1. A two-step channel error estimation and compensation method for a distributed multi-channel SAR is characterized by comprising the following steps:

based on inertial navigation data, coarse correction of envelope walking is achieved; for the corrected data, the space-variant phase error between channels is roughly compensated based on inertial navigation data;

reordering the data after envelope walking coarse correction and inter-channel space-variant phase error coarse compensation according to the correlation between the antenna sampling data; and estimating residual envelope walking and compensating by adopting a distance frequency domain phase increment method, and finely estimating and compensating residual space-variant phase errors by adopting a distance time domain phase increment method.

2. The distributed multi-channel SAR-oriented two-step channel error estimation and compensation method according to claim 1, wherein the envelope walking includes distance space-variant envelope walking and inter-channel space-variant envelope walking.

3. The distributed multi-channel SAR-oriented two-step channel error estimation and compensation method according to claim 2, wherein the specific process of implementing coarse correction of envelope walking based on inertial navigation data is as follows:

solving the slope distance difference from the target at different distances in the scene to each channel and the reference channel by using inertial navigation information, and performing linear fitting on the slope distance difference; and based on the obtained fitting parameters, performing CS operation, correcting a distance space-variant part of envelope walking, multiplying a signal obtained after the CS operation by a linear phase in a distance frequency domain, correcting an inter-channel space-variant part of envelope walking, and completing coarse envelope walking correction based on inertial navigation data.

4. The distributed multi-channel SAR-oriented two-step channel error estimation and compensation method of claim 3, wherein the CS operation is: and setting a scaling equation and a frequency domain matching function based on the obtained fitting parameters, processing the echo signal, and multiplying the processed result by the residual phase to realize the envelope walking distance space-variant correction.

5. The two-step channel error estimation and compensation method for the distributed multi-channel SAR according to claim 4, wherein the scaling equation is as follows:

s_sc(t)＝exp{jπα_mK_r(t-t_ref)²}

wherein the content of the first and second substances,K_rrepresenting the chirp rate, t, of the transmitted signal_ref＝2r₀C denotes a reference time, c denotes a speed of light, t denotes a time, beta_mIs the slope in the fitting parameters.

6. The two-step channel error estimation and compensation method for the distributed multi-channel SAR according to claim 4, wherein the scaling equation is as follows:

wherein the content of the first and second substances,f_τdenotes the distance frequency, K_rRepresenting the chirp rate, beta, of the transmitted signal_mIs the slope in the fitting parameters.

Background

Synthetic Aperture Radar (SAR) is an active microwave remote sensing device firstly proposed in the united states in the 50 th century, has the characteristics of all-weather, two-dimensional, high resolution, strong tradition and the like, and plays an important role in the fields of disaster monitoring, mapping and the like. Along with the improvement of the stealth performance and the air performance requirements of the airborne platform, the size and the weight of various sensors installed on the platform are all subjected to more severe requirements. By adopting the distributed shaped-mounted multi-channel SAR system, the design of the whole system has higher flexibility, and the requirements of platform stealth and pneumatic performance are met.

The azimuth multi-channel SAR system can realize high-resolution wide-range imaging, but the aperture of the antenna is too large. The multi-channel distributed SAR system can divide a traditional multi-channel large-aperture antenna into a plurality of small-aperture antennas to be distributed and shaped and installed at each part of a machine body, and the influence of SAR equipment on the stealth performance and the air performance of a platform is reduced. However, distributed SAR channels are sparse and have a large vertical track baseline. The channel sparseness causes poor correlation among channels, the channel amplitude and phase error and the sampling time error cannot be estimated by applying the method to the traditional phase increment method, and the operation amount of the subspace method is large. The vertical track base line can generate the skew distance difference between channels, so that the data envelope moving and the phase error between the channels are caused, the azimuth Doppler reconstruction cannot be directly carried out, and the compensation is needed. And when the imaging width is larger, the large distance space variation exists in the slope distance difference, the traditional compensation method only considers the phase error distance space variation, ignores the distance space variation of envelope walking and influences the imaging quality. Further research into distributed multi-channel SAR channel error estimation and compensation methods is therefore needed.

Disclosure of Invention

In view of this, the invention provides a two-step channel error estimation and compensation method for a distributed multi-channel SAR, which effectively solves the problems of sparse channels of the distributed multi-channel SAR and difficult error estimation and compensation caused by a vertical track baseline, and is suitable for high-resolution wide-range imaging of the distributed multi-channel SAR.

