Two-dimensional fuzzy suppression method of waveform diversity alternate emission sparse SAR mode

Abstract

The invention discloses a two-dimensional fuzzy suppression method for a waveform diversity alternate transmission sparse SAR (Synthetic Aperture Radar) mode, which comprises the following steps of: firstly, constructing NLFM (Non-linear Frequency Modulation) wave waveform diversity with different power spectral densities, and carrying out MF (Multi-Frequency) imaging processing on an SAR based on a waveform diversity alternate transmission mode so as to realize initial suppression on range ambiguity; and secondly, aiming at azimuth ambiguity generated by periodic phase-amplitude modulation introduced by alternately transmitting different waveforms, constructing a sparse SAR imaging model based on incomplete Doppler spectrum observation, converting the sparse SAR imaging model into an optimization reconstruction problem based on L1 norm regularization, and solving by adopting a threshold iterative recovery algorithm to obtain a sparse SAR imaging model. Therefore, the residual range ambiguity is further weakened while the azimuth ambiguity is suppressed. The method is based on image domain processing, effective suppression of azimuth ambiguity can be realized in the SAR in the alternate transmission mode without acquiring system parameters and specific transmission waveform information in advance, suppression of range ambiguity is considered, and the method has the characteristics of simplicity in implementation and high applicability.

Description

Technical Field

[0001] This invention belongs to the field of Synthetic Aperture Radar (SAR) imaging and sparse microwave imaging based on waveform diversity alternating transmission mode, and specifically relates to a two-dimensional ambiguity suppression method for waveform diversity alternating transmission sparse SAR mode. Background Technology

[0002] SAR, as an advanced active microwave Earth remote sensing method, plays an irreplaceable role in fields such as Earth observation, disaster emergency response and monitoring, resource survey and exploration, and military reconnaissance. A core technical challenge facing spaceborne SAR systems is the inherent constraint between high azimuth resolution and a wide range swathe. To achieve high azimuth resolution, a shorter antenna is needed to illuminate the target for a longer period to obtain a higher Doppler bandwidth. According to the Nyquist sampling theorem, the system needs to use a high pulse repetition frequency (PRF) to avoid azimuth ambiguity caused by spectral aliasing. However, a wide range swathe requires a sufficiently long pulse repetition interval (PRI) between pulses to prevent range ambiguity caused by the superposition of consecutive pulse echoes. Therefore, in traditional SAR systems, there is an inherent trade-off between azimuth ambiguity suppression and range ambiguity suppression in system design.

[0003] Alternating transmission mode SAR based on waveform diversity improves system performance by increasing the degree of freedom in waveform design, especially in range ambiguity suppression. Compared with other methods, this mode can suppress range ambiguity without increasing system complexity or changing the PRF. However, the transmission of different waveforms by adjacent pulses introduces periodic phase-amplitude modulation in the azimuth direction, resulting in azimuth ambiguity. Although azimuth ambiguity can be reduced by truncating the Doppler spectrum, the azimuth resolution is inevitably lost due to the reduction in equivalent bandwidth. Taking advantage of the property of compressed sensing that it can recover high-dimensional signal information from low-dimensional observations, this invention describes the azimuth ambiguity suppression problem under alternating transmission waveforms as sparse reconstruction under incomplete Doppler spectrum observations. By solving the L1 norm regularization problem, it can effectively suppress azimuth ambiguity while further reducing residual range ambiguity, achieving joint suppression of range-azimuth two-dimensional ambiguity and ambiguity-free sparse reconstruction of the observation scene. Summary of the Invention

[0004] Purpose of the invention: The purpose of this invention is to provide a two-dimensional ambiguity suppression method for waveform diversity alternating transmission sparse SAR mode, which is used to jointly suppress range-azimuth two-dimensional ambiguity, thereby achieving ambiguity-free sparse reconstruction of the observation scene.

