CN104076343B - Satellite-borne three-channel SAR-GMTI self-adaptive clutter suppression method - Google Patents

Abstract

The invention discloses a satellite-borne three-channel SAR-GMTI self-adaptive clutter suppression method and relates to self-adaptive clutter suppression. The method comprises the first step of obtaining distance Doppler domain target echo signals after distance compression, the second step of obtaining distance Doppler domain echo signals after distance compression, the third step of obtaining data X1 of a first distance unit to be detected of three channels, the fourth step of obtaining data of multiple Doppler units subjected to self-adaptive clutter suppression in the first distance unit to be detected, and the fifth step of making the number l of the distance unit to be detected to be increased by one, repeating the second step, the third step and the fourth step until l is equal to L, namely completing clutter suppression of L distance units, and then outputting data Y of the L distance units subjected to clutter suppression. The satellite-borne three-channel SAR-GMTI self-adaptive clutter suppression method is used for suppressing ground clutter received by a three-channel SAR-GMTI system.

Description

Satellite-borne three-channel SAR-GMTI self-adaptive clutter suppression method

Technical Field

The invention belongs to the technical field of Radar, relates to self-adaptive clutter suppression, and particularly relates to a satellite-borne three-channel Synthetic Aperture Radar (SAR) -Ground Moving Target Indication (GMTI) self-adaptive clutter suppression method for suppressing Ground clutter received by a three-channel SAR-GMTI system.

Background

The proposal of the synthetic aperture concept and the invention of the SAR are a major breakthrough in the development history of the radar technology in the twentieth century. The SAR obtains high resolution in distance by emitting large bandwidth signals, and forms a large synthetic aperture by means of relative motion between a radar platform and a target, so that the high resolution in direction is obtained. The acquisition of high resolution of distance and direction enables the SAR to acquire large swath two-dimensional images similar to optical imaging all weather, all day long and far away, and the information acquisition capability of the radar is greatly improved. In view of the above advantages, SAR has gained widespread attention and application in recent years. The satellite-borne three-channel SAR-GMTI system becomes a research hotspot due to the important role of the system in traffic monitoring and battlefield reconnaissance.

For the space-borne SAR-GMTI system, because the radar works in a downward-looking state, a large amount of ground clutter is inevitably contained in an echo, and the frequency spectrum of the ground clutter may overlap with the frequency spectrum of a target, so that the target may be submerged by the strong clutter, and the detection of the space-borne SAR-GMTI system on the target is seriously influenced. In order to solve the above problems, an effective clutter suppression method is needed to suppress ground clutter and improve the detection performance of the system on the target.

The offset Phase center Antenna (DPCA) technique, which suppresses clutter by subtracting two images from each other, is one of the most commonly used two-channel SAR-GMTI techniques. However, DPCA techniques are designed for two-channel SAR-GMTI systems. When the number of channels of the SAR-GMTI system is more than two, the DPCA technology cannot fully accumulate the energy of the target signal. Furthermore, since DPCA techniques only use two degrees of freedom to suppress clutter, clutter suppression capability is very limited. Thus, DPCA techniques are not optimal for multi-channel SAR-GMTI systems. In order to solve the problems, Ender, Cerutti-Maori and the like of Germany high energy physics and radar technical research institute propose a multi-channel SAR-GMTI method based on adaptive clutter suppression, and the method carries out adaptive clutter suppression in a range-Doppler domain. The inventor finds that the method ignores the correlation of the data of the adjacent Doppler units, thereby limiting the clutter suppression performance of the multi-channel SAR-GMTI system and further limiting the detection capability of the system on the target.

Disclosure of Invention

Aiming at the defects of the multi-channel SAR-GMTI range-Doppler domain self-adaptive clutter suppression method, the invention provides a satellite-borne three-channel SAR-GMTI self-adaptive clutter suppression method, which performs self-adaptive clutter suppression on data of each Doppler unit in a range-Doppler domain by combining two adjacent Doppler unit data.

In order to achieve the purpose, the invention is realized by adopting the following technical scheme.

