A frequency-modulated single-channel SAR-GMTI method

Abstract

The present invention relates to a frequency-modulated single-channel SAR-GMTI method, which belongs to the field of radar image technology. The method comprises: establishing a single-channel airborne synthetic aperture radar signal model in combination with imaging geometry; two-dimensional frequency domain imaging processing with variable Doppler modulation frequency; and Doppler fuzzy processing and target detection. Compared with traditional multi-channel methods, the proposed method can effectively avoid the correlation problem between channels and reduce the requirements and dependence on system hardware. At the same time, compared with other traditional single-channel methods, the proposed method belongs to the focusing method, which can effectively improve the probability of detecting moving targets under the same false alarm probability. The proposed method is relatively robust to the Doppler center estimation error and has better clutter suppression capability and detection performance in measured data. While detecting moving targets, the proposed method can also estimate the radial velocity of the moving target, thereby relocating the target more accurately.

Description

Technical Field

[0001] The invention belongs to the technical field of radar images, and in particular relates to a frequency-modulated single-channel SAR-GMTI method. Background Art

[0002] After acquiring high-resolution synthetic aperture radar (SAR) images, the most important task is to detect moving targets in the region of interest (ROI). However, in complex clutter environments, moving targets are often overwhelmed by strong ground clutter. Traditional SAR ground moving target detection techniques primarily rely on strong inter-channel correlation to effectively suppress ground stationary clutter, thereby achieving ground moving target indication (GMTI). Key approaches include offset phase center antenna (DPCA), space-time adaptive processing (STAP), and along-track interferometry (ATI). Multi-channel SAR ground moving target detection methods are generally robust, but their effectiveness is primarily limited by inter-channel correlation, which can be affected by several factors, such as amplitude and phase errors between channels, image registration issues, and non-uniform azimuth sampling. To mitigate the effects of inter-channel inconsistencies, the use of single-channel data from the SAR system to detect moving targets is considered. Traditional single-channel SAR-GMTI methods primarily utilize the Doppler shift introduced by the moving target's radial velocity to detect moving targets outside the main clutter zone. To better detect slow-moving targets within the main clutter region, a defocusing filter is designed using different Doppler modulation frequencies to generate a set of defocused synthetic aperture radar (SAR) images. For each defocused image, a stationary target remains essentially stationary, while a moving target, due to its radial velocity, experiences a shift or significant defocus. Comparing these defocused images allows for effective detection of moving targets. However, defocused images prevent optimal energy accumulation in moving targets, resulting in a reduced signal-to-noise ratio (SCNR), which ultimately affects the probability of detection. To improve this defocusing approach, multiple Doppler perspectives can be used to generate a set of focused images, which can then be compared to detect targets. However, dividing the Doppler spectrum into different perspectives is equivalent to using multi-channel data for moving target detection, and different Doppler perspectives also suffer from inter-perspective correlation. More importantly, the Doppler center of a SAR system can deviate due to factors such as motion error. This deviation increases the inter-perspective correlation, severely impacting moving target detection results. Summary of the Invention

[0003] Technical problems to be solved

[0004] Aiming at the perspective inconsistency problem existing in single-channel airborne SAR radar when detecting targets through images with different Doppler perspectives, a focused imaging and compensation method with variable Doppler modulation frequency is proposed. This method can eliminate channel inconsistency while reducing the requirement and dependence on the Doppler center estimation accuracy, and effectively improve the detection probability of moving targets.

[0005] Technical Solution

[0006] A frequency-modulated single-channel SAR-GMTI method, characterized by the following steps:

[0007] Step 1: Build a single-channel airborne synthetic aperture radar signal model based on imaging geometry

[0008] 1a) Establish a model for a single-channel antenna transmitting a linear frequency modulation waveform signal;

[0009] 1b) Combined with the waveform signal model, the instantaneous slant range expression of the moving target and the clutter point is given;

[0010] 1c) Combined with the instantaneous slant range expression, the received signal model of the clutter point and the moving target is established;

[0011] Step 2: Variable Doppler modulation frequency two-dimensional frequency domain imaging processing

[0012] 2a) According to the stationary point theorem, the received signal is converted to the range frequency domain and the frequency modulated secondary phase in the signal is compensated;

[0013] 2b) constructing a variable Doppler frequency modulation filter and compensating the echo signal in the range frequency domain and azimuth time domain;

