阅读说明：本技术 一种多通道sar复图像域相位和基线误差联合估计方法 (Multi-channel SAR complex image domain phase and baseline error joint estimation method ) 是由 向吉祥 孙光才 张瑜 邢孟道 于 2021-07-20 设计创作，主要内容包括：本发明公开了一种多通道SAR复图像域相位和基线误差联合估计方法,包括：对多通道回波数据预处理；进行基带多普勒频率线性相位校正；将各通道的频率线性相位校正数据分别进行SAR成像处理得到多通道图像集；利用图像子空间方法进行通道相位误差估计并对校正各通道图像相位；对相位校正后的多通道图像集进行通道基线误差估计,利用估计得到的通道基线误差沿方位基线值的更新；利用更新得到的沿方位基线值,重复对多通道图像集利用图像子空间方法进行通道相位误差估计至进行沿方位基线值的更新的步骤至满足迭代终止条件；利用迭代终止时的沿方位基线值和通道相位误差得到HRWS SAR图像。本发明能减小图像域误差估计计算量,提高成像质量。(The invention discloses a method for jointly estimating the phase and baseline error of a multi-channel SAR complex image domain, which comprises the following steps: preprocessing multi-channel echo data; performing baseband Doppler frequency linear phase correction; respectively carrying out SAR imaging processing on the frequency linear phase correction data of each channel to obtain a multi-channel image set; estimating channel phase errors by using an image subspace method and correcting the image phase of each channel; channel baseline error estimation is carried out on the multi-channel image set after phase correction, and updating of the channel baseline error obtained through estimation along the azimuth baseline value is utilized; repeating the steps from channel phase error estimation to updating of the value along the orientation baseline of the multi-channel image set by using an image subspace method until the iteration termination condition is met by using the value along the orientation baseline obtained by updating; and obtaining the HRWS SAR image by using the value along the azimuth baseline and the channel phase error when the iteration is terminated. The invention can reduce the error estimation calculated amount of the image domain and improve the imaging quality.)

1. A multi-channel SAR complex image domain phase and baseline error joint estimation method is characterized by comprising the following steps:

respectively preprocessing the acquired multi-channel echo data of the MACs SAR system;

respectively carrying out baseband Doppler frequency linear phase correction on the preprocessed multi-channel echo data to obtain frequency linear phase correction data of each channel;

respectively carrying out SAR imaging processing on the frequency linear phase correction data of each channel, and integrating the obtained images of each channel to obtain a multi-channel image set;

estimating the phase error of the channel by using an image subspace method for the multi-channel image set, and performing phase correction on each channel image according to the estimation result of the phase error of the channel;

channel baseline error estimation is carried out on the multi-channel image set after phase correction, and updating along the azimuth baseline value is carried out by utilizing the channel baseline error obtained by estimation;

repeatedly performing channel phase error estimation on the multi-channel image set by using an image subspace method by using the updated value along the orientation baseline until the step of updating the value along the orientation baseline is performed until an iteration termination condition is met;

and obtaining the HRWS SAR image by using the value along the azimuth baseline and the channel phase error when the iteration is terminated.

2. The method for jointly estimating the phase and the baseline error of the multi-channel SAR complex image domain according to claim 1, wherein the preprocessing the acquired multi-channel echo data of the MACs SAR system respectively comprises:

respectively carrying out channel amplitude error correction on the multi-channel echo data;

and respectively carrying out Doppler center correction on the multi-channel echo data subjected to the channel amplitude error correction.

3. The method for jointly estimating the phase and the baseline error of the multi-channel SAR complex image domain according to claim 2, wherein the respectively performing the channel amplitude error correction on the multi-channel echo data comprises:

and adjusting the average amplitude value of the multi-channel echo data to be consistent.

4. The method for jointly estimating the phase and the baseline error of the multi-channel SAR complex image domain according to claim 2 or 3, wherein the step of respectively performing Doppler center correction on the multi-channel echo data after the channel amplitude error correction comprises the following steps:

and moving the frequency spectrum center of each channel echo data after channel amplitude error correction to a zero frequency position.

5. The multi-channel SAR complex image domain phase and baseline error joint estimation method of claim 1, wherein the performing baseband Doppler frequency linear phase correction on the preprocessed multi-channel echo data respectively comprises:

and respectively transforming the preprocessed echo data of each channel to an azimuth frequency domain, and then multiplying the data by a corresponding compensation item.