A satellite-borne three-channel SAR-GMTI self-adaptive clutter suppression method is characterized by comprising the following steps:

step 1, establishing a target original echo signal model, and obtaining a range frequency domain target echo signal according to the target original echo signal; constructing a distance frequency domain distance compression filter according to the distance frequency domain target echo signal; obtaining a range-Doppler domain target echo signal after range compression according to the range-frequency domain target echo signal and a range-frequency domain range compression filter;

step 2, the satellite-borne three-channel SAR-GMTI system receives three-channel original echo signals, distance Fourier transform is respectively carried out on the received three-channel original echo signals to obtain distance frequency domain echo signals, then distance compression is respectively carried out on the distance frequency domain echo signals according to a distance frequency domain distance compression filter to obtain distance compressed echo signals, and the distance compressed echo signals are converted into a distance Doppler domain to obtain distance compressed distance Doppler domain echo signals;

step 3, taking out the data x of the No. I distance unit to be detected from the distance Doppler domain echo signal after the distance compression of each channel_l，iIf i represents a channel serial number, i is 1, 2, 3, and L is 1_lExpressed as:

X_l＝[x_l，1，x_l，2，x_l，3]

wherein x is_l，1Data for channel 1, distance unit I, x_l，2Data for channel 2, distance unit I, x_l，3Data for channel 3, distance unit I, x_l，1、x_l，2And x_l，3The dimensions are K × 1, and K is the number of Doppler units needing target detection;

step 4, data X of No. I distance units to be detected in three channels_lIn the method, a space-time data vector z of data of three adjacent Doppler units is constructed_l，k(ii) a Then, according to the range-compressed range-Doppler domain target echo signals obtained in the step 1, constructing space-time guiding vectors D of targets of three adjacent Doppler units_l，k(ii) a Space-time steering vector D of target according to three adjacent Doppler units_l，kSolving for weight vector w_l，k(ii) a Using weight vectors w_l，kFor space-time data vector z_l，kPerforming self-adaptive clutter suppression to obtain self-adaptive clutter suppressed data y of the kth Doppler unit of the No. l distance unit to be detected_l，k(ii) a And then completing the data y after the self-adaptive clutter suppression of each Doppler unit of the No. I distance unit to be detected_l＝[y_l，1，y_l，2，…，y_l，K]^T；

And 5, increasing the number L of the distance units to be detected by 1, repeating the steps 2-4 until L is equal to L, finishing clutter suppression of the L distance units, and outputting data Y after the clutter suppression of the L distance units, wherein the Y is [ Y ═ Y-₁，y₂，…y_l…，y_L]。

The technical scheme has the characteristics and further improvement that:

(1) step 1 comprises the following substeps:

1a) the instantaneous distance of the target to the ith channel is expressed as:

wherein v is_xIs the target azimuth velocity, v_yIs the target range-wise velocity, y₀Is a slow time t_aWhen the ordinate is 0, i is 1, 2, 3.

The target original echo signal received by the ith channel is represented as:

wherein A is₀A complex constant reflecting the scattering power of moving objects, t_rFor fast time, c is the speed of light, w_a(t_a) Is an azimuthal envelope, w_r(t_r) Is the distance envelope, f_cIs a carrier frequency, K_rFor frequency modulation of the system transmission signal, t_aFor slow time, i is 1, 2, 3.

1b) Obtaining a range frequency domain target echo signal according to a formula (2), wherein the expression is as follows:

wherein f is_rIs the distance frequency, t_aIs a slow time, W_r(f_r) Is the distance frequency envelope, f_cIs a carrier frequency, K_rFor frequency modulation of the system transmission signal, A₀A complex constant reflecting the scattering power of the moving object, c is the speed of light, w_a(t_a) Is the azimuth envelope.

1c) According to the expression of the range frequency domain target echo signal, constructing a range frequency domain range compression filter as follows:

wherein f is_rIs a distance frequency, K_rThe system is frequency modulated for the transmitted signal.

1d) The distance compression filter in the distance frequency domain is utilized to carry out distance compression on the target echo signal in the distance frequency domain to obtain the target echo signal after the distance compression, and according to the formula (3) and the formula (2), the target echo signal expression after the distance compression is as follows:

wherein f is_rIs the distance frequency, t_aIs a slow time, W_r(f_r) Is the distance frequency envelope, f_cIs a carrier frequency, K_rFor frequency modulation of the system transmission signal, A₀A complex constant reflecting the scattering power of the moving object, c is the speed of light, w_a(t_a) Is the azimuth envelope.