[0014] 2c) Using the stationary point theorem, transform the compensated signal into the Doppler domain;

[0015] 2d) Designing a range unit migration filter to compensate for range migration in the signal;

[0016] 2e) performing a two-dimensional inverse Fourier transform on the compensated echo signal to obtain a coarsely focused radar image;

[0017] Step 3: Doppler blurring and target detection

[0018] 3a) Expanding the number of azimuth samples by inverse Fourier transform;

[0019] 3b) Calculate the number of azimuth deviation units and compensate for the azimuth offset;

[0020] 3c) extracting the compensated orientation unit;

[0021] 3d) Calculating and estimating the distance offset and radial velocity of the moving target based on the focus position information.

[0022] A further technical solution of the present invention: The received signal models of the moving target and the clutter point in step 1c are respectively:

[0023]

[0024]

[0025]

[0026]

[0027]

[0028] Among them, t a represents slow time, t represents fast time; c represents the speed of electromagnetic wave, A(σ t ) and A(σ c ) represent the complex reflection envelopes of the moving target and clutter scattering points, v represents the speed of the aircraft platform along the x-axis, v a and v r They represent the along-track velocity and radial velocity of the moving target, R 0c is the shortest slant distance to the clutter scattering point, R 0t is the instantaneous slant range when the radar directly illuminates the moving target, x 0t and x 0c Represent the initial azimuth positions of the moving target and clutter scattering points respectively.

[0029] A further technical solution of the present invention: the radar image in step 2e is:

[0030]

[0031]

[0032]

[0033] Θ v (t,t a )=∫∫Θ v (f r , f a )exp(j2πf r t+j2πf a t a )df r df a

[0034] Where β∈(0,1) is the modulation frequency coefficient in the azimuth time domain, β′=1-(1-β)v 2 / (vv a )2 is the loss coefficient, α′=β′vv a ) 2 / βv 2 represents the loss coefficient, Δf a，c，β is the new Doppler bandwidth of the clutter scattering point, Δf a，t，β′ represents the new Doppler bandwidth of the moving target, A(σ c ) represents the scattering intensity of the moving target, A(σ t ) represents the scattering intensity of the clutter point, αf r represents the distance frequency difference, and λ represents the wavelength.

[0035] A further technical solution of the present invention: the distance offset of the moving target described in step 3d is:

[0036]

[0037] in, <M r > is the number of distance units between the two moving targets detected, M r is the non-integer distance of the actual difference, F s Represents the sampling frequency of the synthetic aperture radar system.

[0038] A further technical solution of the present invention: the radial velocity of the moving target in step 3d is:

[0039]

[0040] Among them, B r Indicates the transmission bandwidth.

[0041] A computer system, characterized in that it includes: one or more processors, and a computer-readable storage medium for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the above-mentioned method.

[0042] A computer-readable storage medium is characterized by storing computer-executable instructions, which are used to implement the above method when executed.

[0043] Beneficial effects

[0044] Compared with the prior art, the method of the present invention has the following beneficial effects when performing moving target detection processing on single-channel SAR images:

[0045] Compared with traditional multi-channel methods, the proposed method can effectively avoid the correlation problem between channels and reduce the requirements and dependence on system hardware. At the same time, compared with other traditional single-channel methods, the proposed method belongs to the focusing method, which can effectively improve the probability of detecting moving targets under the same false alarm probability. The proposed method is relatively robust to the Doppler center estimation error and has better clutter suppression ability and detection performance in the measured data. While detecting moving targets, the proposed method can also estimate the radial velocity of the moving target, thereby relocating the target more accurately. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The accompanying drawings are only for the purpose of illustrating particular embodiments and are not to be considered limiting of the present invention. Like reference symbols denote like parts throughout the drawings.

[0047] Figure 1 The strip-type single-channel synthetic aperture radar system adopted by the present invention;

[0048] Figure 2 This is the result of the variable Doppler frequency modulation imaging used in the present invention;

[0049] Figure 3 This is the imaging result after the TDCV used in the present invention compensates for the azimuth offset;

[0050] Figure 4 This is a processing flow chart of the TDCV method adopted in the present invention. DETAILED DESCRIPTION

[0051] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only intended to illustrate the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

[0052] The present invention provides a frequency-modulated single-channel SAR-GMTI method, which comprises the following steps:

[0053] (1) Establish a single-channel airborne synthetic aperture radar signal model based on imaging geometry;

[0054] (2) Two-dimensional frequency domain imaging processing with variable Doppler modulation frequency;

[0055] (3) Doppler fuzzy processing and target detection.

