A SAR-MMTI clutter suppression method based on frequency diverse array

Abstract

A clutter suppression method for SAR-MMTI based on a frequency diverse array, which relates to the field of radar technology. The present invention is to solve the problem that the existing methods do not have the ability to resolve range ambiguity under wide-swath conditions and cannot be applied to SAR radar systems for clutter suppression. The present invention makes full use of the relationship between the carrier frequency difference and the PRF, avoiding the problems of inability to resolve ambiguity and poor ambiguity resolution performance. At the same time, for the time-varying characteristics of the transmitting pattern of the frequency diverse array, the quasi-static transmitting beamforming conditions are analyzed. By using the echo signals from different transmitting units, the signal power is increased. The present invention can match the echo characteristics of SAR radars with a large number of coherent integration pulses, suppress range-ambiguous clutter at the same time, and further improve the signal-to-clutter ratio and target detection performance. It can be used for airborne radar ground / sea moving target detection.

Description

Technical Field

[0001] The invention belongs to the technical field of radar. Background Art

[0002] Synthetic aperture radar (SAR) has the ability to observe the earth around the clock and in all weather conditions, and has unique advantages in applications such as ocean monitoring, environmental monitoring, mapping, and military. After years of development, the synthetic aperture radar system has gradually shifted from a low-resolution, narrow-band imaging mode to a high-resolution, wide-band imaging mode. The wide-band SAR imaging system based on a frequency diversity array overcomes the contradiction that the traditional phased array antenna SAR cannot simultaneously meet the Doppler and range unambiguity requirements when selecting PRF (pulse repetition frequency). It uses tiny carrier frequency increments between array elements to generate a range- and angle-dependent emission pattern, which in turn provides favorable conditions for resolving range ambiguity.

[0003] The SAR-MMTI system, which combines the SAR system with the marine moving target indication technology (MMTI), realizes the integration of imaging and moving target detection, and can monitor the surface ships in real time, providing all-round and effective information for marine traffic control. Before using the SAR echo based on the frequency diversity array to detect the surface target, it is necessary to suppress the clutter in the scene to improve the signal-to-clutter ratio and the detection performance of the target. However, the traditional STAP (space-time adaptive processing) method is mainly used in pulse Doppler radar, and its signal-to-clutter ratio improvement ability is limited. It is not suitable for SAR radar systems with more coherent accumulation pulses. More importantly, it does not have the ability to resolve distance ambiguity under wide-band conditions. Summary of the invention

[0004] The present invention aims to solve the problem that the existing methods are not capable of resolving range ambiguity under wide-band conditions and cannot be applied to SAR radar systems for clutter suppression. A SAR-MMTI clutter suppression method based on a frequency diversity array is now provided.

[0005] A SAR-MMTI clutter suppression method based on frequency diversity array is proposed. The spatial rectangular coordinate system XYZ is established with the radar height direction as the Z axis, the moving direction as the Y axis, and the radar mapping width direction as the X axis. u To divide the X-axis into multiple range ambiguity zones for spacing, the antennas on the radar are arranged in a frequency diversity array;

[0006] The SAR-MMTI clutter suppression method includes the following steps:

[0007] Step 1: Make M transmitting array element antennas transmit orthogonal signals simultaneously, and N receiving array element antennas receive M reflected echo signals, where M and N are both positive integers;

[0008] Step 2: Perform de - carrier processing on the echo signals received by each receiving array element antenna respectively, and separate the M echo signals after frequency conversion using M matched filters;

[0009] Step 3: Discretize the distance from the radar to the detection target along the range dimension of the echo signal into N r range cells, and perform range - dependence compensation on each of the separated echo signals in sequence according to the range cells using a compensation factor, so that the transmitted spatial frequencies of the echo signals within the same range ambiguity region are the same;

[0010] Step 4: Perform range migration correction on the echo signals after range - dependence compensation, and then perform discrete Fourier transform on the corrected echo signals in the azimuth dimension of the echo signals, so that the echo signals are converted to the range - Doppler domain;

[0011] Step 5: Discretize the Doppler frequency in the range - Doppler domain into N a Doppler cells;