1e) Performing range inverse Fourier transform and azimuth Fourier transform on the range-compressed target echo signal to obtain a range-compressed range-Doppler-domain target echo signal, wherein the range-compressed range-Doppler-domain target echo signal has an expression:

wherein, t_rFor fast time, B is the bandwidth of the transmitted signal, A₀A complex constant reflecting the scattering power of the moving object, c is the speed of light, f_aIs the Doppler frequency, W_a(f_a) Is the Doppler frequency envelope, λ is the signal wavelength, v_aIs radar platform velocity, v_xIs the target azimuth velocity, v_yIs the target range-wise velocity, y₀Is a slow time t_aD is the spacing of the equivalent phase centers of adjacent channels, on the ordinate of the target at 0.

(2) Step 4 comprises the following substeps:

4a) data X of No. I distance unit to be detected in three channels_lIn the method, data of k-1, k and k +1 Doppler units form a space-time data vector z_l，k：

Wherein,data representing the 1 st channel # l distance cell # k doppler cell,data representing the 2 nd channel # l distance cell # k doppler cell,data representing the 3 rd channel # l distance cell # k doppler cell,indicating the Doppler frequency of the kth Doppler cell, superscript^TRepresenting a non-conjugate transpose, the k-1, k, and k +1 doppler cells are three adjacent doppler cells.

4b) Constructing space-time steering vectors D of targets positioned at the k-1, k and k +1 Doppler units of the No. l range unit according to the expression (6) of range-Doppler domain target echo signals_l，k：

Wherein,a steering vector for a target of the kth doppler cell being the l-th range cell;a steering vector of a target of the kth-1 th Doppler unit which is the ith distance unit;a steering vector of a target of the (l) th distance element (k + 1) th Doppler element;

4c) solving for the weight vector w by solving the following equation (10)_l，kTo obtain

Wherein,is the covariance matrix of the kth doppler cell,D_l，kspace-time steering vectors of targets of the (l) th distance units (k-1, k and k + 1) th Doppler unit,data representing the 1 st channel # l distance cell # k doppler cell,data representing the 2 nd channel # l distance cell # k doppler cell,data representing the 3 rd channel # l range cell # k doppler cell.

4d) Using weight vectors w_l，kFor space-time data vector z_l，kAnd (3) carrying out self-adaptive clutter suppression:

wherein, y_l，kIs the data after the self-adaptive clutter suppression of the kth Doppler unit of the No. l distance unit to be detected, w_l，kAs weight vectors, superscript^HRepresenting a conjugate transpose.

4e) Increasing K by 1, and repeating the steps 4a) to 4d) until K is equal to K, wherein K is the number of Doppler units needing target detection, and data after adaptive clutter suppression of each Doppler unit of the No. I distance unit to be detected is obtained: y is_l＝[y_l，1，y_l，2，…，y_l，K]^T。

Compared with the prior art, the invention has the following advantages:

1) the invention considers the coherence of the data of adjacent Doppler channels, and adopts a method of combining the data of two adjacent Doppler units to carry out self-adaptive clutter suppression on the data of each Doppler unit, thereby better suppressing clutter;

2) the invention can completely accumulate the energy of the signals of all channels, can obviously improve the final output signal-to-noise ratio of the system, and is beneficial to improving the detection performance of the system to the target;

4) the invention adopts a self-adaptive method to carry out clutter suppression, can adapt to the internal motion of clutter and the mismatch of system channels, and has wider application range.

Drawings

The invention is further described with reference to the following figures and detailed description.

FIG. 1 is a flow chart of an implementation of the present invention;

FIG. 2 is a perspective view of a three-channel SAR-GMTI system on a planar vehicle with an oblique distance; wherein the abscissa represents the azimuth direction and the ordinate represents the direction of the slope;

FIG. 3 is a diagram showing simulation results of range-Doppler domain data before clutter suppression after range compression; wherein the abscissa represents the range gate and the ordinate represents the doppler frequency;

FIG. 4 is a graph of simulation results of range-Doppler domain data after clutter suppression using the techniques of the present invention; wherein the abscissa represents the range gate and the ordinate represents the doppler frequency;

FIG. 5 is a comparison graph of clutter suppression in the range-Doppler domain; wherein the abscissa represents the doppler frequency and the ordinate represents the clutter suppression ratio.