[0005] Technical solution: The two-dimensional ambiguity suppression method for waveform diversity alternating transmission sparse SAR mode described in this invention is implemented as follows:

[0006] (1) Use the asymmetric PWL function and ALGA to generate NLFM waveform diversity with different power spectral densities;

[0007] (2) Perform MF imaging on the echo data of NLFM waveform diversity alternating transmission mode SAR to generate SAR image of the observation scene;

[0008] (3) Using SAR images, construct a sparse SAR imaging model based on incomplete Doppler spectrum;

[0009] (4) The sparse SAR imaging model based on incomplete Doppler spectrum is transformed into an optimization reconstruction problem based on L1 norm regularization;

[0010] (5) Solve the optimization and reconstruction problem in step (4) by using the threshold iterative recovery algorithm to achieve unambiguous sparse reconstruction of the observed scene.

[0011] Furthermore, the implementation process of step (1) is as follows:

[0012] The baseband radar signal is:

[0013] (1)

[0014] in, Represents the pulse duration. It represents fast time. For signal phase; select a weighting function with a peak-to-side-lobe ratio between -30dB and -40dB. Generate the initial NLFM waveform:

[0015] (2)

[0016] Instantaneous frequency modulation , for about Functions:

[0017] (3)

[0018] Where, constant , For the transmitted signal bandwidth;

[0019] use The asymmetric PWL function of order 1 reconstructs the NLFM wave generated in step (11), resulting in a new positive instantaneous frequency. and negative instantaneous frequency They are respectively:

[0020] (4)

[0021] (5)

[0022] in, and Represents the start / end time of each waveform segment. and This represents the frequency corresponding to the end of each waveform segment. and The modulation frequency of each waveform segment:

[0023] (6)

[0024] (7)

[0025] Based on the instantaneous frequency given by formulas (4) and (5) and formula (2), generate the NLFM waveform based on the asymmetric PWL function;

[0026] Make the duration of each waveform equal, that is , , Through optimization and common The problem involves optimizing the NLFM waveform with several parameters to reduce sidelobes without broadening the impulse response width. This problem is formulated as a nonlinear constrained optimization problem.

[0027] (8)

[0028] in, , This represents a nonlinear inequality constraint; the generated NLFM waveform is optimized using ALGA, and the optimization problem is equivalent to:

[0029] (9)

[0030] in, For the fitness function, For the Lagrange multiplier estimates, The offset is non-negative.

[0031] Furthermore, the implementation process of step (2) is as follows:

[0032] Constructing distance compression operators in RDA Distance migration correction operator and azimuth compression operator Alternating firing NLFM waveforms The distance compression operator is expressed as follows:

[0033] (10)

[0034] in, , For the first One transmitted waveform, Indicates the period index of the transmitted waveform. This represents the number of samples in the azimuth direction. , , and These represent the Fourier transform operators and inverse Fourier transform operators for the azimuth and range directions, respectively; based on the constructed operators, the matched filtering process... Represented as:

[0035] (11)

[0036] in, For Hadamard product; for SAR echo data in alternating transmission mode Perform MF processing:

[0037] (12)

[0038] in, SAR image of the target observation scene.

[0039] Furthermore, the implementation process of step (3) is as follows:

[0040] Constructing a sparse SAR imaging model based on incomplete Doppler spectrum:

[0041] (13)

[0042] in, Represents image reconstruction using the MF imaging algorithm Backscattering coefficient of the target observation scene The differences between them For rectangular window functions, the specific expression is as follows:

[0043] (14)

[0044] in, The Doppler bandwidth of the echo signal. For Doppler frequency, The center frequency of the Doppler wave. The scaling factor is used to control the range of the observed Doppler spectrum.

[0045] Furthermore, the implementation process of step (4) is as follows:

[0046] For the constructed sparse SAR imaging model, scene reconstruction is achieved by solving the following L1 regularization problem:

[0047] (15)

[0048] in, It is a reconstructed, unambiguous, sparse SAR image. Regularization parameters are used to control the sparsity of the target observation scene.