6. The method for jointly estimating the phase and the baseline error of the multi-channel SAR complex image domain according to claim 5, wherein the form of construction of the compensation term comprises:

H_m(f_a)＝exp(-j2πx_mf_a/v)

wherein H_m(f_a) Represents the compensation term, x, corresponding to the mth channel in the azimuth frequency domain_mDenotes the relative position of the mth channel and the reference channel, f_aIndicating azimuth frequency and v indicating radar velocity.

7. The method for jointly estimating the phase and the baseline error of the multi-channel SAR complex image domain according to claim 1, wherein the estimating the phase error of the channel by using the image subspace method for the multi-channel image set comprises:

obtaining a first covariance matrix corresponding to the multi-channel image set;

decomposing the eigenvalue of the first covariance matrix to obtain a first noise subspace corresponding to the multi-channel image set;

based on the first noise subspace, solving an optimal solution of a first function which is constructed in advance to obtain a channel phase error estimation value matrix; wherein the first function is an optimal objective function of a channel phase error estimation problem.

8. The method of claim 7, wherein the expression of the first function comprises:

wherein Γ represents a diagonal matrix composed of phase error terms between channels; a is_iIndicating a steering vector corresponding to the azimuth ambiguity with the sequence number i; argmin {. } represents a variable value at which an objective function in { } takes a minimum value; i represents the change serial number of the orientation fuzzy quantity, and the value range of I is [ -I, I]I is larger than 0, and (2I +1) azimuth ambiguities are counted; { }^HRepresenting a conjugate transpose operation on a matrix; u shape_nRepresenting a first noise subspace.

9. The method for jointly estimating the phase and baseline error of the multi-channel SAR complex image domain according to claim 1 or 7, wherein the estimating the channel baseline error of the phase-corrected multi-channel image set comprises:

obtaining a second covariance matrix corresponding to the multi-channel image set after phase correction;

decomposing the eigenvalue of the second covariance matrix to obtain a second noise subspace corresponding to the multichannel image set after the phase correction;

based on the second noise subspace, solving the optimal solution of a second function which is constructed in advance by using a least square method to obtain a channel baseline error estimation value matrix; wherein the second function is an optimal objective function of the measurement error estimation problem along the azimuth baseline.

10. The method of jointly estimating the phase and the baseline error of the multi-channel SAR complex image domain according to claim 9, wherein the expression of the second function includes:

wherein, Δ X represents a channel baseline error estimation value matrix; a is_iIndicating a steering vector corresponding to the azimuth ambiguity with the sequence number i; argmin {. } represents a variable value at which an objective function in { } takes a minimum value; i represents the change serial number of the orientation fuzzy quantity, and the value range of I is [ -I, I]I is larger than 0, and (2I +1) azimuth ambiguities are counted; { }^HRepresenting conjugate transposing of matricesOperating; u denotes the second noise subspace.

Technical Field

The invention belongs to the field of radars, and particularly relates to a method for jointly estimating a domain phase and a baseline error of a multi-channel SAR complex image.

Background

Synthetic Aperture Radar (SAR) is an active earth observation system, can be installed on flight platforms such as airplanes, satellites and spacecrafts, and has the characteristics of all-time, all-weather and high-resolution earth observation. Therefore, the SAR system has unique application advantages in the fields of ocean monitoring, agricultural census and the like.

With the widespread application of SAR systems in the civilian and military fields, High Resolution Wide Swath (HRWS) imaging has become a growing concern worldwide. For conventional single-channel SAR systems, to achieve HRWS imaging and avoid range ambiguity, a low PRF (pulse repetition frequency), i.e., a low azimuth sampling rate, is typically employed. However, a low PRF will result in a low doppler bandwidth, which will mean that the azimuth size of the radar antenna cannot be set large enough to guarantee a high azimuth resolution, i.e. a high azimuth solution and a wide width (wide swath) are contradictory in conventional single-channel SAR systems.

To solve this conflict, the relevant scholars propose an azimuth multi-channel sar (macs sar) system in combination with digital beamforming techniques. The MACs SAR system transmits chirp signals in a single channel with a low PRF, i.e., the PRF is lower than the doppler bandwidth of the signals of each receiving channel of the system, and all antenna channels receive echoes simultaneously. Due to the low PRF, the echo of each receive channel is blurred in the doppler domain. In order to fully utilize the information carried by the azimuth angle signal, the echoes of all the receiving channels are combined by reconstructing the Doppler ambiguity-free signal. Essentially, this MACs SAR system operates in a low PRF mode and uses multiple azimuth receive channels to equivalently increase the azimuth sampling rate. Numerous SAR systems have demonstrated the feasibility and effectiveness of this model, such as the new multi-frequency polarized airborne SAR of DLR, the second commercial geostationary SAR satellite in canada, the high-resolution three-number geostationary SAR satellite in china.

