CN114910906A

CN114910906A - Bistatic ISAR sparse aperture maneuvering target imaging method and system

Info

Publication number: CN114910906A
Application number: CN202210475635.1A
Authority: CN
Inventors: 胡文华; 薛东方; 朱瀚神; 郭宝锋; 曾慧燕
Original assignee: Army Engineering University of PLA
Current assignee: Army Engineering University of PLA
Priority date: 2022-04-29
Filing date: 2022-04-29
Publication date: 2022-08-16

Abstract

The embodiment of the specification provides a bistatic ISAR sparse aperture maneuvering target imaging method and a bistatic ISAR sparse aperture maneuvering target imaging system, wherein the method comprises the steps of compressing acquired fundamental frequency echo pulses of a target to be measured to acquire distance frequency domain-azimuth time domain signals; establishing a Bi-ISAR sparse aperture maneuvering target echo signal model containing a translation error term according to the distance frequency domain-azimuth time domain signal; establishing a phase compensation term, reconstructing an image by using a 2D-JLBI algorithm, updating the phase compensation term based on the translation track parameter corresponding to the IC search optimal image, and solving a model to obtain and output a target reconstruction image based on a set convergence condition. The invention converts the translation compensation problem into the problem of compressed sensing two-dimensional joint sparse reconstruction image and maneuvering target motion parameter estimation, thereby not only avoiding a large amount of calculation of vectorization and row-by-row processing, but also fully utilizing the gain of echo two-dimensional compression, avoiding the loss of imaging performance and having strong robustness to noise.

Description

Bistatic ISAR sparse aperture maneuvering target imaging method and system

Technical Field

The document relates to the technical field of radar imaging analysis, in particular to a bistatic ISAR sparse aperture maneuvering target imaging method and system.

Background

Inverse Synthetic Aperture Radar (ISAR) can generate a fine two-dimensional reflectivity image of an observation target, and plays an important role in the fields of target tracking, identification and the like. Compared with the traditional single-base ISAR imaging, the bistatic (multi-base) radar adopts a mode of separating a transmitter and a receiver, so that the flexibility of the system is enhanced, the imaging probability is improved, in principle, ISAR realizes distance direction high resolution by performing pulse compression on each echo, and direction high resolution is realized by a virtual aperture formed by target motion.

Translation compensation is a key step of ISAR imaging, the existing ISAR sparse aperture imaging algorithm is mainly based on a compressed sensing theory, and the idea of reconstructing and imaging a sparse aperture signal by using a compressed sensing technology can be divided into two categories:

firstly, the distance direction and the azimuth direction are separately processed, and although the mode has high speed and good performance, the coupling of the distance dimension and the azimuth dimension is damaged, and the imaging performance is reduced to some extent; secondly, distance and azimuth coupling processing, wherein the solution of the two-dimensional coupling echo matrix reconstruction problem mainly comprises the following ideas:

vectorization processing, which has high reconstruction accuracy but is only suitable for the situation of low scene complexity, and when the imaging scene is complex, the calculated amount is large and the real-time performance is poor;

processing line by line and column by column, and a large amount of redundant processing exists in the mode, so that the calculation efficiency is low;

the block processing shortens the operation time by the mode of firstly reconstructing blocks and then synthesizing imaging, but causes certain imaging performance loss,

and fourthly, reconstructing by using a joint sparse reconstruction algorithm, but the performance of the general joint sparse algorithm is poorer under the condition of lower signal-to-noise ratio.

However, the imaging problem solved by the method is basically the assumption that translational compensation is completed, and the problem of translational compensation on the echo under the condition of not considering the sparse aperture is solved.

Disclosure of Invention

One or more embodiments of the present specification provide a bistatic ISAR sparse aperture maneuvering target imaging method, comprising the steps of:

compressing the obtained fundamental frequency echo pulse of the target to be detected to obtain a distance frequency domain-azimuth time domain signal; establishing a Bi-ISAR sparse aperture maneuvering target echo signal model containing a translation error term according to the distance frequency domain-azimuth time domain signal; establishing a phase compensation term, reconstructing an image by using a 2D-JLBI algorithm, updating the phase compensation term based on the translation track parameter corresponding to the IC search optimal image, and solving a model to obtain and output a target reconstruction image based on a set convergence condition.

One or more embodiments of the present specification provide a bistatic ISAR sparse aperture maneuvering target imaging system, based on a bistatic inverse synthetic aperture radar system, comprising

A data acquisition unit: the Bi-ISAR system performs pulse compression on each echo to obtain a distance frequency domain-azimuth time domain signal; obtaining an initial searching starting point of a target through narrow-band speed measurement;

an echo signal model construction unit: according to the distance frequency domain-azimuth time domain signal obtained by the data acquisition unit, a Bi-ISAR sparse aperture maneuvering target echo signal model containing a translation error term is established;

a model solving unit: establishing a phase compensation item, reconstructing an image by using a 2D-JLBI algorithm, updating the phase compensation item based on the translation track parameter corresponding to the IC search optimal image, and solving a model to obtain and output a target reconstruction image based on the set search range, step length and cycle time conditions.

One or more embodiments of the present specification provide a computer apparatus comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, the processor implementing the bistatic ISAR sparse aperture maneuvering target imaging method described above when executing the computer program.

One or more embodiments of the present specification provide a computer-readable storage medium storing a computer program which, when executed by a processor, implements the bistatic ISAR sparse aperture maneuvering target imaging method described above.