[0012] Step 6: Construct the transmit and receive steering vectors of the range ambiguity region of interest within the d - th Doppler cell, and synthesize the transmit and receive steering vectors into the composite steering vector of the detection target in the d - th Doppler cell, where d = 1, 2,..., N a ;

[0013] Step 7: Select the echo signals converted to the range - Doppler domain in 20 range cells adjacent to the i - th range cell as the training sample set, where i = 1, 2,..., N r and remove the samples containing non - uniform clutter and interference in the training sample set to obtain the final sample set of the i - th range cell;

[0014] Step 8: Estimate the clutter - free covariance matrix of the i - th range cell using the final sample set of the i - th range cell, calculate the optimal adaptive weight vector of the detection target in the d - th Doppler cell using the composite steering vector of the detection target in the d - th Doppler cell and the clutter - free covariance matrix of the i - th range cell, and use this optimal adaptive weight vector to suppress the clutter in the i - th range cell;

[0015] Step 9: Judge whether i < N r is satisfied. If so, set i = i + 1 and return to Step 7, otherwise execute Step 10;

[0016] Step 10: Judge whether d < N a is satisfied. If so, set d = d + 1 and return to Step 6, otherwise complete the suppression of clutter in SAR - MMTI.

[0017] Further, the above SAR-MMTI clutter suppression method based on frequency diverse array further includes:

[0018] Step Eleven: Perform azimuth compression on the echo signals after clutter suppression in all Doppler cells to complete the image reconstruction of the detected target.

[0019] Further, the conditions satisfied by the carrier frequency difference Δf of the orthogonal signals transmitted by adjacent transmitting array elements include:

[0020]

[0021]

[0022] Δφ = 2π(M - 1)·Δf·T << 2π,

[0023] where ΔR is the slope range observed by the radar, c is the speed of light, T is the pulse width of the orthogonal signal transmitted by the radar, PRF is the pulse repetition frequency, z represents the integer part, μ represents the decimal part, and μ ≥ 1 / M, and Δφ is the phase difference between the first and the Mth transmitting array elements.

[0024] Further, the impulse response h m (τ) of the mth matched filter at time τ is:

[0025]

[0026] And it satisfies the following orthogonality condition:

[0027]

[0028] where is the orthogonal signal transmitted by the mth transmitting array element at time τ, represents the conjugate of, is the orthogonal signal transmitted by the m'th transmitting array element at time τ - t, Δf is the carrier frequency difference of the orthogonal signals transmitted by adjacent transmitting array elements, m' = 1, 2,..., M, m = 1, 2,..., M, m ≠ m', and j is the imaginary flag, represents any time.

[0029] Further, the compensation factor h i of the ith range cell is constructed according to the following formula:

[0030]

[0031] where c is the speed of light, R i is the slant range of the ith range cell, [·] Trepresents the transpose of a matrix, Δf is the carrier frequency difference of the orthogonal signals transmitted by adjacent transmitting array elements, and j is the imaginary flag.

[0032] Further, the transmit and receive steering vectors for the range ambiguity region of interest within the d-th Doppler cell are respectively:

[0033]

[0034]

[0035] where, a T (f d , p) and a R (f d ) are respectively the transmit and receive steering vectors for the range ambiguity region of interest within the d-th Doppler cell, f d is the Doppler frequency of the detected target within the d-th Doppler cell, p is the sequence number of the range ambiguity region where the detected target is located, f dc is the Doppler frequency shift caused by the radial velocity of the detected target, d T is the spacing between adjacent two transmitting array element antennas, d R is the spacing between adjacent two receiving array element antennas, v is the radar moving speed, v a is the azimuth velocity of the detected target, Δf is the carrier frequency difference of the orthogonal signals transmitted by adjacent transmitting array elements, and c is the speed of light;

[0036] The composite steering vector of the detected target in the d-th Doppler cell is:

[0037]

[0038] where, represents the Kronecker product.