However, due to the existence of non-ideal factors such as space temperature variation, certain amplitude and phase errors usually exist among channels of the MACs SAR system, and channel mismatch is caused. Mismatching among the channels can seriously affect the performance of image reconstruction, and cause the phenomenon of orientation blurring in an imaging result, thereby seriously reducing the imaging quality and causing poor HRWS imaging effect. Therefore, estimation and correction of the mismatch between channels in MACs SAR systems has become a critical issue in practical operation. For the phase error which is difficult to process, the existing method mainly carries out estimation processing in a time domain, and carries out iterative estimation on the phase error between channels by using a minimum entropy method for imaging.

Disclosure of Invention

In order to solve the problems in the prior art, the invention provides a multi-channel SAR complex image domain phase and baseline error joint estimation method. The technical problem to be solved by the invention is realized by the following technical scheme:

respectively preprocessing the acquired multi-channel echo data of the MACs SAR system;

respectively carrying out baseband Doppler frequency linear phase correction on the preprocessed multi-channel echo data to obtain frequency linear phase correction data of each channel;

respectively carrying out SAR imaging processing on the frequency linear phase correction data of each channel, and integrating the obtained images of each channel to obtain a multi-channel image set;

estimating the phase error of the channel by using an image subspace method for the multi-channel image set, and performing phase correction on each channel image according to the estimation result of the phase error of the channel;

channel baseline error estimation is carried out on the multi-channel image set after phase correction, and updating along the azimuth baseline value is carried out by utilizing the channel baseline error obtained by estimation;

repeatedly performing channel phase error estimation on the multi-channel image set by using an image subspace method by using the updated value along the orientation baseline until the step of updating the value along the orientation baseline is performed until an iteration termination condition is met;

and obtaining the HRWS SAR image by using the value along the azimuth baseline and the channel phase error when the iteration is terminated.

In an optional implementation manner, the preprocessing the acquired multi-channel echo data of the MACs SAR system includes:

respectively carrying out channel amplitude error correction on the multi-channel echo data;

and respectively carrying out Doppler center correction on the multi-channel echo data subjected to the channel amplitude error correction.

In an optional embodiment, the performing channel amplitude error correction on the multi-channel echo data respectively includes:

and adjusting the average amplitude value of the multi-channel echo data to be consistent.

In an optional embodiment, the performing doppler center correction on the multi-channel echo data after channel amplitude error correction respectively includes:

and moving the frequency spectrum center of each channel echo data after channel amplitude error correction to a zero frequency position.

In an optional embodiment, the performing baseband doppler frequency linear phase correction on the preprocessed multi-channel echo data respectively includes:

and respectively transforming the preprocessed echo data of each channel to an azimuth frequency domain, and then multiplying the data by a corresponding compensation item.

In an optional embodiment, the form of the compensation term includes:

H_m(f_a)＝exp(-j2πx_mf_a/v)

wherein H_m(f_a) Represents the compensation term, x, corresponding to the mth channel in the azimuth frequency domain_mDenotes the relative position of the mth channel and the reference channel, f_aIndicating azimuth frequency and v indicating radar velocity.

In an optional embodiment, the performing, by using an image subspace method, channel phase error estimation on the multi-channel image set includes:

obtaining a first covariance matrix corresponding to the multi-channel image set;

decomposing the eigenvalue of the first covariance matrix to obtain a first noise subspace corresponding to the multi-channel image set;

based on the first noise subspace, solving an optimal solution of a first function which is constructed in advance to obtain a channel phase error estimation value matrix; wherein the first function is an optimal objective function of a channel phase error estimation problem.

In an optional embodiment, the expression of the first function includes:

wherein Γ represents a diagonal matrix composed of phase error terms between channels; a is_iIndicating a steering vector corresponding to the azimuth ambiguity with the sequence number i; argmin {. } represents a variable value at which an objective function in { } takes a minimum value; i represents the change serial number of the orientation fuzzy quantity, and the value range of I is [ -I, I]I is larger than 0, and (2I +1) azimuth ambiguities are counted; { }^HRepresenting a conjugate transpose operation on a matrix; u shape_nRepresenting a first noise subspace.

In an alternative embodiment, the performing a channel baseline error estimation on the phase-corrected multi-channel image set includes:

obtaining a second covariance matrix corresponding to the multi-channel image set after phase correction;

decomposing the eigenvalue of the second covariance matrix to obtain a second noise subspace corresponding to the multichannel image set after the phase correction;

based on the second noise subspace, solving the optimal solution of a second function which is constructed in advance by using a least square method to obtain a channel baseline error estimation value matrix; wherein the second function is an optimal objective function of the measurement error estimation problem along the azimuth baseline.