The technical scheme for realizing the invention is as follows:

a two-step channel error estimation and compensation method for a distributed multi-channel SAR comprises the following steps:

based on inertial navigation data, coarse correction of envelope walking is achieved; for the corrected data, the space-variant phase error between channels is roughly compensated based on inertial navigation data;

reordering the data after envelope walking coarse correction and inter-channel space-variant phase error coarse compensation according to the correlation between the antenna sampling data; and estimating residual envelope walking and compensating by adopting a distance frequency domain phase increment method, and finely estimating and compensating residual space-variant phase errors by adopting a distance time domain phase increment method.

Preferably, the envelope walking of the present invention includes distance space-variant envelope walking and inter-channel space-variant envelope walking.

Preferably, the specific process of implementing coarse correction of envelope walking based on inertial navigation data of the present invention is as follows:

solving the slope distance difference from the target at different distances in the scene to each channel and the reference channel by using inertial navigation information, and performing linear fitting on the slope distance difference; and based on the obtained fitting parameters, performing CS operation, correcting a distance space-variant part of envelope walking, multiplying a signal obtained after the CS operation by a linear phase in a distance frequency domain, correcting an inter-channel space-variant part of envelope walking, and completing coarse envelope walking correction based on inertial navigation data.

Preferably, the CS of the present invention operates as: and setting a scaling equation and a frequency domain matching function based on the obtained fitting parameters, processing the echo signal, and multiplying the processed result by the residual phase to realize the envelope walking distance space-variant correction.

Preferably, the scaling equation of the present invention is:

s_sc(t)＝exp{jπα_mK_r(t-t_ref)²}

wherein the content of the first and second substances,K_rrepresenting the chirp rate, t, of the transmitted signal_ref＝2r₀C denotes a reference time, c denotes a speed of light, t denotes a time, beta_mIs the slope in the fitting parameters.

Preferably, the scaling equation of the present invention is:

wherein f is_τDenotes the distance frequency, K_rWhich represents the chirp rate of the transmitted signal,β_mis the slope in the fitting parameters.

Has the advantages that:

the invention provides a distributed multi-channel SAR-oriented two-step channel error estimation and compensation method, which is characterized in that data after envelope walking coarse correction and inter-channel space-variant phase error coarse compensation are reordered according to correlation among antenna sampling data, so that the correlation among adjacent data is good, and therefore, the error can be accurately estimated by adopting a traditional time domain correlation method, and the channel error estimation and compensation of the distributed multi-channel SAR space-variant is realized.

Secondly, the invention provides a two-step channel error estimation and compensation method for a distributed multi-channel SAR, which corrects the linearly-changed envelope walking through CS operation without blocking an image, and greatly reduces the complexity of operation compared with the traditional method for correcting the envelope walking through blocking.

Drawings

Fig. 1 is a flowchart of a two-step channel error estimation and compensation method for a distributed multi-channel SAR according to the present invention;

FIG. 2 is a schematic diagram of distributed SAR multi-channel data spatial sampling;

FIG. 3 is an azimuthal cross-sectional view of a point target simulation imaging result;

fig. 4 is a view of a scene simulation imaging result.

Detailed Description

Embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

It should be noted that, in the case of no conflict, the features in the following embodiments and examples may be combined with each other; moreover, all other embodiments that can be derived by one of ordinary skill in the art from the embodiments disclosed herein without making any creative effort fall within the scope of the present disclosure.

It is noted that various aspects of the embodiments are described below within the scope of the appended claims. It should be apparent that the aspects described herein may be embodied in a wide variety of forms and that any specific structure and/or function described herein is merely illustrative. Based on the disclosure, one skilled in the art should appreciate that one aspect described herein may be implemented independently of any other aspects and that two or more of these aspects may be combined in various ways. For example, an apparatus may be implemented and/or a method practiced using any number of the aspects set forth herein. Additionally, such an apparatus may be implemented and/or such a method may be practiced using other structure and/or functionality in addition to one or more of the aspects set forth herein.

As shown in fig. 1, an embodiment of the present application provides a two-step channel error estimation and compensation method for a distributed multi-channel SAR, including the following steps:

s1: and compensating the envelope walking of the space variation among the echo data channels based on the inertial navigation data.

When the baseline vertical component is large and the imaging width is wide, the distance space variability of envelope walking caused by the inter-channel vertical baseline is not negligible. Over a large range, the envelope walk may be approximated by a linear fit. The traditional method carries out envelope walking correction in blocks, and when the number of blocks is large, the operation is complex. The CS operation can correct the envelope walk of linear variation, so the CS operation is introduced in this step to perform one-time space-variant envelope walk correction on the whole wide-width scene. The process of the step is as follows: solving the slope distance difference from the target at different distances in the scene to each channel and the reference channel by using inertial navigation information, and performing linear fitting on the slope distance difference; based on the obtained fitting parameters, CS operation is carried out to correct the space-variant part of the envelope walking distance, the CS operation is used for setting a scaling equation and a frequency domain matching function, echo signals are processed, and then the processed result is multiplied by the residual phase to realize the space-variant correction of the envelope walking distance; and on a distance frequency domain, multiplying a signal obtained after the CS operation by a linear phase to finish the envelope walking coarse correction based on the inertial navigation data.