[0039] Further, when the sample x in the training sample set satisfies the following formula, it is excluded as a sample containing non-uniform clutter and interference:

[0040]

[0041] where, R c is the clutter covariance matrix estimated from the training sample set, η is the decision threshold and η = αNM, α is a coefficient greater than 1, (·) H represents the conjugate transpose of a matrix.

[0042] Further, the expression of the clutter-free covariance matrix R c (f d ) for the i-th range cell is:

[0043]

[0044] Among them, E[·] represents the expectation operation, and C(i, f d ) is the clutter matrix of f in the i-th range cell, and f d is the Doppler frequency of the detected target in the d-th Doppler cell. d

[0045] Furthermore, under the condition of the maximum output signal-to-noise ratio, the optimal adaptive weight vector W opt (f d , p) of the detected target in the d-th Doppler cell is:

[0046]

[0047] Among them, d(f d , p) is the synthetic steering vector of the detected target in the d-th Doppler cell.

[0048] Furthermore, the echo signal after clutter suppression is:

[0049]

[0050] The echo signal after azimuth dimension compression is:

[0051]

[0052] Among them, W opt (f d , p) is the optimal adaptive weight vector of the detected target in the d-th Doppler cell, is the range-Doppler domain echo signal after range migration correction, and (·) H represents the conjugate transpose of the matrix, H(f d ) is the impulse response of the azimuth dimension matching filter, f d is the Doppler frequency of the detected target in the d-th Doppler cell, p is the serial number of the range ambiguity region where the detected target is located, t m is the azimuth dimension time, is the fast time of the range dimension.

[0053] The beneficial effects of the radar signal adaptive processing method based on a frequency diversity array according to the present invention are as follows:

[0054] 1. By making full use of the relationship between the carrier frequency difference and PRF, the problems of ambiguity resolution failure and poor ambiguity resolution performance are avoided. Meanwhile, for the time-varying characteristics of the transmitting pattern of the frequency diversity array, the quasi-static transmitting beamforming conditions are analyzed. Finally, the relationship between the main lobe of the pattern and the observed slant range is considered. This provides guidance for the theoretical research of the frequency diversity array in radar imaging and the design of transmitting signal parameters.

[0055] 2. A space-time adaptive three-dimensional processing is carried out in the Doppler-receiving element-transmitting element domain according to the characteristics of SAR echoes. Compared with the clutter suppression method in the conventional SAR system, the method of the present invention makes full use of the range dimension degrees of freedom provided by the frequency diversity system and can independently suppress the clutter in multiple ambiguous regions. At the same time, for a single ambiguous region, the echo signals from different transmitting units are utilized, and coherent superposition can be achieved at the target, and the signal power can be further increased by 2M times. Since the clutter power has relatively weak correlation, the improvement amplitude is relatively small, thereby increasing the signal-to-clutter ratio and enhancing the detection performance of the target.

[0056] In summary, the present invention can match the echo characteristics of the SAR radar with a large number of coherent accumulation pulses, simultaneously suppress the range ambiguous clutter, and further improve the signal-to-clutter ratio and the target detection performance. It can be used for the detection of ground / seasurface moving targets by airborne radars and has good application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 is a schematic diagram of the usage scenario of the present invention;

[0058] Figure 2 is a flowchart of the present invention;

[0059] Figure 3 are the transmitting patterns corresponding to the resolvable ambiguity cases when designing the carrier frequency increment and the transmitting spatial frequency distribution diagrams of different range ambiguous regions, where (a) is the case of Δf < PRF and (b) is the case of Δf > PRF;

[0060] Figure 4 are the transmitting patterns corresponding to the non-resolvable ambiguity cases when designing the carrier frequency increment and the transmitting spatial frequency distribution diagrams of different range ambiguous regions;

[0061] Figure 5 are the imaging result diagrams of two range ambiguous regions under the frequency diversity system, where (a) represents ambiguous region 1 and (b) represents ambiguous region 2;

[0062] Figure 6 is for Figure 5 are the imaging result diagrams of adding sea clutter with sea state level 2 on the basis of

[0063] Figure 7To add the imaging result diagrams under sea state 5 clutter on the basis of Figure 5 (a) represents the ambiguous area 1 and (b) represents the ambiguous area 2;