The method comprises the steps of firstly establishing a Bi-ISAR sparse aperture maneuvering target echo signal model containing a translation error term, using a quadratic polynomial to fit the movement locus of the maneuvering target, establishing a phase compensation term, reconstructing an image by using a 2D-JLBI algorithm, updating the phase compensation term based on the translation locus parameters corresponding to the IC search optimal image, converting the translation compensation problem into the problem of compressed sensing two-dimensional combined sparse reconstruction image and maneuvering target movement parameter estimation, avoiding a large amount of calculations of vectorization and row-by-row processing, fully utilizing the gain of echo two-dimensional compression, avoiding the loss of imaging performance and having strong robustness on noise. The problem of carry out translation compensation to the echo under the sparse aperture condition of not considering in the current bistatic ISAR sparse aperture maneuvering target imaging technique is solved.

Drawings

In order to more clearly illustrate one or more embodiments or prior art solutions of the present specification, the drawings that are needed in the description of the embodiments or prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments described in the present specification, and that other drawings can be obtained by those skilled in the art without inventive exercise.

FIG. 1 is a flow chart provided by the method of the present invention;

FIG. 2 is a schematic diagram of a Bi-ISAR imaging geometry system provided by the method of the present invention;

FIG. 3 is a flow chart of a Bi-ISAR sparse aperture maneuvering target translation compensation imaging algorithm provided by the method of the invention;

FIG. 4 shows a Bi-ISAR simulation scene and a scattering point model in a simulation experiment provided by the method of the present invention; wherein, the diagram (a) is a Bi-ISAR simulation scene; FIG. (b) is a scattering point model; graph (c) is a scattering point model;

FIG. 5 is an image of the imaging result obtained by respectively using the method, the 2D-FISTA algorithm and the 2D-SL0 algorithm under the condition of different aperture losses in the simulation experiment provided by the method of the present invention;

FIG. 6 is an image of the imaging result obtained by the method, the 2D-FISTA algorithm and the 2D-SL0 algorithm under different SNR conditions in the simulation experiment provided by the method of the present invention;

FIG. 7 is a schematic diagram of a system framework provided by the system of the present invention;

FIG. 8 is a block diagram of a computer apparatus according to the present invention.

Detailed Description

In order to make those skilled in the art better understand the technical solutions in one or more embodiments of the present disclosure, the technical solutions in one or more embodiments of the present disclosure will be clearly and completely described below with reference to the drawings in one or more embodiments of the present disclosure, and it is obvious that the described embodiments are only a part of the embodiments of the present disclosure, and not all embodiments. All other embodiments that can be derived by a person skilled in the art from one or more of the embodiments described herein without making any inventive step shall fall within the scope of protection of this document.

The invention is described in detail below with reference to specific embodiments and the accompanying drawings.

Method embodiment

According to an embodiment of the present invention, there is provided a bistatic ISAR sparse aperture maneuvering target imaging method, as shown in fig. 1, the bistatic ISAR sparse aperture maneuvering target imaging method provided by the present invention, the bistatic ISAR sparse aperture maneuvering target imaging method according to the embodiment of the present invention includes the steps of:

compressing the acquired fundamental frequency echo pulse of the target to be detected to acquire a distance frequency domain-azimuth time domain signal;

secondly, establishing a Bi-ISAR sparse aperture maneuvering target echo signal model containing a translation error term according to the distance frequency domain-azimuth time domain signal; fitting the motion trail of the maneuvering target by a quadratic polynomial to establish a phase compensation term;

and thirdly, reconstructing the image by using a 2D-JLBI algorithm, updating a phase compensation item based on the translational track parameter corresponding to the IC search optimal image, and solving a model to obtain and output a target reconstructed image based on the set search range, step length and cycle time conditions.

The method of this embodiment proposes a method of combining 2D-JLBI with Image Contrast (IC) based search to solve the problem. Firstly, a Bi-ISAR sparse aperture maneuvering target echo signal model containing a translation error term is established, a quadratic polynomial is used for fitting the movement locus of the maneuvering target, a phase compensation term is established, a 2D-JLBI algorithm is used for reconstructing an image, the phase compensation term is updated based on translation locus parameters corresponding to an IC search optimal image, the translation compensation problem is converted into the problem of compressed sensing two-dimensional combined sparse reconstruction image and maneuvering target movement parameter estimation, a large amount of calculation of vectorization and line-by-line processing is avoided, the gain of echo two-dimensional compression is fully utilized, the loss of imaging performance is avoided, and the robustness to noise is strong. The problem of carry out translation compensation to the echo under the condition of not considering sparse aperture in the current bistatic ISAR sparse aperture maneuvering target imaging technique is solved.

Preferably, in this embodiment, the implementation steps of the specific step one and the step two are specifically as follows:

firstly, as shown in fig. 2, a Bi-ISAR imaging geometry system established for the present embodiment, where Tr is a transmitting station, Re is a receiving station, L is a radar baseline length, and E is an equivalent single-base radar position; suppose the target maneuvers in space with velocity v and acceleration a. Imaging start time t ₀ With the target centroid at O and the dihedral angle at β ₀ Establishing a right-hand coordinate system xOy by taking the centroid of the target as an origin and the bisector of the ground angle as a y-axis, wherein the coordinate of the scattering point P in the coordinate system is (x) _p ,y _p ) OP length d and angle alpha with x-axis ₀ . At t _p At that moment, the target centroid shifts to O _p The point, coordinate system x 'Oy' is translated from coordinate system xOy by O _p As the origin, bistaticThe angular bisector is a v axis, and a right-hand coordinate system uO is established _p v, the coordinate of the scattering point P in the coordinate system is recorded as

O _p The included angle between the P axis and the u axis is alpha _m The equivalent rotation angle of the equivalent monostatic radar is theta _m 。

The following steps are completed based on the Bi-ISAR imaging geometric system:

s101, performing pulse compression on the acquired echoes each time to obtain a one-dimensional range profile of the echoes each time;

the Bi-ISAR of this embodiment transmits a Linear Frequency Modulation (LFM) signal, whose mathematical expression is as follows:

in the formula: a represents the backscattering amplitude, rect (-) represents a rectangular window function and

t denotes total time, PRT is pulse repetition period, t _m Where mrt represents a slow time,

denotes fast time, M ═ 1: M]M denotes the total number of pulses, T _p Is the pulse width, f _c Is the carrier frequency, mu is the tuning frequency.