In an optional embodiment, the expression of the second function includes:

wherein, Δ X represents a channel baseline error estimation value matrix; a is_iIndicating a steering vector corresponding to the azimuth ambiguity with the sequence number i; argmin {. } represents a variable value at which an objective function in { } takes a minimum value; i represents the change serial number of the orientation fuzzy quantity, and the value range of I is [ -I, I]I is larger than 0, and (2I +1) azimuth ambiguities are counted; { }^HRepresenting a conjugate transpose operation on a matrix; u denotes the second noise subspace.

In the scheme provided by the embodiment of the invention, the processing flow of the existing phase error estimation method is improved, a mechanism for processing multi-channel SAR data based on post-imaging correction channel errors is provided, and the multi-channel phase error and baseline error are jointly estimated in a complex image domain. Compared with the prior frequency domain estimation method, the clear region of the original data cannot be distinguished effectively, the embodiment of the invention adopts a processing mode of imaging first and then estimating, because the energy of the imaged signal is more concentrated than that before imaging and the echo data is localized, the blocking processing can be facilitated, and when the sub-image with high signal-to-noise ratio is selected to carry out the channel phase error estimation, the better error estimation effect can be obtained. Compared with the existing time domain estimation method, the method provided by the embodiment of the invention does not need to carry out cyclic search on the image, so that the calculated amount can be reduced, the baseline error estimation is introduced into the image domain and is combined with the channel phase error estimation, the reduction of multi-channel signal reconstruction and orientation ambiguity suppression performance caused by the inaccurate orientation baseline and the inconsistent channel phase can be reduced, the calculated amount of the image domain error estimation can be reduced, and the quality of the HRWS SAR image is improved.

Therefore, the multi-channel SAR data processing mechanism based on the post-imaging correction channel error separates imaging from error estimation, and has great significance for exploring new application scenes in the future.

The present invention will be described in further detail with reference to the accompanying drawings and examples.

Drawings

Fig. 1 is a schematic flow chart of a method for jointly estimating a domain phase and a baseline error of a multi-channel SAR complex image according to an embodiment of the present invention;

fig. 2 is a simplified flowchart schematic diagram of a multi-channel SAR complex image domain phase and baseline error joint estimation method according to an embodiment of the present invention;

FIG. 3 is a schematic diagram of the geometry of a MACs SAR system provided by an embodiment of the present invention;

FIG. 4(a) is a grayscale map corresponding to an imaging result obtained without using channel error estimation and compensation;

FIG. 4(b) is a grayscale image corresponding to the imaging results obtained by the method of the embodiment of the invention;

FIG. 5(a) is a slice along the 1360 th range cell of the imaging results of FIG. 4 (a);

fig. 5(b) is a slice along the 1360 th range cell of the imaging result of fig. 4 (b).

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

In order to improve the error estimation result of the MACs SAR system and obtain the HRWS SAR image with better quality, the embodiment of the invention provides a joint estimation method for the phase and baseline error of a multi-channel SAR complex image domain.

It should be noted that the MACs SAR system according to the embodiment of the present invention may be an airborne radar system, and the system may store echo data in a storage hard core, and an execution subject of the multi-channel SAR complex image domain phase and baseline error joint estimation method provided by the embodiment of the present invention may be a data processing device, such as a computer, and the data processing device may acquire the multi-channel echo data recorded by the MACs SAR system through data transmission, data replication, and the like, so as to perform a subsequent processing process, and a specific data processing tool may be Matlab and the like.

As shown in fig. 1, a method for jointly estimating a domain phase and a baseline error of a multi-channel SAR complex image according to an embodiment of the present invention may include the following steps:

and S1, respectively preprocessing the acquired multi-channel echo data of the MACs SAR system.

The MACs SAR system provided by the embodiment of the invention has (2I +1) M-channel SAR signals with fuzzy orientation, namely echo data of M channels in total; the two-dimensional (distance and orientation) frequency domain signal of the mth channel can be represented in the form:

wherein S is_c,m(f_r,f_a) Representing a two-dimensional frequency domain signal of an mth channel, wherein M is more than or equal to 1 and less than or equal to M, and M is a natural number more than 1; i represents a change number of the orientation ambiguity number; the value range of I is [ -I, I]I is greater than 0; gamma-shaped_c,mAnd Δ θ_mRespectively representing the amplitude error and the phase error of the mth channel; exp represents an exponential function with a natural constant e as the base; j represents the imaginary part of the complex number; s₀' represents an ideal two-dimensional signal spectrum without azimuthal ambiguity; f. of_rRepresents a range frequency; f. of_aRepresenting the azimuth frequency; f. of_dRepresents the doppler center frequency; f. of_pRepresents the Pulse Repetition Frequency (PRF); x is the number of_mRepresenting the relative positions of the mth channel and the reference channel, i.e. the corresponding radar antenna baseline; Δ x_mRepresenting the azimuth baseline error of the radar antenna corresponding to the mth channel; v denotes the radar speed. Wherein S is_c,mAnd Γ_c,mC in the subscript represents a channel, m being a variable; f. of_r、f_aAnd f_pIs a self-defined parameter.