Detailed Description

Referring to fig. 1, the satellite-borne three-channel SAR-GMTI adaptive clutter suppression method of the present invention is described, which is used for suppressing the ground clutter received by a three-channel SAR-GMTI system, and comprises the following specific steps:

step 1, establishing a target original echo signal model, and obtaining a range frequency domain target echo signal according to the target original echo signal; constructing a distance frequency domain distance compression filter according to the distance frequency domain target echo signal; and obtaining a range-Doppler domain target echo signal after range compression according to the range-frequency domain target echo signal and the range-frequency domain range compression filter.

The observation geometry of the slant range planar spaceborne three-channel SAR-GMTI system is shown in figure 2: the distance between the equivalent phase centers of the channel 1 and the channel 2 is d, the distance between the equivalent phase centers of the channel 2 and the channel 3 is d, and the speed v of the radar platform is_a. In operation of the system, channel 2 transmits a signal. When receiving signals, the three channels receive simultaneously. At a slow time t_aWhen the phase center of channel 1 is 0, the phase center of channel 2 is 0, the phase center of channel 3 is 0, the phase center of channel 2 is 2, the phase center of channel 3 is 0, the phase center of channel 3 is 2, and the phase center of the object is 0, y₀). The target moves at a constant speed, and the speed along the azimuth direction is v_xVelocity in the direction of distance is v_y。

For the convenience of the following derivation, a model of the target raw echo signal is given below.

1a) The instantaneous distance of the target to the ith channel is expressed as:

wherein v is_xIs the target azimuth velocity, v_yIs the target range-wise velocity, y₀Is a slow time t_aOn the ordinate of the object, i denotes the channel number, i is 1, 2, 3.

The target original echo signal received by the ith channel is represented as:

wherein A is₀A complex constant reflecting the scattering power of moving objects, t_rFor fast time, c is the speed of light, w_a(t_a) Is an azimuthal envelope, w_r(t_r) Is the distance envelope, f_cIs a carrier frequency, K_rFor frequency modulation of the system transmission signal, t_aFor slow time, i is 1, 2, 3.

1b) Obtaining a range frequency domain target echo signal according to a formula (2), wherein the expression is as follows:

wherein f is_rIs the distance frequency, t_aIs a slow time, W_r(f_r) Is the distance frequency envelope, f_cIs a carrier frequency, K_rFor frequency modulation of the system transmission signal, A₀A complex constant reflecting the scattering power of the moving object, c is the speed of light, w_a(t_a) Is the azimuth envelope.

In order to improve the efficiency, the invention respectively carries out distance compression on three-channel original echo signals received by the system in a distance frequency domain in a phase multiplication mode.

1c) According to the expression of the range frequency domain target echo signal, constructing a range frequency domain range compression filter as follows:

wherein f is_rIs a distance frequency, K_rThe system is frequency modulated for the transmitted signal.

1d) The distance compression filter in the distance frequency domain is utilized to carry out distance compression on the target echo signal in the distance frequency domain to obtain the target echo signal after the distance compression, and according to the formula (3) and the formula (2), the target echo signal expression after the distance compression is as follows:

wherein,_fr is the distance frequency, t_aIs a slow time, W_r(f_r) Is the distance frequency envelope, f_cIs a carrier frequency, K_rFor frequency modulation of the system transmission signal, A₀A complex constant reflecting the scattering power of the moving object, c is the speed of light, w_a(t_a) Is the azimuth envelope.

1e) Performing range inverse Fourier transform and azimuth Fourier transform on the range-compressed target echo signal to obtain a range-compressed range-Doppler-domain target echo signal, wherein the range-compressed range-Doppler-domain target echo signal has an expression:

wherein, t_rFor fast time, B is the bandwidth of the transmitted signal, A₀A complex constant reflecting the scattering power of the moving object, c is the speed of light, f_aIs the Doppler frequency, W_a(f_a) Is the Doppler frequency envelope, λ is the signal wavelength, v_aIs radar platform velocity, v_xIs the target azimuth velocity, v_yIs the target range-wise velocity, y₀Is a slow time t_aD is the spacing of the equivalent phase centers of adjacent channels, on the ordinate of the target at 0.

And 2, the satellite-borne three-channel SAR-GMTI system receives three-channel original echo signals, distance Fourier transform is respectively carried out on the received three-channel original echo signals to obtain distance frequency domain echo signals, distance compression is respectively carried out on the distance frequency domain echo signals according to a distance frequency domain distance compression filter to obtain distance compressed echo signals, and the distance compressed echo signals are converted into a distance Doppler domain to obtain distance compressed distance Doppler domain echo signals.

