vector at the initial moment, v S represents the GEO SAR velocity vector (first-order slant range term), a S represents the GEO SAR acceleration vector (slant range second-order term), T0 represents the position vector of the target, v T represents the velocity vector of the target, a T represents the acceleration vector of the target, t represents the azimuth time, and the above position and velocity model can be expanded to higher-order terms as needed.

[0053] Step 2: Based on the radar-target motion model and the basic principle framework of RD positioning established above, the characteristics of the range Doppler information change caused by the target motion are analyzed.

[0054] Based on the radar-target motion model and the basic principle framework of RD positioning established above, the variation characteristics of range Doppler information caused by target motion are analyzed.

[0055] Taking a simple linear motion scenario as an example (considering only the first-order slant range term), define the unit slant range vector in the range direction as R and the unit vector in the azimuth direction as x. The range condition equation from the GEO SAR to the target can be expressed as:

[0056] R ta ||=||ST||=R0+M range ·R T (2)

[0057] Among them, R tarepresents the slant distance vector from the target T to the antenna phase center S at the imaging center moment. The second term in the equation is the slant distance equation at the imaging center moment. The third term in the equation represents the target slant distance information in the image domain. R0 represents the reference slant distance of pulse compression. M range It represents the distance sampling interval, R T Indicates the range coordinate of the moving target.

[0058] Correspondingly, according to the relationship between the radar platform speed and the target slant range vector, the target Doppler shift condition equation can be derived as follows:

[0059]

[0060] Where λ is the wavelength of the electromagnetic wave, f dc is the Doppler center frequency.

[0061] Since the SAR imaging principle assumes the target is stationary, it is necessary to consider the impact of target motion on the signal slant range history based on the traditional imaging model. The azimuth signal of the target can be expressed as:

[0062]

[0063] Among them, n0 is a unit vector, indicating the slant range direction. The second and third rows of the above formula represent the phase terms introduced by the target motion speed. The term will cause the target Doppler center frequency to shift, resulting in azimuth position shift. The two terms represent the residual frequency modulation phase generated by the target motion speed, which causes the target to be defocused in the image domain.

[0064] In summary, within the subsequence, the azimuth position offset caused by the target movement speed Its size can be expressed as:

[0065]

[0066] in, Indicates the target movement speed at t a The radial projection size of the azimuth moment, with the value being positive away from the radial direction.

[0067] Step 3: Based on the azimuth position offset model caused by the target motion speed, establish an RD object-image relationship model as shown below:

[0068]

[0069] Among them, θ(·,·) means finding the projection of the first vector on the second vector, x krepresents the azimuth direction vector (zero Doppler direction) of the k-th frame sequence, the x vector without a superscript represents the azimuth coordinate axis direction in the previously defined positioning coordinate system, and R represents the range coordinate axis direction in the positioning coordinate system. It means projecting the offset of the k-th frame sequence to the direction of the final positioning coordinate system, so that the azimuth offset of the k-th frame sequence acts on the x and R dimensions of the positioning coordinate system. Indicates the azimuth position offset caused by the target motion speed in the k-th frame sequence, Represents the real coordinates of the moving target in the kth frame sequence, k = 0, 1, 2...N, N is the number of frames involved in the calculation.

[0070] Step 4: Based on the object-image relationship model under the multi-frame sequence, a time series RD equation is established to solve the target position and speed parameters.

[0071]

[0072] in, Indicates the azimuth coordinates and range coordinates of the target in the k-th frame image, t k represents the azimuth time of the center moment of the kth frame, Represents the distance equation of the target at the center of the k-th frame image.

[0073] Taking a two-frame RD sequence as an example, the temporal RD positioning equation can be degenerated into:

[0074]

[0075] In order to achieve object-image positioning matching, it is necessary to estimate the target position velocity parameters in the positioning equation. Only the estimation problem of the (x, R) two-dimensional plane is considered, and the positioning error caused by the pulse pressure in the range direction is ignored. Therefore, 3 parameters to be estimated.

[0076] By solving the above equation using the least squares estimation (LS), we can obtain the vectorized expression of the observation equation:

[0077]

[0078] Among them, there is a corresponding approximate relationship: t2-t1=Δt. In the above formula, L is the observation quantity, H is the observation matrix, and β is the position and velocity parameter to be estimated.