[0064] Figure 8 The figure is the result diagram after range ambiguity clutter suppression by using the method of the present invention under sea state 2. (a) represents the ambiguous area 1 and (b) represents the ambiguous area 2;

[0065] Figure 9 The figure is the result diagram after range ambiguity clutter suppression by using the method of the present invention under sea state 5. (a) represents the ambiguous area 1 and (b) represents the ambiguous area 2;

[0066] Figure 10 The figure is the improvement factor curve diagram after clutter suppression in the two ambiguous areas. (a) represents the ambiguous area 1 and (b) represents the ambiguous area 2. Detailed implementation manners

[0067] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention. It should be noted that, without conflict, the embodiments in the present invention and the features in the embodiments may be combined with each other.

[0068] The SAR system accumulates a sufficient number of pulses to improve the cross-range resolution. At the same time, the echo in each slow-time dimension will contain signals transmitted at different times, irradiated to different range ambiguity areas and then returned. For a conventional SAR system, it is impossible to separate the echoes from different ambiguity areas. For the SAR-MMTI system under the frequency diversity architecture, the current processing methods cannot adapt to a large number of accumulated pulses. Therefore, the method described in the present invention is used for processing. The clutter suppression method includes the following steps:

[0069] Detailed implementation manner 1: As Figure 1 shown, a spatial rectangular coordinate system XYZ is established with the height direction of the radar as the Z axis, the moving direction as the Y axis, and the mapping width direction of the radar as the X axis. The mapping width covered by the radar pattern is divided into multiple range ambiguity areas with the maximum unambiguous range R u as the spacing. The antennas on the radar are arranged in a frequency diversity array manner. The radar moves uniformly along the Y axis (azimuth axis) at a speed v and irradiates the mapping area at a squint angle. Point P is an arbitrary point in the scene as the detection target, v r represents the radial velocity of the detection target, v a represents the azimuth velocity of the detection target, RB Indicates the closest distance from the radar to the detection target.

[0070] A SAR-MMTI clutter suppression method based on a frequency diverse array according to this embodiment includes the following steps:

[0071] Step 1: Make M transmitting array element antennas transmit orthogonal signals simultaneously, and N receiving array element antennas all receive M reflected echo signals, that is, each receiving array element antenna receives M echo signals. Both M and N are positive integers. For a pulse modulation radar, each echo signal is stored row by row. For example, the echo of the first pulse is placed in the first row, and the echo of the second pulse is placed in the second row, thus forming a two-dimensional data matrix. The dimension defined in the row direction is the range dimension, and the dimension defined in the column direction is the azimuth dimension.

[0072] The carrier frequency differences of different orthogonal signals need to consider the following conditions:

[0073] First, consider the relationship between the carrier frequency difference Δf of the orthogonal signals transmitted by adjacent transmitting array element antennas and the pulse repetition frequency PRF. Since in the FDA (Frequency Diverse Array)-MIMO system, the signals transmitted by the transmitting array element antennas are orthogonal to each other, and only an equivalent pattern can be formed at the receiving end. The distance is the round-trip of transmission and reception. At this time, the distance change period is corrected to W s = c / 2Δf (c is the speed of light), and at the same time, the maximum unambiguous distance R of the radar u = c / 2PRF. Therefore, according to the value of Δf, the relative sizes of R u and W s can be adjusted. The relationship between the frequency step Δf and PRF is expressed as:

[0074]

[0075] where z represents the integer part, μ represents the fractional part, and μ≥1 / M.

[0076] If z = 0, that is, Δf < PRF, then as shown in Figure 3 (a), one pattern period will contain one or even multiple range ambiguity regions. If z > 0, that is, Δf > PRF, then as shown in Figure 3 (b), one ambiguity period will span multiple pattern periods. In either case, the prerequisite for unambiguous resolution is that the system can detect and distinguish the transmitted spatial frequencies corresponding to different ambiguity periods. If, as shown in Figure 4 the transmitted spatial frequencies of different ambiguity regions overlap, then range unambiguous resolution cannot be achieved. Therefore, the fractional part should be designed to be at least 1 / M.