Step 3, taking out the data x of the No. I distance unit to be detected from the distance Doppler domain echo signal after the distance compression of each channel_l，iIf i represents a channel serial number, i is 1, 2, 3, and L is 1_lExpressed as:

X_l＝[x_l，1，x_l，2，x_l，3]

wherein x is_l，1Data for channel 1, distance unit I, x_l，2Data for channel 2, distance unit I, x_l，3Data for channel 3, distance unit I, x_l，1、x_l，2And x_l，3The dimensions are all K × 1, K is the number of Doppler units needed to perform target detection.

Step 4, data X of No. I distance units to be detected in three channels_lIn the method, a space-time data vector z of data of three adjacent Doppler units is constructed_l，k(ii) a Then, according to the range-compressed range-Doppler domain target echo signals obtained in the step 1, constructing space-time guiding vectors D of targets of three adjacent Doppler units_l，k(ii) a Space-time steering vector D of target according to three adjacent Doppler units_l，kSolving for weight vector w_l，k(ii) a Using weight vectors w_l，kFor space-time data vector z_l，kPerforming self-adaptive clutter suppression to obtain self-adaptive clutter suppressed data y of the kth Doppler unit of the No. l distance unit to be detected_l，k(ii) a And then completing the data y after the self-adaptive clutter suppression of each Doppler unit of the No. I distance unit to be detected_l＝[y_l，1，y_l，2，…，y_l，K]^T。

4a) Data X of No. I distance unit to be detected in three channels_lIn the method, the data of the k-1, k and k +1 Doppler units are selected to form space timeData vector z_l，k：

Wherein,data representing the 1 st channel # l distance cell # k doppler cell,data representing the 2 nd channel # l distance cell # k doppler cell,data representing the 3 rd channel # l distance cell # k doppler cell,indicating the Doppler frequency of the kth Doppler cell, superscript^TRepresenting a non-conjugate transpose, the k-1, k, and k +1 doppler cells are three adjacent doppler cells.

4b) Constructing space-time steering vectors D of targets positioned at the k-1, k and k +1 Doppler units of the No. l range unit according to the expression (6) of range-Doppler domain target echo signals_l，k：

Wherein,a steering vector for a target of the kth doppler cell being the l-th range cell;for the k-1 Doppler unit of the No. l distance unitA steering vector of a target of the element;a steering vector of a target of the (l) th distance element (k + 1) th Doppler element;the expression is as follows:

wherein,denotes the Doppler frequency of the kth Doppler cell, λ is the signal wavelength, v_aIs radar platform velocity, v_xIs the target azimuth velocity, v_yD is the spacing of the equivalent phase centers of adjacent channels for the target range velocity.

4c) Solving for the weight vector w by solving the following equation (10)_l，kTo obtain

Wherein,is the covariance matrix of the kth doppler cell,D_l，kspace-time steering vectors of targets of the (l) th distance units (k-1, k and k + 1) th Doppler unit,data representing the 1 st channel # l distance cell # k doppler cell,data representing the 2 nd channel # l distance cell # k doppler cell,data representing the 3 rd channel # l range cell # k doppler cell.

In sub-step 4c) to optimize the spur suppression performance, i.e. to maximize the output signal-to-noise ratio, the weight vector w_l，kThe constraint condition in the above equation (10) is satisfied.

4d) Using weight vectors w_l，kFor space-time data vector z_l，kAnd (3) carrying out self-adaptive clutter suppression:

wherein, y_l，kIs the data after the self-adaptive clutter suppression of the kth Doppler unit of the No. l distance unit to be detected, w_l，kAs weight vectors, superscript^HRepresenting a conjugate transpose.

4e) Increasing K by 1, and repeating the steps 4a) to 4d) until K is equal to K, wherein K is the number of Doppler units needing target detection, and data after adaptive clutter suppression of each Doppler unit of the No. I distance unit to be detected is obtained: y is_l＝[y_l，1，y_l，2，…，y_l，K]。

And 5, increasing the number L of the distance units to be detected by 1, repeating the steps 2-4 until L is equal to L, finishing clutter suppression of the L distance units, and outputting data Y after the clutter suppression of the L distance units, wherein the Y is [ Y ═ Y-₁，y₂，…y_l…，y_L]。

The effect of the present invention will be further explained with the simulation experiment.