[0056] in:

[0057] Step (1) mainly includes the following steps:

[0058] a) Establish a single-channel antenna transmitting linear frequency modulation waveform signal model;

[0059] b) Combined with the waveform signal model, the instantaneous slant range expression of the moving target and the clutter point is given;

[0060] c) Combined with the instantaneous slant range expression, a receiving signal model of the clutter point and the moving target is established.

[0061] Step (2) mainly includes the following steps:

[0062] a) According to the stationary point theorem, the received signal is converted into the range frequency domain and the frequency modulated secondary phase in the signal is compensated;

[0063] b) constructing a variable Doppler frequency modulation filter and compensating the echo signal in the range frequency domain and azimuth time domain;

[0064] c) using the stationary point theorem, transforming the compensated signal into the Doppler domain;

[0065] d) Design a range unit migration filter to compensate for the range migration in the signal;

[0066] e) Performing a two-dimensional inverse Fourier transform on the compensated echo signal to obtain a coarsely focused radar image.

[0067] Step (3) mainly includes the following steps:

[0068] a) Expanding the number of azimuth samples by inverse Fourier transform;

[0069] b) Calculate the number of azimuth deviation units and compensate for the azimuth offset;

[0070] c) extracting the compensated orientation unit;

[0071] d) Calculate and estimate the distance offset and radial velocity of the moving target based on the focus position information.

[0072] In order to enable those skilled in the art to better understand the present invention, the present invention is described in detail below with reference to specific embodiments.

[0073] (1) Airborne single-channel synthetic aperture radar signal model

[0074] Assume that the airborne single-channel SAR system operates in the strip-front side-view mode, as Figure 1 As shown. Among them, the x-axis represents the ideal trajectory, and the y-axis represents the distance coordinate. The linear frequency modulation waveform transmitted by a single-channel antenna is shown as follows:

[0075]

[0076] Where t represents the fast time (sampling time within a pulse), Tp represents the pulse duration, f c and γ represent the center frequency and modulation frequency of the linear frequency modulation signal respectively, and the rectangular window function rect is expressed as

[0077]

[0078] exist Figure 1 In, R t (t a ) and R c (t a ) represent the slant ranges of the moving target and the clutter scattering point, respectively. Therefore, the point target receiving signals of the moving target and the clutter scattering point can be expressed as:

[0079]

[0080]

[0081] Among them, t a represents slow time (azimuth time), c represents electromagnetic wave speed, A(σ t ) and A(σ c ) represent the complex reflection envelopes of the moving target and clutter scattering points respectively. Assume that the aircraft platform moves along the x-axis at a speed v, and the along-track speed and radial speed of the moving target are v respectively. a and v r Then the slope distance formula can be approximately expressed as:

[0082]

[0083]

[0084] where R 0c is the shortest slant distance to the clutter scattering point, R 0t is the instantaneous slant range when the radar directly illuminates the moving target, x 0t and x 0c They represent the initial azimuth positions of the moving target and the clutter scattering point respectively. The azimuth zero moment refers to the moment when the radar beam illuminates the scattering point. Considering the assumption that vt a <<R 0c ,(vv a )t a <<R 0t , so the high-order phase terms can be ignored. In addition, it is assumed that the moving target moves on a plane at a constant speed throughout the synthetic aperture time.

[0085] (2) Variable Doppler modulation frequency two-dimensional frequency domain imaging algorithm

[0086] According to the stationary point principle, the original received signals (2) and (3) can be converted into the range frequency domain:

[0087]

[0088]

[0089] where w a (t a ) is the envelope of the azimuth-time domain, W r (f r ) is the envelope in the range-frequency domain. By compensating the frequency modulation secondary phase in equations (6) and (7), the signal form of the moving target and the clutter scattering point in the range-frequency domain-azimuth-time domain can be expressed as:

[0090]

[0091]

[0092] In fact, the Doppler variable frequency modulation (TDCV) filter can be considered as a filter that compensates for the range unit migration. After obtaining the signal expressions (8) and (9) in the range frequency domain and the azimuth time domain, the TDCV filter can be obtained by compensating for the partial Doppler frequency modulation phase.