[0049] Furthermore, the implementation process of step (5) is as follows:

[0050] Unambiguous sparse reconstruction of the observed scene is achieved through iterative recovery; the threshold iterative algorithm is input to the SAR image obtained in step (2). and the window function obtained in step (3) ; Let's define the target scenario for sparse reconstruction. The initial value is 0, and the iteration parameter is... The error parameter is The maximum number of iterations is When the condition is met, the number of iterations is [number]. And residuals When this happens, perform the following steps:

[0051] S1, Estimated residual data values:

[0052] (16)

[0053] (17)

[0054] S2, Update gradient values:

[0055] (18)

[0056] in, This is a parameter that controls the convergence speed of the algorithm, and is usually set to a constant;

[0057] S3, Update the parameters controlling the sparsity of the target observation scene. :

[0058] (19)

[0059] in, Indicates amplitude value Sort in descending order The largest component;

[0060] S4, Threshold shrinkage of the target observation scene:

[0061] (20)

[0062] S5, Estimate the residuals of the reconstructed image:

[0063] (twenty one)

[0064] If satisfied and residual If so, continue iterating, that is... Repeat steps S1 to S5; if the condition is not met, end the iteration and output the recovered sparse unblurred image. .

[0065] The device according to the present invention includes a memory and a processor, wherein:

[0066] Memory is used to store computer programs that can run on a processor;

[0067] A processor, configured to, while running the computer program, perform the steps of the two-dimensional blur suppression method for waveform diversity alternating transmission of sparse SAR modes as described above.

[0068] The present invention discloses a storage medium storing a computer program, which, when executed by at least one processor, implements the steps of the two-dimensional blur suppression method for waveform diversity alternating transmission of sparse SAR mode as described above.

[0069] Beneficial effects: Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention suppresses the azimuth ambiguity of SAR based on waveform diversity alternating transmission mode and further weakens the residual range ambiguity, thus alleviating the contradiction between azimuth ambiguity suppression and range ambiguity suppression in the design of traditional SAR systems; The present invention is based on image domain processing, which does not require prior knowledge of system parameters and specific transmission waveform information, and has the characteristics of simple implementation and strong applicability. Attached Figure Description

[0070] Figure 1 This is a flowchart of the present invention;

[0071] Figure 2 A schematic diagram is set up for the simulation experiment target;

[0072] Figure 3 The MF imaging results are shown when five NLFM waveforms are transmitted alternately.

[0073] Figure 4 A schematic diagram illustrating the intensity of ambiguity analysis between the target and its orientation;

[0074] Figure 5The imaging results of this invention are used when five NLFM waveforms are transmitted alternately;

[0075] Figure 6 The upsampling result diagram of this invention is used when alternately transmitting five NLFM waveforms;

[0076] Figure 7 The MF imaging result is shown when a single NLFM waveform is emitted.

[0077] Figure 8 The MF imaging results are shown when two NLFM waveforms are transmitted alternately.

[0078] Figure 9 The imaging results of this invention are used when two NLFM waveforms are transmitted alternately. Detailed Implementation

[0079] The invention will now be further described with reference to the accompanying drawings.

[0080] like Figure 1 As shown, this invention provides a two-dimensional ambiguity suppression method for waveform diversity alternating transmission sparse synthetic aperture radar modes, comprising the following steps:

[0081] Step 1: Generate nonlinear frequency modulation (NLFM) waveform diversity with different power spectral densities based on the asymmetric piecewise linear (PWL) function and the augmented Lagrangian generative algorithm (ALGA).

[0082] Baseband radar signals can be represented as:

[0083] (1)

[0084] in, Represents the pulse duration. It represents fast time. For signal phase. Select a weighting function with peak-to-sidelobe ratio (PSLR) between -30 dB and -40 dB. To generate the initial NLFM waveform:

[0085] (2)

[0086] Instantaneous frequency modulation , for about Functions:

[0087] (3)

[0088] Where, constant , This refers to the bandwidth of the transmitted signal.