[0079] Step 5: Considering the influence of the image envelope on the accuracy of the target position parameter solution, the weighted least squares estimation (WLS) method is used to solve the vector equation:

[0080] β=(H T C L -1 H)-1 H T C L -1 L (9)

[0081] Among them, C L is the covariance matrix of the image positioning error. Assuming that the GEO SAR subsequences are independent of each other, all non-diagonal elements are 0, C L It can be expressed as:

[0082]

[0083] In the example of two frames, C L It can be expressed as:

[0084]

[0085] in, Indicates the azimuth resolution of the first frame image. Indicates the distance resolution of the first frame image. and Same thing.

[0086] The standard deviation vector (RMSE) of the least squares estimate of the velocity parameter (i.e., positioning accuracy) can be expressed as:

[0087] σ β =diag(H T C L -1 H) -1 (11)

[0088] Step 6: Comprehensive positioning accuracy σ β The resolution level of the subframe image is set as the target of the signal-to-clutter ratio (SCR) of ship target imaging detection. The required number of subframes N is taken as the optimization parameters, and the multi-objective optimization genetic algorithm is used to achieve joint optimization.

[0089] The position and velocity parameter estimation has been completed by executing steps 1 to 5 above. In order to ensure that the estimated positioning accuracy and signal-to-noise ratio meet the requirements, this step is used to optimize the optimization parameters to ensure that the estimation result after step 6 is better.

[0090] Signal-to-noise ratio expression:

[0091]

[0092] Among them, σ t represents the target RCS within the unit resolution unit, σ crepresents the RCS of the clutter within the unit resolution unit, C represents the clutter coherence factor, if the clutter is fully coherent, C = 1, if the clutter is non-coherent, C = 1 / N, and N represents the number of accumulated pulses.

[0093] Normalized backscattering coefficient modeling of sea surface: The NRL clutter model is used as an example to represent the average backscattering coefficient. The model inputs frequency, ground-grazing angle, and sea state parameters to simulate the sea clutter radar cross section (NRCS) (also known as σ0), such as Figure 2 shown.

[0094] The multi-objective optimization genetic algorithm is used to solve the optimal positioning parameters. The multi-objective optimization genetic algorithm flow chart is as follows: Figure 3 shown.

[0095] According to the frame number division parameters obtained by the optimized configuration, the imaging simulation of the moving target is carried out and the velocity parameters are estimated.

[0096] At this point, all steps are completed.

[0097] Next, an implementation example is given with specific parameters.

[0098] In this example, a set of GEO SAR and moving target parameters are given as shown in Table 1. Considering the L-band system, the signal-to-clutter ratio, and the accumulation time of each frame is 60 seconds, two frames are used as an example to solve the positioning parameters.

[0099] According to the parameters in Table 1, the imaging resolution can be obtained as 30m*93m. Figure 4 shown.

[0100] Table 1 GEO SAR and moving target parameters

[0101]

[0102]

[0103] Table 2 Positioning parameter calculation accuracy

[0104] Azimuth positioning deviation 3.02km Distance positioning deviation 30m Azimuth velocity estimation bias 0.0045m / s Range velocity estimation bias 0.066m / s

[0105] The final positioning result is as follows Figure 5 shown.

[0106] It is particularly important to point out that this target positioning method essentially leverages the extremely long integration time of geosynchronous orbit synthetic aperture radar (GEO SAR) to solve the problem of locating moving targets at sea. Numerical optimization methods are used to address target positioning accuracy and target detection in the presence of sea clutter. Theoretically, this method can also leverage GEO SAR's multi-frame temporal information to address the defocusing issue during the imaging of moving targets. This requires expanding the target and radar motion parameters to higher-order terms based on the example order, and using multi-frame sequences to estimate higher-order phase terms in the imaging, thereby further improving resolution and positioning accuracy. Furthermore, based on the concept of this method, distributed geosynchronous orbit SAR systems can be used in the future for multi-angle joint positioning, addressing the problem of high-precision monitoring of moving targets at sea using distributed GEO SAR.

[0107] Of course, the present invention may have many other cases. Without departing from the spirit and essence of the present invention, those skilled in the art may make various corresponding changes and modifications based on the present invention, but these corresponding changes and modifications should all fall within the scope of protection of the claims attached to the present invention.