[0077] Secondly, consider the quasi-static transmit beamforming condition. The pattern of a frequency diverse array antenna has time-varying characteristics. For a system operating in a pulsed regime such as SAR, if the maximum phase difference between the frequency diverse arrays is much less than 2π within the pulse duration, it can be approximately considered that its transmit pattern is time-invariant within the pulse duration. During the pulse duration, let the phase difference Δφ between the first and the M-th transmit array elements be:

[0078] Δφ = 2π(M - 1)·Δf·T << 2π,

[0079] where T is the pulse width of the orthogonal signal transmitted by the radar.

[0080] Finally, consider the slant range over which the main lobe of the transmit beam persists within the pulse width. When the slope range observed by the radar is ΔR, then:

[0081]

[0082] Furthermore, it can be obtained that:

[0083]

[0084] Step 2: Before transmitting the signal, the radar needs to modulate the signal frequency from the base frequency to a specific frequency, called the carrier frequency. Therefore, when receiving the signal, conversely, the signal is returned from the carrier frequency to the base frequency. This process is called down-conversion processing or de-carrier processing. The de-carrier processing is performed on the echo signals received by each receiving array element antenna, and M matched filters are used to separate the M echo signals from different transmitting array element antennas after the frequency conversion processing.

[0085] where the impulse response h m (τ) of the m-th matched filter at time τ is:

[0086]

[0087] and satisfies the following orthogonality condition:

[0088]

[0089] where is the orthogonal signal transmitted by the m-th transmit array element antenna at time τ, denotes the conjugate of, is the orthogonal signal transmitted by the m'-th transmit array element antenna at time τ - t, Δf is the carrier frequency difference of the orthogonal signals transmitted by adjacent transmit array element antennas, m' = 1, 2,..., M, m = 1, 2,..., M, m ≠ m', j is the imaginary flag, denotes any time.

[0090] Step 3: First, construct the compensation factor for the range cells, and then discretize the range from the radar to the detected target along the range dimension of the echo signal into N r range cells. Finally, use the compensation factor to perform range-dependent compensation on each separated echo signal in sequence according to the range cells, so that the transmitted spatial frequencies of the echo signals within the same range ambiguity region are the same, and at the same time, it is convenient to separate the targets and clutter in different range ambiguity regions.

[0091] Among them,

[0092] the compensation factor h of the i-th range cell i is constructed according to the following formula:

[0093]

[0094] where R i is the slant range of the i-th range cell, and [·] T represents matrix transpose.

[0095] Step 4: Perform range migration correction on the echo signal after range-dependent compensation, and then perform discrete Fourier transform on the corrected echo signal along the azimuth dimension of the echo signal, so that the echo signal is converted to the range-Doppler domain.

[0096] Step 5: Discretize the Doppler frequency in the range-Doppler domain into N a Doppler cells.

[0097] Step 6: In the Doppler-receiving array element-transmitting array element domain, the steering vector compensates for the phase differences of the echo signals of moving targets in different channels. For the receiving steering vector, the phase difference is caused by the baseline length between different receiving channels and the radial velocity of the moving target; for the transmitting steering vector, the phase difference is related not only to the baseline length of different transmitting channels and the target radial velocity, but also to the range ambiguity region where the target is located. Construct the transmitting and receiving steering vectors of the range ambiguity region of interest within the d-th Doppler cell, and synthesize the transmitting and receiving steering vectors into the composite steering vector of the detected target in the d-th Doppler cell, d = 1, 2,..., N a .

[0098]

[0099]

[0100] where a T (f d , p) and a R (f d ) are the transmitting and receiving steering vectors of the range ambiguity region of interest within the d-th Doppler cell respectively, and f dTo detect the Doppler frequency of the target in the d-th Doppler cell, p is the serial number of the range ambiguity region where the target is located, and f dc is the Doppler frequency shift caused by the radial velocity of the detected target, and d T is the spacing between adjacent transmitting array element antennas, and d R is the spacing between adjacent receiving array element antennas, v is the radar moving speed, and v a is the azimuth velocity of the detected target, Δf is the carrier frequency difference of the orthogonal signals transmitted by adjacent transmitting array element antennas, and c is the speed of light.