[0089] use The asymmetric PWL function of order 1 reconstructs the NLFM wave generated in step 1-1, resulting in a new positive instantaneous frequency. and negative instantaneous frequency They are respectively:

[0090] (4)

[0091] (5)

[0092] in, and Represents the start / end time of each waveform segment. and This represents the frequency corresponding to the end of each waveform segment. and The modulation frequency of each waveform segment:

[0093] (6)

[0094] (7)

[0095] Based on the instantaneous frequency given by formulas (4) and (5) and formula (2), an NLFM waveform based on the asymmetric PWL function is generated.

[0096] Make the duration of each waveform equal, that is , , Through optimization and common The goal is to optimize the NLFM waveform by considering several parameters to reduce sidelobes without broadening the impulse response width (IRW). This problem can be formulated as a nonlinear constrained optimization problem:

[0097] (8)

[0098] in, , This represents a nonlinear inequality constraint. The generated NLFM waveform is optimized using ALGA; the optimization problem is equivalent to:

[0099] (9)

[0100] in, For the fitness function, For the Lagrange multiplier estimates, The offset is non-negative.

[0101] Step 2: Use the Range-Doppler Algorithm (RDA) to image the echo data of the NLFM waveform diversity alternating transmission mode SAR to generate the SAR image of the observation scene.

[0102] The three main operations in constructing RDA are distance compression operators. Distance migration correction operator and azimuth compression operator Alternating firing NLFM waveforms The distance compression operator can be expressed as:

[0103] (10)

[0104] in, , For the first One transmitted waveform, Indicates the period index of the transmitted waveform. This represents the number of samples taken in the azimuth direction. , , and These represent the Fourier transform operators and inverse Fourier transform operators for the azimuth and range directions, respectively. Based on the constructed operators, the matched filter (MF) process... It can be represented as:

[0105] (11)

[0106] in, For Hadamard product. For SAR echo data in alternating transmission mode. Perform MF processing:

[0107] (12)

[0108] in, SAR image of the target observation scene.

[0109] Step 3: Using SAR images, construct a sparse SAR imaging model based on incomplete Doppler spectrum.

[0110] Constructing a sparse SAR imaging model based on incomplete Doppler spectrum:

[0111] (13)

[0112] in, Represents image reconstruction using the MF imaging algorithm Backscattering coefficient of the target observation scene The differences between them For rectangular window functions, the specific expression is as follows:

[0113] (14)

[0114] in, The Doppler bandwidth of the echo signal. For Doppler frequency, The center frequency of the Doppler wave. The scaling factor is used to control the range of the observed Doppler spectrum.

[0115] Step 4: The sparse SAR imaging model based on incomplete Doppler spectrum is transformed into an optimization reconstruction problem based on L1 norm regularization.

[0116] For the sparse SAR imaging model constructed by formula (13), scene reconstruction can be achieved by solving the following L1 regularization problem:

[0117] (15)

[0118] in, It is a reconstructed, unambiguous, sparse SAR image. Regularization parameters are used to control the sparsity of the target observation scene.

[0119] Step 5: Solve the optimization and reconstruction problem in Step 4 using the threshold iterative recovery algorithm to achieve unambiguous sparse reconstruction of the observed scene.

[0120] To address the optimization problem in step 4, unambiguous sparse reconstruction of the observed scene can be achieved through iterative recovery. The threshold iterative algorithm is input to the SAR image obtained in step 2. and the window function in step 3 Let's define the target scenario for sparse reconstruction. The initial value is 0, and the iteration parameter is... The error parameter is The maximum number of iterations is When the condition is met, the number of iterations is [number]. And residuals When needed, perform the following steps.

[0121] 1) Estimate the residual data values:

[0122] (16)

[0123] (17)

[0124] 2) Update gradient values:

[0125] (18)

[0126] in, This is a parameter that controls the convergence speed of the algorithm, and it is usually set to a constant.

[0127] 3) Update the parameters controlling the sparsity of the observation scene for the target. :

[0128] (19)

[0129] in, Indicates amplitude value Sort in descending order The largest component.