[0108] In summary, the above are only preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for locating moving targets at sea using geosynchronous orbit synthetic aperture radar (RD) based on time series information, characterized in that: The process includes the following: First, a radar-target motion model is established based on the position vectors and velocity vectors of the satellite and the target, and the azimuth position offset caused by the target motion velocity is calculated using the radar-target motion model; Secondly, a time series RD object-image relationship model is established according to the azimuth position offset, and a time series RD equation is obtained based on the relationship model; Finally, the position and velocity of the target are solved by using weighted least squares estimation for the time series RD equation.

2. The method for locating moving targets at sea using geosynchronous orbit synthetic aperture radar (RD) based on time series information according to claim 1, characterized in that: The radar-target motion model is: Among them, S0 represents the GEO SAR position vector at the initial moment, v S represents the GEO SAR velocity vector, a S represents the GEO SAR acceleration vector, T0 represents the position vector of the target, v T represents the velocity vector of the target, a T represents the acceleration vector of the target, and t represents the azimuth time.

3. The method for locating moving targets at sea using geosynchronous orbit synthetic aperture radar (RD) based on time series information according to claim 1, characterized in that: The azimuth position offset caused by the target motion speed is: in, Indicates the target movement speed at t a The radial projection size of the azimuth time, R0 represents the reference slant distance of pulse compression, v S represents the GEO SAR velocity vector, v T Represents the velocity vector of the target.

4. The method for locating moving targets at sea using geosynchronous orbit synthetic aperture radar (RD) based on time series information according to claim 3, characterized in that: The azimuth position offset establishes a time series RD object-image relationship model as follows: Among them, θ(·,·) means finding the projection of the first vector on the second vector, x k Indicates the azimuth direction vector of the k-th frame sequence, x indicates the azimuth coordinate axis direction in the positioning coordinate system, and R indicates the distance coordinate axis direction in the positioning coordinate system. Indicates projecting the offset of the k-th frame sequence to the final positioning coordinate system direction, Indicates the azimuth position offset of the k-th frame sequence, k = 0, 1, 2...N, where N is the number of frames involved in the calculation.

5. The method for locating moving targets at sea using geosynchronous orbit synthetic aperture radar (RD) based on time series information according to claim 4, characterized in that: The time series RD equation is: in, Indicates the azimuth coordinates and range coordinates of the target in the k-th frame image, t k represents the azimuth time of the center moment of the kth frame, Represents the distance equation of the target at the center of the k-th frame image.

6. The method for locating moving targets at sea using geosynchronous orbit synthetic aperture radar (RD) based on time series information according to claim 1, characterized in that: The position velocity parameter β to be estimated is: β=(H T C L -1 H) -1 H T C L -1 L Among them, C L is the covariance matrix of the image positioning error, L is the observation quantity, and H is the observation matrix.

7. The method for locating moving targets at sea using geosynchronous orbit synthetic aperture radar (RD) based on time series information according to claim 1, characterized in that: The following steps are also included: With comprehensive positioning accuracy σ β The resolution level of the subframe image is set as the target of the signal-to-clutter ratio (SCR) of ship target imaging detection. The number of frames N involved in the calculation is the optimization parameter, and the multi-objective optimization genetic algorithm is used to achieve joint optimization.

8. The method for locating moving targets at sea using geosynchronous orbit synthetic aperture radar (RD) based on time series information according to claim 7, characterized in that: The comprehensive positioning accuracy σ β for: σ β =diag(H T C L -1 H) -1 。 9. The method for locating moving targets at sea using geosynchronous orbit synthetic aperture radar (RD) based on time series information according to claim 7, characterized in that: The signal-to-noise ratio expression is: Among them, σ t represents the target RCS within the unit resolution unit, σ c represents the clutter RCS within the unit resolution unit, and C represents the clutter coherence factor.

10. The method for locating moving targets at sea using geosynchronous orbit synthetic aperture radar (RD) based on time series information according to claim 9, characterized in that: If the clutter is fully coherent, C=1; if the clutter is non-coherent, C=1 / N, where N represents the number of accumulated pulses, i.e., the number of frames involved in the calculation.

Images