The obtained target echo signal is down-converted to a fundamental frequency echo form, and pulse compression is performed as follows:

wherein K represents the number of scattering points, σ _k The backscattering coefficient corresponding to the k-th scattering point, a (-) represents the complex envelope of the signal, R _k (t _m ) Represents the k-th scattering point at a slow time t _m The sum of the distances between the two stations at the moment and the receiving and transmitting stations (the echo signal analysis adopts the assumption of 'go-stop'), and c is the propagation speed of the electromagnetic waves;

R _k (t _m ) The calculation is as follows:

wherein R is _T (t _m ) Representing the equivalent centre of rotation of the maneuvering target at t _m Instantaneous distance, beta, from time of day to transmitting and receiving stations _m Corresponds to t _m The angle of the double base at the moment of time,

indicating the distance change due to the rotation of the target.

S102, performing distance dimension compression on the one-dimensional range profile to obtain a distance frequency domain-azimuth time domain signal, which is specifically as follows:

in which A (f) is

Fourier transform of (3); performing matched filtering processing on the formula (4) and neglecting the imaging-independent items, and obtaining the signals of the azimuth time domain-distance frequency domain as follows:

s103, establishing a Bi-ISAR sparse aperture maneuvering target echo signal model containing a translation error term based on the signals of the azimuth time domain and the distance frequency domain, and concretely realizing the following steps:

in general, it is required that the Bi-ISAR imaging accumulation time is short, and in this case, it is considered that the target is uniformly accelerated and rotated at a small rotation angle during the observation period, and the following approximation can be made

cosθ

_m 1, and in addition, assume a dihedral angle β (t) during imaging _m ) Constant β, therefore, the backscatter signal of equation (5) can be written as:

discretizing the formula (6) to obtain,

wherein the content of the first and second substances,

representing the effect of the translational component of the object on the echo envelope, and in the second phase term,

the term does not change with the azimuth time and the distance frequency and can be regarded as a constant term to be ignored;

and

negligible in small angle imaging;

f _r ＝f _s n is the distance sampling frequency, f _s Is the sampling rate and N is the number of sampling points.

y _q ，x _p Respectively, after discretizing the scene, corresponding distance dimension and azimuth dimension imaging units, y _k The representation shows the kth scattering point.

Accordingly, formula (7) can be expressed as follows:

in the formula, parameter

Depending only on the target rotation parameter, the same for all scatter points. Integrating the echo to fast time to convert two-dimensional echo signal into a row of azimuth signal, combining with smoothing ₀ Reconstructing norm to determine gamma and

the approximation degree of (2) is determined by taking the value of gamma at the time of the minimum entropy as the estimated value

It can be seen from (8) that the distance coordinate of the target has a fourier transform pair relationship with the distance frequency, and the orientation coordinate has a matching fourier transform pair relationship with the orientation dimension time, which contains the parameter γ. Considering the relation among the random sparsity of direction, the maneuvering condition of the target and observation noise, echo signals, translation errors and imaging scenes, the Bi-ISAR sparse aperture maneuvering target echo signal model can be expressed in a matrix form as follows:

S＝E⊙(AXB)+N _o (9)

Matrix E:

Matrix A:

Matrix B:

wherein, < > indicates a Hadamard product, S ∈ C ^N×R An echo signal matrix representing a distance frequency domain-an azimuth time domain; e is an element ofC ^N×R Envelope walk and phase shift for echo caused by translational components representing the presence of a target, i.e.

R _T (t _n ) Obtaining the rough motion information of the target; considering the situation of random sparse azimuth, A belongs to C ^N×M A matching Fourier transform matrix containing a parameter gamma and corresponding to the azimuth dimension is represented; b is belonged to C ^R×R A Fourier transform matrix representing a distance dimension; x belongs to C ^M×R Representing a discretized imaged scene; n is a radical of _o ∈C ^N×R Representing observation noise; n, M and R respectively represent the number of azimuth sampling pulses, the number of scene azimuth units and the number of echo distance dimension samples.

Since the target has mobility, the translation error term about the target usually has a high-order term changing with slow time, and in order to establish a phase compensation term, the embodiment uses a quadratic polynomial to fit the motion track of the target along the baseline direction of the bistatic radar, i.e. the target is assumed to have the following motion states:

by combining the model of the formula (10) with the sparsity of the Bi-ISAR imaging scene, the translation error correction problem of the Bi-ISAR target is equivalent to solving the optimization problem shown as follows, namely, the optimization model is established under the condition of sparsest scene:

in the formula, | · the luminance | | _F Frobenius norm, | | · | | | torry representing matrix ₁ L representing a matrix ₁ And a norm, wherein a is a vector formed by fitting polynomial coefficients to the motion trajectory, and λ represents a regularization parameter of a sparse term and is used for controlling estimation accuracy.