Simulation 1, data simulation before clutter suppression.

The SAR system simulation parameters are shown in a table 1, when the radar works in a front side view mode, a moving target exists in an observation scene, the azimuth speed of the moving target is zero, and the distance speed of the moving target is 10 m/s. The simulation result is shown in fig. 3, which shows the range-doppler domain data after range compression before clutter suppression, i.e. the target echo signal after range compression; the colors of the pixel elements of the graph represent the amplitude of the data, and it can be seen from the simulation results that the target is completely swamped, and if clutter suppression is not performed, the target cannot be detected. The range gates in the simulation of the present invention are range cells.

TABLE 1

Carrier frequency	5.4GHz	Speed of radar	7500m/s
				Distance bandwidth	50MHz	Center distance of scene	924km
Distance sampling frequency	75MHz	Bandwidth of azimuth	2000Hz
				Pulse repetition frequency	3000Hz	Pulse width	20μs
Signal to noise ratio	15dB	Noise to noise ratio	15dB
				Number of channels	3	Base length	2.5m

And 2, performing data simulation after clutter suppression by using the technology.

The parameter setting in the simulation is the same as that in the simulation 1, the simulation result is shown in figure 4, and the figure shows the range-doppler domain data after the range compression after the clutter suppression, namely the data after the clutter suppression of all range units is completed; the color of the pixel unit of the graph represents the amplitude of data, and as can be seen from simulation results, the target is clearly visible after clutter suppression, and the target can be easily detected, which shows that the method can well suppress clutter.

And (3) simulating, and comparing clutter suppression effects in the range-Doppler domain.

The parameter settings in this simulation are the same as those in simulation 1, and the simulation results are shown in fig. 5, where the solid line indicates the clutter suppression ratio when DPCA technique is used, the ' line indicates the clutter suppression ratio when conventional adaptive clutter suppression technique is used, and the ' + ' line indicates the clutter suppression ratio when the present invention is used. From simulation results, it can be seen that in the whole doppler bandwidth, the clutter suppression performance of the conventional adaptive clutter suppression technology is better than that of the DPCA technology, and the clutter suppression performance of the present invention is better than that of the DPCA technology and the conventional adaptive clutter suppression technology.

Claims (3)

1. A satellite-borne three-channel SAR-GMTI self-adaptive clutter suppression method is characterized by comprising the following steps:

step 1, establishing a target original echo signal model, and obtaining a range frequency domain target echo signal according to the target original echo signal; constructing a distance frequency domain distance compression filter according to the distance frequency domain target echo signal; obtaining a range-Doppler domain target echo signal after range compression according to the range-frequency domain target echo signal and a range-frequency domain range compression filter;

step 2, the satellite-borne three-channel SAR-GMTI system receives three-channel original echo signals, distance Fourier transform is respectively carried out on the received three-channel original echo signals to obtain distance frequency domain echo signals, then distance compression is respectively carried out on the distance frequency domain echo signals according to a distance frequency domain distance compression filter to obtain distance compressed echo signals, and the distance compressed echo signals are converted into a distance Doppler domain to obtain distance compressed distance Doppler domain echo signals;

step 3, taking out the data x of the No. I distance unit to be detected from the distance Doppler domain echo signal after the distance compression of each channel_l，iIf i represents a channel serial number, i is 1, 2, 3, and L is 1_lExpressed as:

X_l＝[x_l，1，x_l，2，x_l，3]

wherein x is_l，1Data for channel 1, distance unit I, x_l，2Data for channel 2, distance unit I, x_l，3Data for channel 3, distance unit I, x_l，1、x_l，2And x_l，3The dimensions are K × 1, and K is the number of Doppler units needing target detection;

step 4, data X of No. I distance units to be detected in three channels_lIn the method, a space-time data vector z of data of three adjacent Doppler units is constructed_l，k(ii) a Then, according to the range-compressed range-Doppler domain target echo signals obtained in the step 1, constructing space-time guiding vectors D of targets of three adjacent Doppler units_l，k(ii) a Space-time steering vector D of target according to three adjacent Doppler units_l，kSolving for weight vector w_l，k(ii) a Using weight vectors w_l，kFor space-time data vector z_l，kPerforming self-adaptive clutter suppression to obtain self-adaptive clutter suppressed data y of the kth Doppler unit of the No. l distance unit to be detected_l，k(ii) a And then completing the data y after the self-adaptive clutter suppression of each Doppler unit of the No. I distance unit to be detected_l＝[y_l，1，y_l，2，…，y_l，K]^T；