[0101] The composite steering vector of the detected target in the d-th Doppler cell is:

[0102]

[0103] Among them, represents the Kronecker product.

[0104] Step Seven: Select the echo signals transformed into the range-Doppler domain in 20 range cells adjacent to the i-th range cell as the training sample set of the covariance matrix, where i = 1, 2,..., N r .

[0105] When the sample x in the training sample set satisfies the following formula, it is excluded as a sample containing non-uniform clutter and interference, and the final sample set of the i-th range cell is obtained.

[0106]

[0107] Among them, R c is the clutter covariance matrix estimated from the training sample set, η is the decision threshold and η = αNM, α is a coefficient greater than 1, and (·) H represents the conjugate transpose of the matrix.

[0108] Step Eight: After screening the samples, according to the clutter signal model, use the final sample set of the i-th range cell to estimate the clutter-free covariance matrix R c (f d ):

[0109]

[0110] Among them, E[·] represents the expectation operation, C(i, f d ) is the clutter matrix of f d in the i-th range cell, and f d is the Doppler frequency of the detected target in the d-th Doppler cell.

[0111] Using the composite steering vector d(f of the detected target in the d-th Doppler celld , p) and the clutter covariance matrix R of the i-th range cell c (f d ) to calculate the optimal adaptive weight vector W of the detected target in the d-th Doppler cell opt (f d , p):

[0112]

[0113] Finally, use the optimal adaptive weight vector W of the detected target in the d-th Doppler cell opt (f d , p) to suppress the clutter in the i-th range cell. The echo signal after clutter suppression is:

[0114]

[0115] is the range-Doppler domain echo signal after range migration correction.

[0116] Step Nine: Determine whether i < N is satisfied r . If yes, set i = i + 1 and return to Step Seven. Otherwise, execute Step Ten.

[0117] Step Ten: Determine whether d < N is satisfied a . If yes, set d = d + 1 and return to Step Six. Otherwise, complete the clutter suppression of SAR-MMTI.

[0118] Specific Embodiment Two: This embodiment further describes a method for suppressing clutter in SAR-MMTI based on a frequency diversity array described in Specific Embodiment One. This embodiment further includes Step Eleven: Perform azimuth compression on the echo signals after clutter suppression in all Doppler cells to complete the image reconstruction of the detected target.

[0119] The echo signal after azimuth compression is:

[0120]

[0121] H(f d ) is the impulse response of the azimuth matching filter, t m is the azimuth time, is the fast time in the range dimension. Specific Example

[0123] Simulation parameters: For the SAR-MMTI system based on frequency diverse array, the carrier frequency f0 = 9.6 GHz, the pulse repetition frequency PRF is 2000 Hz, the carrier frequency difference Δf is 2500 Hz, the radar speed v is 200 m / s, the radar altitude is 5000 m, the system bandwidth is 100 MHz, the pulse width is 2 μs, and the number of transmitting and receiving channels is 8 each. In two range ambiguity regions, 2 moving targets, 1 stationary target and 1 moving target, 1 stationary target are respectively set.

[0124] Simulation content:

[0125] Simulation 1, under the above simulation parameters, directly image the targets in the scene. The results are as Figure 5 shown.

[0126] Simulation 2, under the above simulation parameters, add 2-level and 5-level sea clutter that follows the K-distribution amplitude model to the scene respectively, and the signal-to-clutter ratios after pulse compression are set to 5 dB and -5 dB respectively. The results are as Figure 6 and Figure 7 shown.

[0127] It can be seen from Figure 6 and Figure 7 that the positions of the targets become gradually blurred under the clutter background, and the positions of the targets cannot be seen under the 5-level sea state.

[0128] Simulation 3, under the above simulation parameters, use the method of the present invention to suppress the sea clutter and stationary targets in the scene, and the results are as Figure 8 and Figure 9 shown.