[0130] 4) Threshold shrinkage in target observation scenarios:

[0131] (20)

[0132] 5) Estimate the residuals of the reconstructed image:

[0133] (twenty one)

[0134] If satisfied and residual If so, continue iterating, that is... Repeat steps 1) to 5). If the condition is not met, end the iteration and output the recovered sparse unblurred image. .

[0135] The present invention also provides an apparatus comprising a memory and a processor, wherein: the memory is configured to store a computer program capable of running on the processor; and the processor is configured to, when running the computer program, execute the steps of the two-dimensional blur suppression method for waveform diversity alternating transmission of sparse SAR modes as described above.

[0136] The present invention also provides a storage medium storing a computer program, which, when executed by at least one processor, implements the steps of the two-dimensional blur suppression method for waveform diversity alternating transmission of sparse SAR modes as described above.

[0137] To verify the feasibility and effectiveness of this invention, comparative experiments were conducted from two dimensions: orientation ambiguity suppression and distance ambiguity suppression. Firstly, as... Figure 2 As shown, three point targets T1, T2, and T3 were set in the observation scene with intensities of 0.06, 0.56, and 1, respectively, with T2 located at the center of the scene. Five different NLFM waveforms were designed and emitted alternately, and the MF imaging results are shown below. Figure 3 As shown (the three target points are marked in yellow), Figure 4 Further annotations were added. Figure 3 The results obtained by processing the target T1 intensity and T3 first-azimuth ambiguity intensity using the sparse SAR ambiguity suppression method of this invention are as follows: Figure 5 As shown, Figure 6 for Figure 5 Local upsampling results. Secondly, to assess the interference caused by the first distance blur, a single point target was set at the center of the observation scene. Figure 7 The image shows the MF imaging results when a single NLFM waveform is emitted (the yellow box indicates the single point target). Figure 8 The results of MF imaging after alternating transmission of two NLFM waveforms are shown. Figure 9 The results obtained using the method of this invention are shown (the black box shows the magnified result of the point target). Experimental results show that significant azimuth blurring exists in the MF imaging results when five NLFM waveforms are alternately transmitted, and the azimuth blurring intensity of T3 is higher than that of target T1. The method of this invention can suppress all azimuth blurring and recover target T1. Furthermore, compared to a single waveform, alternating waveform transmission can reduce range blurring, while the method of this invention can further weaken residual range blurring while suppressing azimuth blurring, achieving joint suppression of two-dimensional range-azimuth blurring.

[0138] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A two-dimensional ambiguity suppression method for sparse SAR modes using waveform diversity alternating transmission, characterized in that, Includes the following steps: (1) Use the asymmetric PWL function and ALGA to generate NLFM waveform diversity with different power spectral densities; (2) Perform MF imaging on the echo data of NLFM waveform diversity alternating transmission mode SAR to generate SAR image of the observation scene; (3) Using SAR images, construct a sparse SAR imaging model based on incomplete Doppler spectrum; (4) The sparse SAR imaging model based on incomplete Doppler spectrum is transformed into an optimization reconstruction problem based on L1 norm regularization; (5) Solve the optimization and reconstruction problem in step (4) by using the threshold iterative recovery algorithm to achieve unambiguous sparse reconstruction of the observed scene.

2. The two-dimensional ambiguity suppression method for waveform diversity alternating transmission sparse SAR mode according to claim 1, characterized in that, The implementation process of step (1) is as follows: The baseband radar signal is: (1) in, Represents the pulse duration. It represents fast time. For signal phase; select a weighting function with a peak-to-side-lobe ratio between -30dB and -40dB. Generate the initial NLFM waveform: (2) Instantaneous frequency modulation , for about Functions: (3) Where, constant , For the transmission signal bandwidth; use The asymmetric PWL function of order 1 reconstructs the NLFM wave generated in step (11), resulting in a new positive instantaneous frequency. and negative instantaneous frequency They are respectively: (4) (5) in, and Represents the start / end time of each waveform segment. and This represents the frequency corresponding to the end of each waveform segment. and The modulation frequency of each waveform segment: (6) (7) Based on the instantaneous frequency given by formulas (4) and (5) and formula (2), generate the NLFM waveform based on the asymmetric PWL function; Make the duration of each waveform equal, that is , , Through optimization and common The problem involves optimizing the NLFM waveform with several parameters to reduce sidelobes without broadening the impulse response width. This problem is formulated as a nonlinear constrained optimization problem. (8) in, , This represents a nonlinear inequality constraint; the generated NLFM waveform is optimized using ALGA, and the optimization problem is equivalent to: (9) in, For the fitness function, For the Lagrange multiplier estimates, The offset is non-negative.