When the step is implemented specifically:

firstly, the flying height and flying attitude of the platform are obtained according to inertial navigation information, the coordinates of each channel of the radar are calculated, and the slope distance difference delta R from targets at different distances in a scene to each channel and a reference channel is calculated_m(r), where m denotes a channel number and r is a skew distance of the target to the reference channel.

For Δ R_m(r) performing a linear fit:

ΔR_m(r)≈ΔR_m,0+β_m(r-r₀)

wherein, Δ R_m,0And beta_mFor the parameters obtained by fitting, Δ R_m,0Is a constant term, β_mIs the slope, r₀Is the slant distance from the center of the scene to the reference channel.

Secondly, performing CS operation on the echo signal;

the specific process of the CS operation is as follows:

let the required scaling equation be:

s_sc(t)＝exp{jπα_mK_r(t-t_ref)²}

wherein the content of the first and second substances,K_rrepresenting the chirp rate, t, of the transmitted signal_ref＝2r₀C represents the reference time, c represents the speed of light, and t represents time.

Multiplying echo signals by a scaling equation, then carrying out Fourier transform (FFT) change, and then multiplying by a frequency domain matching function to realize matched filtering, wherein the frequency domain matching function is

Wherein f is_τRepresents a range frequency;

performing inverse Fourier transform on the frequency domain signal subjected to matched filtering to convert the frequency domain signal into a time domain signal, and multiplying the time domain signal by the residual phaseAnd completing the CS operation, and correcting the distance space-variant part of the envelope walking.

Thirdly, multiplying the signal obtained after the CS operation by a linear phase on a distance frequency domainCompleting envelope walking coarse correction based on inertial navigation data to obtain data s_m,1(t, η), η, represents slow time.

S2: calculated Delta R according to inertial navigation data_m(r) comparing the data s_m,1(t, η) times the upper phaseCoarse correction of space-variant phase errors among channels based on inertial navigation data is achieved, and data s are obtained_m,2(t, η), whereinTherefore, the phase can also be adjustedIs shown as

S3: and reordering the data after the envelope walking coarse correction and the inter-channel space-variant phase error coarse compensation according to the correlation between the antenna sampling data.

After envelope walking and coarse phase correction are carried out according to inertial navigation information, channel amplitude-phase errors, distance sampling time delay and residual slope distance differences caused by inertial navigation measurement errors are remained. A fine estimation and correction is required. However, the channel spacing is sparse, the correlation is poor, and the error estimation by the traditional time domain correlation method cannot be directly adoptedAnd (4) poor. A data rearrangement is required. FIG. 2 is a schematic diagram of distributed SAR multi-channel data spatial sampling, p_m,nSample data representing the nth PRT (pulse repetition time) of the mth antenna. Assuming that the radar has 3 antennas, the conventional time domain correlation method has a data sequence of [.. p._1,n-1,p_2,n-1,p_3,n-1,p_1,n,p_2,n,p_3,n,p_1,n+1,p_2,n+1,p_3,n+1,...]The correlation between adjacent data is poor, and p is assumed_1,nAnd p_2,n-1Has large data correlation between p_3,n-2And p_1,n+1If the data correlation between the two data is large, the data with large correlation are adjacently arranged, and the data after rearrangement in this embodiment is [_1,n,p_2,n-1,p_3,n-2,p_1,n+1,p_2,n,p_3,n-1,p_1,n+2,p_2,n+1,p_3,n,...](ii) a The data processed in step S2 is reordered according to the above-described order.

S4: and performing correlation operation on the reordered adjacent data, namely performing correlation between the closest points of the spatial sampling positions to obtain phase increment, calculating the Doppler center frequency according to the result of the correlation operation, and calculating the residual envelope quantity walk and compensate on the basis of the Doppler center frequency.

Based on the sorting example listed in step S3, the specific process implemented in this step is:

after rough correction of space-variant envelope walking and rough correction of space-variant phase errors, respectively carrying out distance Fourier transform on the multi-channel data to obtain distance frequency domain data;

wherein, theta_mSuperimposing phase, phi, for point objects in a scene_mFor channel inherent phase error, d_mIs the spacing from the reference channel, v is the platform velocity, θ₀For beam azimuth center squint angle, M is 1,2 … M-1.