In this embodiment, the optimization problem formula (11) is solved through step three, and specifically, the optimization problem is solved by using a method of alternately updating the imaging scene X and the polynomial coefficient a.

The updating of the imaging scene specifically comprises:

firstly, optimizing and updating an imaging scene, wherein the corresponding form is as follows:

at this time, the above equation may be regarded as a two-dimensional matrix sparsity optimization problem that converts the reconstruction imaging problem after the translation compensation into matrix X (i.e., imaging scene matrix), and this embodiment proposes to compensate and reconstruct the scene image in the frequency domain using a 2D-JLBI algorithm, calculate the contrast of the scene image, and store the contrast until the search ends in the set search range and step length loop with the search starting point, as follows.

In this embodiment, in the process of "residual back substitution", the 2D-JLBI algorithm estimates the stagnation step length of each iteration by using a kisking technique; adjusting a weight parameter eta to control the weight of the residual error and the measured value during the back substitution; the algorithm is improved in three ways of reducing the condition number of the perception matrix, so that the iteration times are greatly reduced, and therefore the improved 2D-JLBI algorithm of the embodiment solves (12) the iteration format:

in the above formula, R ^(k) Denotes the residual error, X, of each iteration ^(k+1) The result obtained for each iteration; v ^(k+1) Is an intermediate variable; x ⁽⁰⁾ ＝V ⁽⁰⁾ ＝0；A ^* ＝A ^H (AA ^H ) ^-1 ，B ^* ＝(B ^H B) ^-1 B ^H ；

In this embodiment, by using the above improved 2D-JLBI algorithm, the 2D-JLBI algorithm is implemented as follows:

initialization: selecting proper parameters delta and eta, and initializing X ⁽⁰⁾ ＝V ⁽⁰⁾ ＝0；

The input matrix is a matrix of a, B,

setting the maximum iteration number k _max The parameter mu;

and (3) solving by loop iteration: fork is 0: k _max ；

S201, residual error R is updated ^(k) ；

S202, updating the intermediate variable V ^(k+1) ；

S203, updating the imaging scene X ^(k+1) ；

S204, updating S ^(k+1) ；

S205, judging whether the algorithm is converged, and if the algorithm is converged, jumping out of a loop;

s206, finally outputting the imaging scene

In this embodiment, the quality of the reconstructed scene image mainly depends on the structure of the phase compensation term, and whether the phase compensation term is accurately determined by the estimation of the translational track parameter. That is to say, the accuracy of the translation trajectory parameters directly affects the final imaging effect, that is, the problem of ensuring the image to be optimal can be finally converted into the problem of estimating the motion parameters of the maneuvering target, and the corresponding optimization problem is as follows:

for the optimization problem, an analytic solution cannot be obtained, and the estimated value of the translation trajectory parameter is updated in an IC search-based manner in the present embodiment.

For intuitive measurement of imaging quality, the concept of image contrast, defined as image amplitude, is introduced

The ratio of the standard deviation to the mean of (a) is:

wherein the content of the first and second substances,

is the image complex amplitude and a (-) is the averaging operation. In the established scattering point model, the larger the IC is, the larger the amplitude value at the k-th scattering point is, the larger the amplitude value is than the image mean value, and the image quality is higher and the focusing degree is good at the moment; when the IC is small, it indicates that the amplitude value at the k-th scattering point is closer to the average value of the image, and the image is blurred and has a poor focusing degree. Therefore, the IC function can be used to evaluate the quality of the image.

In addition, in order to better acquire the motion information of the maneuvering target to establish the compensation item, the present embodiment needs to acquire the speed information of the rough target as an initial value. The Bi-ISAR generally uses alternate transmission of wide and narrow band signals as transmission signals, determines the properties of a target by transmitting the wide band signals, improves the information perception, identifies the classified target, simultaneously transmits the narrow band signals by a radar, obtains the distance information of the target by transmitting the narrow band signals, and performs curve fitting on the distance information to obtain the initial estimation of the speed of the target. However, the distance resolution of the narrowband signal is poor, so the accuracy of the obtained speed information is not very high, and further accurate estimation of the speed parameter needs to be completed, that is, in this document, the motion trajectory parameter is updated, a reconstructed image is obtained, and it is assumed that the target rough speed information obtained by the narrowband signal is a ₀ Specifically, the updating process is shown in fig. 3, and specifically includes the following steps:

s301, initializing a contrast matrix Ψ to 0 according to the set search range θ and the step α, and setting an initial value a ₀ As a center, determining a search start point

S302, establishing a phase compensation term

S303, reconstructing an imaging scene X;

s304, calculating the contrast of the imaging scene X, comparing the contrast with the contrast of the imaging scene X calculated last time, and if the contrast value is larger, storing the corresponding contrast value and the imaging scene X into a matrix Ψ;

s305, updating the search distance and order

Judge this moment

Whether the search range is exceeded or not is judged, and if not, the step is jumped to S302; if yes, ending the circulation;

s306, determining the maximum contrast value in the matrix psi and the index P corresponding to the maximum contrast value, and outputting the index P corresponding to the index P

Corresponding image

According to the 2D-JLBI algorithm solving process and the IC searching process, the translation compensation problem of the embodiment is converted into the problems of compressed sensing two-dimensional joint sparse reconstruction and maneuvering target motion parameter estimation by combining the 2D-JLBI and the IC searching method, and referring to fig. 3, the third step specifically comprises the following implementation steps:

a11, initializing a contrast matrix according to a set search range theta and a set step length alpha based on a Bi-ISAR sparse aperture maneuvering target echo signal model to obtain an initial value a ₀ As a center, determining a search start point