And 5, increasing the number L of the distance units to be detected by 1, repeating the steps 2-4 until L is equal to L, finishing clutter suppression of the L distance units, and outputting data Y after the clutter suppression of the L distance units, wherein the Y is [ Y ═ Y-₁，y₂，…y_l…，y_L]。

2. The on-board three-channel SAR-GMTI adaptive clutter suppression method according to claim 1, characterized in that step 1 comprises the following sub-steps:

1a) the instantaneous distance of the target to the ith channel is expressed as:

$R_i\left(t_a\right)=\sqrt{{(y_0+v_y{t}_a)}^2+{\[v_x{t}_a-v_a{t}_a+(i-1)d\]}^2}---\left(1\right)$

wherein v is_xIs the target azimuth velocity, v_yIs the target range-wise velocity, y₀Is a slow time t_aWhen the value is 0, i represents the channel serial number, and i is 1, 2 and 3;

the target original echo signal received by the ith channel is represented as:

$s_i(t_r,t_a)=A_0{w}_a\left(t_a\right)w_r(t_r-\frac{2R_i\left(t_a\right)}{c})\exp \{-j4{\mathrm{\πf}}_c\frac{R_i\left(t_a\right)}{c}+{\mathrm{j\πK}}_r{(t_r-\frac{2R_i\left(t_a\right)}{c})}^2\}---\left(2\right)$

wherein A is₀A complex constant reflecting the scattering power of moving objects, t_rFor fast time, c is the speed of light, w_a(t_a) Is an azimuthal envelope, w_r(t_r) Is the distance envelope, f_cIs a carrier frequency, K_rFor frequency modulation of the system transmission signal, t_aSlow time, i ═ 1, 2, 3;

1b) obtaining a range frequency domain target echo signal according to a formula (2), wherein the expression is as follows:

$s_i(f_r,t_a)=A_0{w}_a\left(t_a\right)W_r\left(f_r\right)\exp \{-j\frac{4\mathrm{\π}(f_r+f_c)}{c}{R}_i\left(t_a\right)\}\exp \{-j\frac{{\mathrm{\πf}}_r^2}{K_r}\}---\left(3\right)$

wherein f is_rIs the distance frequency, t_aIs a slow time, W_r(f_r) Is the distance frequency envelope, f_cIs a carrier frequency, K_rFor frequency modulation of the system transmission signal, A₀A complex constant reflecting the scattering power of the moving object, c is the speed of light, w_a(t_a) Is an orientation envelope;

1c) according to the expression of the range frequency domain target echo signal, constructing a range frequency domain range compression filter as follows:

$H_r\left(f_r\right)=\mathrm{\α}p\left\{j\frac{{\mathrm{\πf}}_r^2}{K_r}\right\}---\left(4\right)$

wherein f is_rIs a distance frequency, K_rFrequency modulation of the system transmission signal;

1d) the distance compression filter in the distance frequency domain is utilized to carry out distance compression on the target echo signal in the distance frequency domain to obtain the target echo signal after the distance compression, and according to the formula (3) and the formula (2), the target echo signal expression after the distance compression is as follows:

$s_{i,rc}(f_r,t_a)=s_i(f_r,t_a)H_r\left(f_r\right)=A_0{w}_a\left(t_a\right)W_r\left(f_r\right)\exp \{-j\frac{4\mathrm{\π}(f_r+f_c)}{c}{R}_i\left(t_a\right)\}---\left(5\right)$

wherein f is_rIs the distance frequency, t_aIs a slow time, W_r(f_r) Is the distance frequency envelope, f_cIs a carrier frequency, K_rFor frequency modulation of the system transmission signal, A₀A complex constant reflecting the scattering power of the moving object, c is the speed of light, w_a(t_a) Is an orientation envelope;

1e) performing range inverse Fourier transform and azimuth Fourier transform on the range-compressed target echo signal to obtain a range-compressed range-Doppler-domain target echo signal, wherein the expression of the range-compressed range-Doppler-domain target echo signal is as follows:

$\begin{array}{c}{s}_{i,rc}(t_r,f_a)=A_0\sin c\{B\[t_r-\frac{y_0}{c}(2+\frac{{\mathrm{\λ}}^2{f}_a^2-4v_y^2}{4{(v_a-v_x)}^2})\]\}{W}_a(f_a+\frac{2v_y}{\mathrm{\λ}})\exp \left\{j\frac{{\mathrm{\π\λy}}_0}{2{(v_a-v_x)}^2}{(f_a+\frac{2v_y}{\mathrm{\λ}})}^2\right\}\\ \×\exp \left\{-j\frac{2\mathrm{\π}d(i-1)}{v_a-v_x}\left(f_a+\frac{2v_y}{\mathrm{\λ}}\right)\right\}\exp \{-j\frac{4\mathrm{\π}}{\mathrm{\λ}}{y}_0\}\end{array}---\left(6\right)$

wherein, t_rFor fast time, B is the bandwidth of the transmitted signal, A₀A complex constant reflecting the scattering power of the moving object, c is the speed of light, f_aIs the Doppler frequency, W_a(f_a) Is the Doppler frequency envelope, λ is the signal wavelength, v_aIs radar platform velocity, v_xIs the target azimuth velocity, v_yIs the target range-wise velocity, y₀Is a slow time t_aD is the spacing of the equivalent phase centers of adjacent channels, on the ordinate of the target at 0.

3. The on-board three-channel SAR-GMTI adaptive clutter suppression method according to claim 2, characterized in that step 4 comprises the following sub-steps:

4a) data X of No. I distance unit to be detected in three channels_lIn the method, the data of the k-1, k and k +1 Doppler units form a space-time data vector z_l，k：

$z_{l,k}={\[x_{l,1}\left(f_{a_{k-1}}\right),x_{l,2}\left(f_{a_{k-1}}\right),x_{l,3}\left(f_{a_{k-1}}\right),x_{l,1}\left(f_{a_k}\right),x_{l,2}\left(f_{a_k}\right),x_{l,3}\left(f_{a_k}\right),x_{l,1}\left(f_{a_{k+1}}\right),x_{l,2}\left(f_{a_{k+1}}\right),x_{l,3}\left(f_{a_{k+1}}\right)\]}^T---\left(7\right)$

Wherein,data representing the 1 st channel # l distance cell # k doppler cell,data representing the 2 nd channel # l distance cell # k doppler cell,data representing the 3 rd channel # l distance cell # k doppler cell,indicating the Doppler frequency of the kth Doppler cell, superscript^TRepresenting a non-conjugate transpose, and k-1, k and k +1 Doppler units are three adjacent Doppler units;

4b) constructing space-time steering vectors D of targets positioned at the k-1, k and k +1 Doppler units of the No. l range unit according to the expression (6) of range-Doppler domain target echo signals_l，k：

$D_{l,k}={\[s_l\left(f_{a_{k-1}}\right),s_l\left(f_{a_k}\right),s_l\left(f_{a_{k+1}}\right)\]}^T---\left(8\right)$

Wherein,a steering vector for a target of the kth doppler cell being the l-th range cell;a steering vector of a target of the kth-1 th Doppler unit which is the ith distance unit;a steering vector of a target of the (l) th distance element (k + 1) th Doppler element;

4c) solving for the weight vector w by solving the following equation (10)_l，kTo obtain

Wherein,is the covariance matrix of the kth doppler cell,D_l，kspace-time steering vectors of targets of the (l) th distance units (k-1, k and k + 1) th Doppler unit,denotes the 1 st channelData for the kth doppler cell from cell # l,data representing the 2 nd channel # l distance cell # k doppler cell,data representing the 3 rd channel # l distance cell # k doppler cell;

4d) using weight vectors w_l，kFor space-time data vector z_l，kAnd (3) carrying out self-adaptive clutter suppression:

$y_{l,k}=w_{l,k}^H{z}_{l,k}---\left(11\right)$

wherein, y_l，kIs the data after the self-adaptive clutter suppression of the kth Doppler unit of the No. l distance unit to be detected, w_l，kAs weight vectors, superscript^HRepresents a conjugate transpose;

4e) increasing K by 1, and repeating the steps 4a) to 4d) until K is equal to K, wherein K is the number of Doppler units needing target detection, and data after adaptive clutter suppression of each Doppler unit of the No. I distance unit to be detected is obtained: y is_l＝[y_l，1，y_l，2，…，y_l，K]^T。