The phase increment of the m-th and m + 1-th channels is

Let C_M(f_τEta) is the cross-correlation of channel 1 at time eta + PRT and channel M at time eta- (M-1) PRT, M is the total number of channels, then

Further, the doppler center frequency is obtained as:

the phase error of channel m with respect to the reference channel is then:

m＝2,3,...,M

by calculatingThe linear slope of the envelope can be used to obtain a fine estimate of the residual envelope walking amount

Data S_m,2(f_τEta) compensationThen, data S is obtained_m,3(f_τ,η)。

S5: for the data S_m,3(f_τEta) by distanceAnd (4) precisely estimating and compensating the residual space-variant phase error by using a time domain phase increment method.

The method specifically comprises the following steps: to S_m,3(f_τEta) inverse Fourier transform to obtain s_m,3(t, η), there is only the residual space-variant phase difference caused by the channel phase error and the inertial navigation measurement error. The phase error of the local area channel can be considered as a constant, the scene is divided into blocks, and each block respectively adopts a phase increment method to estimate the phase error in the block. Similar to the correlation method in S4, the phase increments of the m-th and m + 1-th channels are

Wherein, Δ R_m,e(τ) is the residual slope distance difference.

The remaining channel error is then:

because the slant range difference extends the distance to the space variation, the subblock phase difference can be estimated in a block mode, and the phase error of the space variation of the whole scene along the distance is estimated by adopting a weighted least square method for fittingFinal compensation

The compensated data can be subjected to imaging processing after realizing frequency spectrum reconstruction by adopting a traditional azimuth inverse filtering method.

The effect of the present invention is further explained by point target simulation and scene simulation test.

The simulation system is 4 channels with mixed baselines which are linearly and uniformly distributed, the parameters are shown in table 1, the slope distance of a point target from a reference channel is 140km, and the point target is positioned near the near end of a scene.

Platform velocity	1700m/s
		Channel spacing (x, y, z)	10m，4.8m，4m
Azimuth dimension of antenna	1.5m
		Transmission bandwidth	200MHz
Pulse repetition frequency	900Hz
		Height of platform	20km
Extent of swath	130km-200km

And evaluating the height of the azimuth direction fuzzy side lobe of the imaging result, and verifying the effectiveness of the distance space-variant envelope walking compensation method based on the CS principle. In the contrast experiment, when the compensation envelope moves, the CS operation is carried out by taking the scene center as a reference distance, the linear phase is directly multiplied in the distance frequency domain by taking the scene center slope distance as the reference distance, and the linear phase is directly multiplied in the distance frequency domain by taking the point target slope distance as the reference distance. At the target position after compensation, the residual envelope motion between adjacent channels is respectively: 0.0026m, 0.0577m, 0 m. The three experiments are consistent in processing flow except for the envelope walking correction, the processing is carried out according to the flow chart shown in fig. 1, the processing result is shown in fig. 3, and the corresponding fuzzy sidelobe height is-67 dB, -43dB and-73 dB, so that in the simulation experiment, the fuzzy sidelobe of the CS method considering the envelope walking distance space variation is improved by 24dB compared with the unified correction method taking the scene center as reference. And (4) accurately correcting the envelope walking result by taking the target position as a reference.

The scene simulation is a mapping zone edge scene, the distance sampling point is 2200, and the azimuth sampling point is 10000. The imaging result is shown in fig. 4, fig. 4a is the imaging result after estimation and compensation by the conventional method, and fig. 4b is the processing result by the method of the present embodiment. The boxes in fig. 4 identify the face objects of strong scattering coefficients in the scene. The area a is the blur component of the face target. In fig. 4a, the blur component is very noticeable. In fig. 4b, the blur component is hardly visible. The region B is a river region, and the scattering coefficient is weak. A blurring component exists that would normally be raised by some onshore strong scattering. The blur component of the lower image is significantly stronger than the upper one. It can be seen that the imaging result blurring inhibition capability of the method provided by the patent is obviously enhanced, and the imaging quality is good.

Therefore, the invention provides a two-step channel error estimation and compensation method for a distributed multi-channel SAR, which corrects channel amplitude and phase errors, sampling time delay and channel distance space-variant slant distance difference by adopting coarse compensation based on inertial navigation data and fine estimation and compensation based on data. And the CS method is adopted to realize the space-variant envelope walking compensation. The method adopts a phase increment method to estimate the residual error after data rearrangement, and makes up for the defects of the prior art.

The present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof, and it will be understood by those skilled in the art that various changes and modifications may be made herein without departing from the spirit and scope of the invention as defined in the appended claims.