[0093]

[0094]

[0095] Where β∈(0,1) is the modulation frequency coefficient in the azimuth time domain, β′=1-(1-β)v 2 / (vv a ) 2 is the loss coefficient. Multiplying equations (10) and (11) with equations (8) and (9) respectively yields:

[0096]

[0097]

[0098] Obviously, the value of β determines the time-bandwidth product (TBP). Only when the TBP is large enough can the stationary point principle be used to roughly derive expressions for the Fourier transform and inverse Fourier transform. Assuming that the TBP is large enough, the signal is converted to the Doppler domain according to the stationary point principle. The signals of moving point targets and clutter scattering points can be expressed as:

[0099]

[0100]

[0101] As can be seen from the above formula, the mutual coupling term of range and azimuth (range unit migration term) still remains. The filter designed to compensate for the range unit migration is as follows:

[0102]

[0103]

[0104] where α′=β′vv a ) 2 / βv 2 represents the loss coefficient. Then the output signal is:

[0105]

[0106]

[0107] in

[0108]

[0109] Similarly, performing a two-dimensional inverse Fourier transform on the above signal, we finally get:

[0110]

[0111]

[0112] where Δf a,c,β is the new Doppler bandwidth of the clutter scattering point, and

[0113]

[0114] Θ v (t,t a )=∫∫Θ v (f r , f a )exp(j2πf r t+J2πf a t a )df r df a (twenty four)

[0115] where Δf a,t,β′ represents the new Doppler bandwidth of the moving target. Since the Doppler modulation rate changes in equations (14) and (15), its Doppler bandwidth also changes with the modulation rate change coefficient β. Here we assume that the error caused by the speed of the moving target along the track can be ignored, then β≈β′. The TDCV images in equations (22) and (23) will appear blurred due to the reduced azimuth resolution, as shown in Figure 2 shown.

[0116] (3) Doppler ambiguity processing and ground moving target detection and estimation

[0117] Depend on Figure 2 It can be seen that due to the reduction of the Doppler modulation frequency, the Doppler spectrum is folded, and the positions of the moving target and the clutter scattering points are also offset. The following is a compensation method to solve the Doppler spectrum ambiguity problem. The compensation method is mainly divided into three steps:

[0118] Step 1 (expanding the number of azimuth samples by inverse Fourier transform): Assume that the number of original azimuth units is N a Then we expand the number of azimuth units to N by performing zero-filling inverse Fourier transform on equations (18) and (19). a / β, the signal is converted to:

[0119]

[0120]

[0121] where t a,β =t′ a / β and assuming β≈β′. Combining equations (25) and (26), we can obtain:

[0122]

[0123]

[0124] By expanding the number of azimuth units, the Doppler spectrum solves the folding problem, but the entire scene still shifts along the azimuth direction.

[0125] Step 2 (compensation for azimuthal offset): During the inverse Fourier transform in step 1, since the number of slow-time samples (number of azimuthal units) is newly expanded, the new slow-time center deviates from the original slow-time Fourier transform center. The number of units of the deviation is:

[0126]

[0127] Where <·> represents the rounding operation. This offset can directly offset the entire scene by t off Although the relative positions of the moving target and the clutter scattering points are correct, the absolute bearing units are still wrong.

[0128] Step 3 (Extracting azimuth units): After compensating for the azimuth offset, the relative positions of the moving target and the clutter scattering points are the same as those obtained after the traditional range-Doppler algorithm. The only difference is that there are N aazimuth units and the azimuth resolution is ρ a , and the focused image generated by TDCV has N a / β azimuth units and the azimuth resolution is ρ a / β. Extract the average value from each 1 / β unit as the value of one orientation unit, and stack all the units into a new image, such as Figure 3 After extracting the azimuth unit, the signal expression of the moving target and the clutter scattering point is:

[0129]

[0130]