A12, reconstructing an imaging scene X;

a13, calculating the contrast of the imaging scene, comparing the contrast with the contrast of the imaging scene X calculated last time, and if the contrast value is larger, storing the corresponding contrast value and the imaging scene X into a matrix;

update the search point at the same time, order

Judge this moment

If the search range is exceeded, jumping to the step A12 if the search range is not exceeded; if yes, ending the loop and jumping to the step A14;

a14, judging whether the algorithm is convergent or not according to the set maximum iteration times, and if not, enabling the translation track parameter corresponding to the maximum contrast value of the imaging scene

Jumping to step A12; if yes, go to step A15;

a15, determining the maximum contrast value in the matrix, determining and outputting the corresponding translation track parameter

Corresponding image

Preferably, in this embodiment, the search range, the step size, and the iteration number all need to be set reasonably, where at the beginning of the search, we will obtain the coarse speed information a of the target through the narrow-band signal ₀ And with the circulation search of one time, the more accurate the obtained target speed information is, the smaller the change is basically, so the search range and compensation are not set to be large, and the maximum contrast is extracted each time, and the contrast change is smaller and smaller.

For the setting of the iteration times (loop times), which is related to the aperture condition and the SNR, as the SNR decreases, the iteration times are set to be larger, as the effective aperture is smaller, the iteration times are set to be smaller, the random sparse iteration times are more than the block sparse times, scattering points with too large iteration times are also filtered, too many small imaging noise points are obtained, and the preferred iteration times of the embodiment are set to be 5-10 times.

The method of the present embodiment is described below by using a specific simulation case.

The simulation experiment environment is a Windows 1064-bit operating system and a Matlab R2018b software platform, and the main parameters of a computer used for simulation are as follows: the processor is Intel core i7-6700HQ, the main frequency is 2.60GHz, and the memory is 16.0 GB. The performance of the algorithm of the embodiment is verified in the aspect of aperture loss and echo SNR by experimental simulation. In order to conveniently and intuitively explain the superiority of the algorithm, a Target To Background Ratio (TBR) and an image entropy E are adopted _n As a measure, both are briefly described below:

wherein T and B represent a target in-region signal and a target out-of-region signal, respectively, A represents a target image,

the TBR is the ratio of the signal intensity in the target area to the signal intensity outside the target area, can effectively represent the SNR of imaging, and evaluates the estimation accuracy and the noise suppression performance of imaging, and the larger the value is, the better the value is. Entropy of image E _n The image quality evaluation method is used for reflecting the average information quantity in the image, the overall quality of the target image can be evaluated, and the effect is better when the value is smaller.

The Bi-ISAR sparse aperture maneuvering target simulation scene is shown in fig. 4(a), the target scattering point model is shown in fig. 4(b), the full aperture distance-Doppler imaging result is shown in fig. 4(c), and the simulation parameter setting is shown in table 4. Assuming that the bistatic baseline length is 500km, the target uniformly accelerates from the median of the distance between the transmitting and receiving radars to the direction of the receiving radar at the speed of v 3000m/s and a 100m/s2 at the height of 300km, the median of the distance between the transmitting and receiving radars is taken as an imaging starting point, 300 pulses are intercepted as imaging data, the accumulated rotation angle in the observation time is 2.0 degrees, the bistatic angle change range is (79.49 degrees and 79.61 degrees), and the condition that the rotation angle is small during the observation period is met. Specific experimental parameters are shown in table 1 below.

TABLE 1 imaging parameters

(1) And (5) verifying the performance of the algorithm under different pore diameter loss conditions.

In the case of sparse aperture, comparing the algorithm of the embodiment with the 2D-SL0 search algorithm based on IC search and the imaging result of the 2D-FISTA search algorithm based on IC search by changing the situation of aperture missing, and verifying the effectiveness and superiority of the algorithm of the embodiment. This experiment was mainly studied on the imaging results in the case where the SNR was 10dB, the echo signal was randomly missing and blocky missing and the missing data rates were different, and fig. 5 lists the imaging results in the case where the signal was randomly missing 50%, blocky missing 50%, randomly missing 75%, and blocky missing 75%, respectively.

TABLE 2 comparison of algorithmic imaging indices under different aperture loss conditions

As shown in fig. 5, fig. 5(a) -5 (c) are the imaging results of three algorithms in the case of a 50% random absence of signal; FIGS. 5(d) to 5(f) are the imaging results for 75% random deletions; FIGS. 5(g) to 5(i) are imaging results for 50% of the block missing cases; FIGS. 5(j) to 5(l) show results in the case of 50% random deletions; the imaging indices are shown in table 2.