[0129] Comparing Figure 6 and Figure 8 、 Figure 7 and Figure 9 it can be seen that the method of the present invention can effectively suppress the sea clutter and stationary targets in the scene, effectively improve the signal-to-clutter ratio, and highlight the moving targets.

[0130] Although the present invention has been described herein with reference to specific embodiments, it should be understood that these embodiments are merely examples of the principles and applications of the present invention. Therefore, it should be understood that many modifications can be made to the exemplary embodiments, and other arrangements can be designed, as long as they do not depart from the spirit and scope of the present invention as defined by the appended claims. It should be understood that the different dependent claims and the features described herein can be combined in a manner different from that described in the original claims. It should also be understood that the features described in connection with a single embodiment can be used in other described embodiments.

Claims

1. A clutter suppression method for SAR-MMTI based on frequency diverse array. A spatial rectangular coordinate system XYZ is established with the altitude direction of the radar as the Z-axis, the moving direction as the Y-axis, and the mapping width direction of the radar as the X-axis. The maximum unambiguous range R of the radar u is used as the spacing to divide the X-axis into multiple range ambiguity regions. The antennas on the radar are arranged in the form of a frequency diverse array It is characterized in that The SAR-MMTI clutter suppression method includes the following steps: Step 1: Make M transmitting array element antennas simultaneously transmit orthogonal signals, and N receiving array element antennas all receive M reflected echo signals. Both M and N are positive integers; Step 2: Perform de-carrier frequency processing on the echo signals received by each receiving array element antenna respectively, and use M matched filters to separate the M echo signals after frequency conversion processing; Step 3: Discretize the distance from the radar to the detection target into N r range cells along the range dimension of the echo signal, and use the compensation factor to perform range-dependent compensation on each of the separated echo signals in sequence according to the range cells, so that the transmitted spatial frequencies of the echo signals within the same range ambiguity region are the same; Step 4: Perform range migration correction on the echo signals after distance-dependence compensation, and then perform discrete Fourier transform on the azimuth dimension of the corrected echo signals, so that the echo signals are converted to the range-Doppler domain; Step Five: Discretize the Doppler frequency in the range-Doppler domain into N a Doppler cells; Step 6: Construct the transmit and receive steering vectors of the range ambiguity region of interest in the d-th Doppler cell, and synthesize the transmit and receive steering vectors into the composite steering vector of the detection target in the d-th Doppler cell, where d = 1, 2, ..., N a ; Step 7: Select the echo signals converted to the range-Doppler domain in the 20 range units adjacent to the i-th range unit as the training sample set, i = 1, 2, ..., N r , and remove the samples containing non-uniform clutter and interference in the training sample set to obtain the final sample set of the i-th distance unit; Step 8: Estimate the clutter-free covariance matrix of the i-th range cell by using the final sample set of the i-th range cell, calculate the optimal adaptive weight vector of the detection target in the d-th Doppler cell by using the synthetic steering vector of the detection target in the d-th Doppler cell and the clutter-free covariance matrix of the i-th range cell, and use this optimal adaptive weight vector to suppress the clutter in the i-th range cell; Step Nine: Determine whether i < N is satisfied r . If yes, set i = i + 1 and return to Step Seven; otherwise, execute Step Ten. Step Ten: Determine whether d < N is satisfied a . If yes, set d = d + 1 and return to Step Six; otherwise, complete the suppression of SAR-MMTI clutter.

2. The SAR-MMTI clutter suppression method based on a frequency diverse array according to claim 1, wherein It also includes: Step 11: Perform azimuth dimension compression on the echo signals after clutter suppression in all Doppler cells to complete the image reconstruction of the detection target.

3. A clutter suppression method for SAR-MMTI based on frequency diverse array according to claim 1 or 2, characterized in that, The carrier frequency difference Δf of the orthogonal signals transmitted by adjacent transmitting array element antennas satisfies the conditions including: Δφ = 2π(M - 1)·Δf·T << 2π, where ΔR is the slope range observed by the radar, c is the speed of light, T is the pulse width of the orthogonal signals transmitted by the radar, PRF is the pulse repetition frequency, z represents the integer part, μ represents the decimal part, and μ ≥ 1 / M, and Δφ is the phase difference between the first and the M-th transmitting array element antennas.