3. The two-dimensional ambiguity suppression method for waveform diversity alternating transmission sparse SAR mode according to claim 1, characterized in that, The implementation process of step (2) is as follows: Constructing distance compression operators in RDA Distance migration correction operator and azimuth compression operator Alternating firing NLFM waveforms The distance compression operator is expressed as follows: (10) in, , For the first One transmitted waveform, Indicates the period index of the transmitted waveform. This represents the number of samples in the azimuth direction. , , and These represent the Fourier transform operators and inverse Fourier transform operators for the azimuth and range directions, respectively; based on the constructed operators, the matched filtering process... Represented as: (11) in, For Hadamard product; for SAR echo data in alternating transmission mode Perform MF processing: (12) in, SAR image of the target observation scene.

4. The two-dimensional ambiguity suppression method for waveform diversity alternating transmission sparse SAR mode according to claim 1, characterized in that, The implementation process of step (3) is as follows: Constructing a sparse SAR imaging model based on incomplete Doppler spectrum: (13) in, Represents image reconstruction using the MF imaging algorithm Backscattering coefficient of the target observation scene The differences between them For rectangular window functions, the specific expression is as follows: (14) in, The Doppler bandwidth of the echo signal. For Doppler frequency, The center frequency of the Doppler wave. The scaling factor is used to control the range of the observed Doppler spectrum.

5. The two-dimensional ambiguity suppression method for waveform diversity alternating transmission sparse SAR mode according to claim 1, characterized in that, The implementation process of step (4) is as follows: For the constructed sparse SAR imaging model, scene reconstruction is achieved by solving the following L1 regularization problem: (15) in, It is a reconstructed, unambiguous, sparse SAR image. Regularization parameters are used to control the sparsity of the target observation scene.

6. The two-dimensional ambiguity suppression method for waveform diversity alternating transmission sparse SAR mode according to claim 1, characterized in that, The implementation process of step (5) is as follows: Unambiguous sparse reconstruction of the observed scene is achieved through iterative recovery; the threshold iterative algorithm is input to the SAR image obtained in step (2). and the window function obtained in step (3) ; Define the target scenario for sparse reconstruction. The initial value is 0, and the iteration parameter is... The error parameter is The maximum number of iterations is When the condition is met, the number of iterations is [number]. And residuals When this happens, perform the following steps: S1, Estimated residual data values: (16) (17) S2, Update gradient values: (18) in, This is a parameter that controls the convergence speed of the algorithm, and is usually set to a constant; S3, Update the parameters controlling the sparsity of the target observation scene. : (19) in, Indicates amplitude value The numbers after sorting in descending order The largest component; S4, Threshold shrinkage of the target observation scene: (20) S5, Estimate the residuals of the reconstructed image: (21) If satisfied and residual If so, continue iterating, that is... Repeat steps S1 to S5; if the condition is not met, end the iteration and output the recovered sparse unblurred image. .

7. A device, characterized in that, Includes memory and processor, wherein: Memory is used to store computer programs that can run on a processor; A processor, configured to, while running the computer program, perform the steps of the two-dimensional blur suppression method for waveform diversity alternating transmission of sparse SAR modes as described in any one of claims 1 to 6.

8. A storage medium, characterized in that, The storage medium stores a computer program that, when executed by at least one processor, implements the steps of the two-dimensional ambiguity suppression method for waveform diversity alternating transmission of sparse SAR modes as described in any one of claims 1 to 6.

Images