It can be seen from the results shown in fig. 4 that when the data loss rate is 50%, no matter random loss or block loss, the algorithm of the present embodiment can clearly and accurately complete compensation and reconstruct an image, although the 2D-FISTA algorithm and the 2D-SL0 algorithm can recover an approximate target contour, the false scattering point and defocus phenomenon are severe, the noise suppression effect is poor, and it is shown that the data loss rate is the same, and the reconstruction performance of the algorithm of the present embodiment is better than that of the other two algorithms. Particularly, when the data missing rate reaches 75%, the imaging quality of the latter two algorithms is rapidly reduced, the defocusing situation of scattering points is more serious, the number of false scattering points is more, the anti-noise performance is worse, and the scattering points in the imaging result of the 2D-SL0 algorithm in the random missing situation are almost drowned by noise; in the case of block missing, the defocusing situations of the latter two algorithms are serious, but the algorithm of the present embodiment can well complete compensation and high-quality imaging, which shows that the algorithm of the present embodiment can still perform high-quality imaging under the condition of more missing data. As can also be seen from the data in Table 2, in the same data missing situation, the image entropy value formed by the algorithm of the present embodiment is the smallest, the TBR value is the largest, and then the 2D-FISTA algorithm, the 2D-SL0 algorithm is the worst. Viewed in the transverse direction, the smaller the data loss rate is, the better the reconstruction effect is; under the condition of the same data loss rate, the image compensation recovery effect of random loss is better than that of fast loss, because the coherence among data is damaged more seriously due to the loss of data blocks. In conclusion, the superiority of the algorithm of the embodiment is described.

(2) Algorithm performance verification under different SNR conditions

To further illustrate the robustness of the algorithm, consider that in the case of random missing of 50% sparse aperture data, the algorithm is compared with the two algorithms respectively under different SNRs, the radar parameters are set as above, and the experimental result is shown in fig. 6.

TABLE 3 comparison of algorithmic imaging indices under different SNR conditions

As shown in fig. 6, fig. 6(a) -6 (c) are imaging results of three algorithms with SNR of 20 dB; FIGS. 6(d) to 6(f) show the imaging results when the SNR is 10 dB; FIGS. 6(g) to 6(i) are the imaging results at an SNR of 5 dB; FIGS. 6(j) to 6(l) are results of SNR of 0 dB; the imaging indices are shown in table 3. As can be seen from FIG. 6, when the SNR is high, the 2D-JBLI and 2D-FISTA algorithms can achieve compensation and imaging well, and the 2D-SL0 algorithm is seriously interfered by noise. As the SNR is reduced, the 2D-FISTA and the 2D-SL0 algorithms are affected seriously by noise, the noise suppression effect is poor, scattering points and noise cannot be clearly resolved, and the imaging result of the algorithm of the embodiment is less affected by the SNR. When the SNR is 0dB, in the image which is compensated and reconstructed by the last two algorithms, the scattering points are submerged by noise, and the real scattering points and the virtual false scattering points can not be distinguished completely. As can be seen from the data in Table 3, with the decrease of SNR, the entropy value and TBR value of the imaging result of the algorithm are superior to those of the latter two algorithms, which shows that the algorithm of the embodiment has stronger robustness, good noise tolerance and stronger suppression effect on noise while completing the translation compensation under the sparse aperture.

(3) Conclusion

The embodiment provides a compensation imaging method combining a 2D-JLBI reconstruction algorithm and image contrast search based on aiming at the problem that the translational compensation of a maneuvering target is difficult under the Bi-ISAR sparse aperture condition. In the method provided by the embodiment, the compensation item is established through the initial velocity information to carry out joint sparse reconstruction, the translation compensation is carried out on the sparse aperture echo, imaging is carried out, the image with the highest quality is screened out through an image contrast search mode, the velocity information is further accurately estimated, and therefore the optimal image is circularly solved. The 2D-JLBI combined sparse reconstruction mode avoids the defects that the complexity of traditional reconstruction vectorization operation is high, and the distance direction and azimuth direction coupling performance is reduced in row-by-column processing, and the accuracy of translational compensation is guaranteed based on the image maximum contrast searching mode. Simulation experiments show that the method can complete compensation when sparse aperture missing data is more, obtain high-quality images, has strong noise tolerance and has strong robustness.

System embodiment

According to an embodiment of the present invention, a bistatic inverse synthetic aperture radar (Bi-ISAR) sparse aperture maneuvering target imaging system established based on a bistatic inverse synthetic aperture radar (Bi-ISAR) system and the bistatic ISAR sparse aperture maneuvering target imaging method provided by the above method embodiment is provided, as shown in fig. 8, a schematic diagram of a frame of the bistatic ISAR sparse aperture maneuvering target imaging system provided by this embodiment is provided, and the bistatic ISAR sparse aperture maneuvering target imaging system according to the embodiment of the present invention includes:

a data acquisition unit: the Bi-ISAR system performs pulse compression on each echo to obtain a distance frequency domain-azimuth time domain signal; obtaining an initial searching starting point of a target through narrow-band speed measurement; in this embodiment, the receiver performs the processing of this step after receiving the radar signal.

An echo signal model construction unit: according to the distance frequency domain-azimuth time domain signal obtained by the data acquisition unit, fitting the motion track of the maneuvering target by a quadratic polynomial, and establishing a Bi-ISAR sparse aperture maneuvering target echo signal model containing a translation error term;

a model solving unit: the method comprises the steps of establishing a phase compensation term by fitting a quadratic polynomial to a motion track of a maneuvering target, reconstructing an image by using a 2D-JLBI algorithm, updating the phase compensation term based on a translation track parameter corresponding to an IC search optimal image, and solving a model to obtain and output a target reconstruction image based on set search range, step length and cycle number conditions.