4. A method for suppressing clutter in SAR-MMTI based on a frequency diverse array according to claim 1 or 2, characterized in that, The impulse response h of the m-th matched filter at time τ m (τ) is as follows: And it satisfies the following orthogonal conditions: Among them, is the orthogonal signal transmitted by the m-th transmitting array element antenna at time τ, denotes the conjugate of, is the orthogonal signal transmitted by the m'-th transmitting array element antenna at time τ - t, Δf is the carrier frequency difference of the orthogonal signals transmitted by adjacent transmitting array element antennas, m' = 1, 2,..., M, m = 1, 2,..., M, m ≠ m', j is the imaginary flag, denotes any time.

5. A SAR-MMTI clutter suppression method based on a frequency diverse array according to claim 1 or 2, characterized in that, The compensation factor h for the i-th range cell i is constructed according to the following formula: where c is the speed of light, and R i is the slant range of the i-th range cell, [·] T denotes the transpose of a matrix, Δf is the carrier frequency difference of the orthogonal signals transmitted by adjacent transmitting array elements, and j is the imaginary flag.

6. A clutter suppression method for SAR-MMTI based on frequency diverse array according to claim 1 or 2, characterized in that The transmitting and receiving steering vectors in the range ambiguity region of interest in the d-th Doppler cell are respectively: where a T (f d , p) and a R (f d ) are the transmit and receive steering vectors of the range ambiguity region of interest in the d-th Doppler cell, respectively. f d is the Doppler frequency of the detected target in the d-th Doppler cell, p is the serial number of the range ambiguity region where the detected target is located, f dc is the Doppler frequency offset caused by the radial velocity of the detected target, d T is the spacing between adjacent transmit array element antennas, d R is the spacing between adjacent receive array element antennas, v is the radar moving speed, v a is the azimuth velocity of the detected target, Δf is the carrier frequency difference of the orthogonal signals transmitted by adjacent transmit array element antennas, and c is the speed of light; The synthetic steering vector of the detection target in the d-th Doppler cell is: Among them, represents the Kronecker product.

7. A clutter suppression method for SAR-MMTI based on frequency diverse array according to claim 1 or 2, characterized in that, When the sample x in the training sample set satisfies the following formula, it is excluded as a sample containing non-uniform clutter and interference: Among them, R c is the clutter covariance matrix estimated from the training sample set, η is the decision threshold and η = αNM, where α is a coefficient greater than 1, and (·) H denotes the conjugate transpose of the matrix.

8. A method for suppressing clutter in SAR-MMTI based on a frequency diverse array according to claim 7, characterized in that, The clutter-free covariance matrix R of the i-th range bin c (f d ) is expressed as: Among them, E[·] represents the expectation operation, and C(i, f d ) is the clutter matrix of f in the i-th range cell, d where f d is the Doppler frequency of the detected target in the d-th Doppler cell.

9. A clutter suppression method for SAR-MMTI based on frequency diverse array according to claim 8, characterized in that, Under the condition of the maximum output signal-to-noise ratio, the optimal adaptive weight vector \(W\) of the detection target in the \(d\)th Doppler cell opt (f d , p) is as follows: where d(f d , p) is the synthetic steering vector of the detection target in the d-th Doppler cell.

10. A clutter suppression method for SAR-MMTI based on frequency diverse array according to claim 2, characterized in that, Echo signal after clutter suppression is as follows: The echo signal after azimuth dimension compression is as follows: Among them, W opt (f d , p) is the optimal adaptive weight vector of the detection target in the d-th Doppler cell, is the range-Doppler domain echo signal after range migration correction, (·) H represents the conjugate transpose of the matrix, H(f d ) is the impulse response of the azimuth dimension matched filter, f d is the Doppler frequency of the detection target in the d-th Doppler cell, p is the serial number of the range ambiguity region where the detection target is located, t m is the azimuth dimension time, is the fast time of the range dimension.

Images