[0131] in In the new TDCV image, the clutter scattering point is focused on the point The original position is almost unchanged, while the moving target is focused on the point It can be seen that the focus positions of moving targets and clutter scattering points have changed in the original SAR image and the new TDCV SAR image. Among them, the azimuth positions of all scattering points have not changed, as long as Among them F s represents the sampling frequency of the synthetic aperture radar system, then the distance position of the clutter scattering point remains basically unchanged. The above inequality is valid in the center area of ​​the scene (i.e., x 0c Smaller areas) are basically satisfied. In fact, only the area in the center of the scene is the area of ​​complete synthetic aperture, and it is also the area of ​​concern. For moving targets, the difference in the distance direction in the two focused images will be greater. This is because the moving target has a certain radial velocity, and when the radial velocity is large enough, the distance position of the moving target in the two focused images will differ by more than one distance unit. Because the two images are generated using the same channel data, there is no problem of channel consistency, but the absolute amplitudes of the two images are different because the final focused Doppler bandwidths are different, but their normalized amplitudes are roughly the same. By comparing the two normalized images, a pair of targets can be detected for moving targets with sufficiently large radial velocities, and this pair of targets actually corresponds to one moving target. Assume that the difference between the two detected moving targets is <M r > distance units, where M r is a non-integer distance difference, so the distance offset of the moving target can be roughly written as:

[0132]

[0133] Therefore, the radial velocity of the moving target can be estimated as:

[0134]

[0135] The '±' symbol represents the moving direction of the moving target. The radial velocity estimated by the above formula can be used to relocate the moving target and restore its true position in the synthetic aperture radar image.

[0136] The above description is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with this technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present invention, and these modifications or replacements should all be included in the scope of protection of the present invention.

Claims

1. A frequency-modulated single-channel SAR-GMTI method, characterized in that Here are the steps: Step 1: Build a single-channel airborne synthetic aperture radar signal model based on imaging geometry 1a) Establish a model for a single-channel antenna transmitting a linear frequency modulation waveform signal; 1b) Combined with the waveform signal model, the instantaneous slant range expression of the moving target and the clutter point is given; 1c) Combined with the instantaneous slant range expression, the received signal model of the clutter point and the moving target is established; Step 2: Variable Doppler modulation frequency two-dimensional frequency domain imaging processing 2a) According to the stationary point theorem, the received signal is converted to the range frequency domain and the FM secondary phase in the signal is compensated; 2b) Construct a variable Doppler frequency modulation filter and compensate the echo signal in the range frequency domain and azimuth time domain; 2c) Using the stationary point theorem, transform the compensated signal into the Doppler domain; 2d) Design a range unit migration filter to compensate for the range migration in the signal; 2e) Performing a two-dimensional inverse Fourier transform on the compensated echo signal to obtain a coarsely focused radar image; Step 3: Doppler blurring and target detection 3a) Expanding the number of azimuth samples by inverse Fourier transform; 3b) Calculate the number of azimuth deviation units and compensate for the azimuth offset; 3c) extracting the compensated orientation unit; 3d) Calculate and estimate the distance offset and radial velocity of the moving target based on the focus position information.

2. The frequency-modulated single-channel SAR-GMTI method according to claim 1, characterized in that: The received signal models of the moving target and clutter point described in step 1c are: in, t a Indicates slow time, Indicates fast time; c represents the speed of electromagnetic waves, and Represent the complex reflection envelopes of moving targets and clutter scattering points respectively, Indicates that the aircraft platform is along x The speed of the axis, and represent the along-track velocity and radial velocity of the moving target, respectively. R 0c is the shortest slant distance to the clutter scattering point, R 0t It is the instantaneous slant range when the radar directly illuminates the moving target. and Represent the initial azimuth positions of the moving target and clutter scattering points respectively, and They represent the center frequency and modulation rate of the linear frequency modulation signal respectively.

3. The frequency-modulated single-channel SAR-GMTI method according to claim 2, characterized in that: The radar image described in step 2e is: in, is the modulation frequency coefficient in the azimuth time domain, is the loss coefficient, represents the loss coefficient, is the new Doppler bandwidth of the clutter scattering point, represents the new Doppler bandwidth of the moving target, represents the distance frequency difference, Indicates wavelength.

4. The frequency-modulated single-channel SAR-GMTI method according to claim 3, characterized in that: The distance offset of the moving target described in step 3d: in, is the number of distance units between the two moving targets detected, is a non-integer distance of the actual difference, Represents the sampling frequency of the synthetic aperture radar system.

5. The frequency-modulated single-channel SAR-GMTI method according to claim 4, characterized in that: The radial velocity of the moving target described in step 3d: in, Indicates the transmission bandwidth.

6. A computer system, characterized in that include: One or more processors, and a computer-readable storage medium for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors are enabled to implement the method of claim 1.

7. A computer-readable storage medium, characterized in that Computer-executable instructions are stored, and when the instructions are executed, they are used to implement the method of claim 1.

Images