Preferably, the data acquiring unit comprises

A first data compression module: the device is used for carrying out pulse compression on the echo acquired each time to obtain one-dimensional range profile echo data of the echo each time;

a second data compression module: performing range dimension compression on the one-dimensional range profile echo data to obtain a range frequency domain-azimuth time domain signal;

the model solving unit includes:

a search starting point determining module: for obtaining the rough movement information of the target as the search starting point a by sending the wide and narrow band velocity measurement signal ₀ ；

The phase compensation term construction module: fitting a motion track of the target along the baseline direction of the bistatic radar by using a quadratic polynomial to establish a phase compensation term, and updating the phase compensation term according to a translation track parameter corresponding to the optimal image determined by the IC searching module;

2D-JLBI calculation module: the phase compensation item construction module is connected with the phase compensation item construction module, a scene image is reconstructed according to the phase compensation item constructed by the phase compensation item construction module, the contrast of the scene image is calculated and stored until the search is finished in a set search range and step length circulation by using a search starting point; for the specific calculation process, reference may be made to the steps described in the above method embodiments, which are not described herein again.

An IC search module: transmitting the translation track parameters corresponding to the better images determined by the 2D-JLBI calculation module to a phase compensation item construction module according to the set maximum iteration times; for the specific search process, reference may be made to the steps described in the above method embodiments, which are not described herein again.

A first judgment module: connected with the 2D-JLBI calculation module and the second judgment module, judging whether the current search range exceeds the set search range according to the set search range and the set step length, and if not, commanding

Adding the set step length, and feeding back to the 2D-JLBI calculation module and the second judgment module;

a second judging module: connected with the IC searching module, judging whether the algorithm is convergent or not according to the set iteration times, if not, adding 1 to the current iteration number, and determining the current according to the 2D-JLBI calculating module

Translation track parameter corresponding to contrast maximum value of imaging scene

And feeding back to the phase compensation item building module, and if the phase compensation item is converged, sending a cycle ending signal to the image determining module.

An image determination module: determining the maximum contrast value in the current matrix and the corresponding translation track parameter according to the received signal

Corresponding image

In this embodiment, the solving process of the model solving unit specifically includes:

based on the echo signal model, according to the set search range theta and the step length alpha, initializing a contrast matrix to obtain an initial value a ₀ As a center, determining a search start point

The phase compensation item building module builds a phase compensation item, an imaging scene X is reconstructed through the 2D-JLBI computing module, the contrast of the imaging scene X is computed and compared with the contrast of the imaging scene X computed last time, and if the contrast value is larger, the corresponding contrast value and the imaging scene X are stored in a matrix psi; at the moment, the first judgment module judges whether the current search range exceeds the set range, and if not, the first judgment module orders the current search range to be more than the set range

And transmitting the data to a 2D-JLBI calculation module for repeated calculation, and if the data exceeds the threshold value, determining the current value

Transmitting the data to a second judging module, judging whether the algorithm is converged according to the set maximum iteration times by the second judging module, if not, adding 1 to the current iteration times, and adding the current iteration times to the second judging module

The feedback is sent to a phase compensation item construction module to update a phase compensation item, and the calculation is continued; if the convergence occurs, sending a cycle ending signal to the image determining module, and determining the maximum contrast value in the current matrix and the corresponding translation track parameter by the image determining module at the moment

Corresponding image

As shown in fig. 8, the present invention also provides a computer readable storage medium, on which a computer program is stored, which when executed by a processor implements the bistatic ISAR sparse aperture maneuvering target imaging method in the above-mentioned embodiment, or which when executed by a processor implements the bistatic ISAR sparse aperture maneuvering target imaging method in the above-mentioned embodiment.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

All the embodiments in the present specification are described in a progressive manner, and portions similar to each other in the embodiments may be referred to each other, and each embodiment focuses on differences from other embodiments. In particular, apparatus or system embodiments, which are substantially similar to method embodiments, are described in relative ease, and reference may be made to some descriptions of method embodiments for related points. The above-described embodiments of the apparatus and system are merely illustrative, and the units described as separate parts may or may not be physically separate, and the parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.

Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A bistatic ISAR sparse aperture maneuvering target imaging method is characterized by comprising the following steps:

compressing the obtained fundamental frequency echo pulse of the target to be detected to obtain a distance frequency domain-azimuth time domain signal; establishing a Bi-ISAR sparse aperture maneuvering target echo signal model containing a translation error term according to the distance frequency domain-azimuth time domain signal; and establishing a phase compensation term, reconstructing an image by using a 2D-JLBI algorithm, updating the phase compensation term based on the translation track parameter corresponding to the IC search optimal image, and solving a model to obtain and output a target reconstructed image based on a set convergence condition.

2. The bistatic ISAR sparse aperture maneuvering target imaging method as claimed in claim 1, characterized in that said reconstructing images using 2D-JLBI algorithm, updating phase compensation terms based on the translation trajectory parameters corresponding to the IC search optimal images, solving the model based on the set convergence conditions to obtain and output the target reconstructed images specifically comprises the implementation steps of:

a11, based on the Bi-ISAR sparse aperture maneuvering target echo signal model, according to the set search range theta and the step length alphaInitializing the contrast matrix to an initial value a ₀ As a center, determining a search start point

A12, reconstructing an imaging scene;

a13, calculating the contrast of the imaging scene, comparing the contrast with the contrast of the imaging scene calculated last time, and if the contrast value is larger, storing the corresponding contrast value and the imaging scene into a matrix;

update the search point at the same time, order

Judge this moment

If not, jumping to the step A12; if yes, ending the loop and jumping to the step A14;

Jumping to step A12; if yes, go to step A15;

Corresponding image

3. The bistatic ISAR sparse aperture maneuvering target imaging method of claim 1, wherein the image contrast is a ratio of a standard deviation and a mean of an image amplitude, calculated as:

wherein the content of the first and second substances,

is the image complex amplitude and a (-) is the averaging operation.

4. The bistatic ISAR sparse aperture maneuvering target imaging method according to claim 1, characterized in that the Bi-ISAR sparse aperture maneuvering target echo signal model can be expressed in matrix form as:

S＝E⊙(AXB)+N _o (9)

Matrix E:

Matrix A:

Matrix B:

wherein, < > indicates a Hadamard product, S ∈ C ^N×R An echo signal matrix representing a range frequency domain-an azimuth time domain; e is as large as C ^N ^×R Representing envelope walk and phase shift caused by the translation component of the object to the echo, A ∈ C ^N×M A matching Fourier transform matrix containing a parameter gamma and corresponding to the azimuth dimension is represented; b is belonged to C ^R×R A Fourier transform matrix representing a distance dimension; x belongs to C ^M×R Representing a discretized imaged scene; n is a radical of _o ∈C ^N×R Representing observation noise; n, M and R respectively represent the number of azimuth sampling pulses, the number of scene azimuth units and the number of echo distance dimensional samples;

representing the influence of the target translation component on the echo envelope;

R _k (t _m ) Represents the k-th scattering point at a slow time t _m The sum of the distances between the time and the receiving and transmitting double stations;

c is the electromagnetic wave propagation speed;

f _c is the carrier frequency, f _r ＝f _s N is the distance sampling frequency, f _s Is the sampling rate, and N is the number of sampling points;

y _q to discretize a scene, the distance dimension imaging unit, y _k The representation shows the kth scattering point.

5. The bistatic ISAR sparse aperture maneuvering target imaging method as claimed in claim 4, characterized in that the motion trajectory of the target along the bistatic radar baseline direction is fitted based on quadratic polynomial, and in combination with the sparsity of the Bi-ISAR imaging scene, the translation error correction problem of the Bi-ISAR target is equivalent to solving the optimization problem shown as follows, namely, the optimization model is established with the most sparse scene as the condition:

in the formula, | · the luminance | | _F Frobenius norm, | | · | | | torry representing matrix ₁ L representing a matrix ₁ And a norm, wherein a is a vector formed by fitting polynomial coefficients to the motion trajectory, and λ represents a regularization parameter of a sparse term and is used for controlling the estimation precision.

6. The bistatic ISAR sparse aperture maneuvering target imaging method of any of claims 1-5, characterized by the 2D-JLBI algorithm solving iteration format as follows:

in the above formula, R ^(k) Denotes the residual error, X, of each iteration ^(k+1) The result obtained for each iteration; v ^(k+1) Is an intermediate variable; x ⁽⁰⁾ ＝V ⁽⁰⁾ ＝0；A ^* ＝A ^H (AA ^H ) ^-1 ，B ^* ＝(B ^H B) ^-1 B ^H ；A∈C ^N×M A matching Fourier transform matrix containing a parameter gamma and corresponding to the azimuth dimension is represented; s belongs to C ^N×R An echo signal matrix representing a range frequency domain-an azimuth time domain; b is belonged to C ^R×R A Fourier transform matrix representing a distance dimension; x belongs to C ^M×R Representing a discretized imaged scene; η is a weight parameter.

7. A bistatic ISAR sparse aperture maneuvering target imaging system is based on a bistatic inverse synthetic aperture radar system and is characterized by comprising

an echo signal model construction unit: according to the distance frequency domain-azimuth time domain signal obtained by the data acquisition unit, fitting the motion trail of the maneuvering target through a quadratic polynomial, and establishing a Bi-ISAR sparse aperture maneuvering target echo signal model containing a translation error term;

a model solving unit: and reconstructing the image by using a 2D-JLBI algorithm, updating a phase compensation item based on the translation track parameter corresponding to the IC search optimal image, and solving a model to obtain and output a target reconstruction image based on the set search range, step length and cycle time conditions.

8. The bistatic ISAR sparse aperture maneuvering target imaging system of claim 7, wherein the model solving unit comprises:

a search starting point determining module: for obtaining rough movement information of target as search starting point by sending wide-narrow band speed measurement signal

The phase compensation term construction module: fitting a motion track of the target along the baseline direction of the bistatic radar by using a quadratic polynomial to establish a phase compensation item, and updating the phase compensation item according to a translation track parameter corresponding to the optimal image determined by the IC searching module;

2D-JLBI calculation module: the phase compensation item construction module is connected with the scene image reconstruction module, and the scene image is reconstructed according to the phase compensation item constructed by the phase compensation item construction module, and the scene image contrast is calculated and stored until the search is finished in a set search range and step length circulation by using a search starting point;

an IC search module: transmitting the translation track parameters corresponding to the better images determined by the 2D-JLBI calculation module to a phase compensation item construction module according to the set maximum iteration times;

a second judging module: the system is connected with an IC searching module, whether the algorithm is converged is judged according to the set iteration times, if not, the current iteration number is added by 1, the current searching point is determined according to a 2D-JLBI calculating module, the translation track parameter value corresponding to the maximum contrast value of the imaging scene is taken as a searching starting point and is fed back to a phase compensation item constructing module, and if the algorithm is converged, a cycle ending signal is sent to an image determining module;

an image determination module: and determining the maximum contrast value in the current matrix, and the corresponding translation track parameter and the corresponding image according to the received signal.

9. Computer device comprising a memory, a processor and a computer program stored in said memory and executable on said processor, characterized in that said processor when executing said computer program implements the bistatic ISAR sparse aperture maneuvering target imaging method according to any of claims 1 to 6.

10. A computer readable storage medium storing a computer program, wherein the computer program when executed by a processor implements the bistatic ISAR sparse aperture maneuvering target imaging method as claimed in any of claims 1 to 6.