In an optional implementation manner, the preprocessing is performed on the acquired multi-channel echo data of the MACs SAR system, and includes the following steps:

and S11, respectively carrying out channel amplitude error correction on the multi-channel echo data.

Specifically, the step is to adjust the average amplitude value of the multi-channel echo data to be consistent.

For example, the average amplitude values of the echo data of M channels may be made uniform by averaging/weighted averaging, normalization, and the like, that is, Γ may be implemented_c,mThe same is true. After treatment, gamma in (1)_c,mNo longer present, the results are as follows:

and S12, respectively performing Doppler center correction on the multi-channel echo data after the channel amplitude error correction.

The step is to compensate the Doppler center of each channel signal under the squint condition, and specifically, to move the frequency spectrum center of each channel echo data after channel amplitude error correction to the zero frequency position. F in (2) after treatment_dNo longer present, the results are as follows:

and S2, respectively carrying out baseband Doppler frequency linear phase correction on the preprocessed multi-channel echo data to obtain frequency linear phase correction data of each channel.

In an alternative embodiment, the steps include:

and respectively transforming the preprocessed echo data of each channel to an azimuth frequency domain, and then multiplying the data by a corresponding compensation item.

The construction form of the compensation term is as follows:

H_m(f_a)＝exp(-j2πx_mf_a/v) (4)

wherein H_m(f_a) And the compensation term corresponding to the mth channel in the azimuth frequency domain is shown.

The embodiment of the invention converts the echo data of each channel into an azimuth frequency domain (azimuth Doppler domain), compensates the linear phase difference of the base band Doppler frequency of each channel data caused by an azimuth baseline, and realizes the aim of aligning the multi-channel echo data on an azimuth time domain relative to a reference channel.

And S3, respectively carrying out SAR imaging processing on the frequency linear phase correction data of each channel, and integrating the obtained images of each channel to obtain a multi-channel image set.

The same imaging algorithm is adopted for SAR imaging processing of each channel. The imaging algorithms include, but are not limited to, Range Doppler (RD), line tone Scaling (CS), and extension algorithms thereof.

In an alternative embodiment, the imaging algorithm may be a classical frequency domain corrected range walking RD algorithm. The imaging process of the algorithm is briefly described below, and mainly includes the following steps.

1) The frequency linear phase correction data of each channel is subjected to FFT (Fast Fourier transform) to transform the data into a distance frequency domain.

2) And multiplying each channel data obtained after transformation by the corresponding compensation item of frequency domain correction distance walking.

Wherein, the compensation term of the frequency domain correction distance walk can be constructed as:

wherein H_rmc(f_a,f_r) A compensation item representing the frequency domain correction distance walking corresponding to the mth channel; r_bRepresents the skew distance; f. of_aMMaximum doppler representing the echo; and c represents the speed of light.

3) And (4) performing azimuth matched filtering on the channel data processed in the step (2).

Specifically, the data of each channel processed in step 2 is transformed to a distance time domain by using inverse FFT, and then multiplied by a matched filtering reference function.

Since imaging the azimuth signal may be equivalent to matched filtering the azimuth signal, the matched filtering reference function for the baseband frequency may be constructed as:

4) and transforming the processed data of each channel in the step 3) into an azimuth time domain through two-dimensional inverse Fourier transform to obtain a multi-channel image set formed by multi-channel images.

After azimuth matched filtering and two-dimensional inverse Fourier transform to an azimuth time domain, real points in the image can be focused in the two-dimensional time domain, but due to the fact that residual distance migration exists in ghost images caused by azimuth blurring, the ghost images are not completely focused. The multi-channel image set at this time can be expressed in the form:

wherein, F_cA multi-channel image set representing a multi-channel image composition in a matrix form; Γ represents a diagonal matrix composed of phase error terms between channels; a represents a matrix formed by image domain steering vectors containing antenna baseline errors;representing different frequency band images constituting a multi-channel image; n denotes a gaussian white noise matrix. Specifically, the method comprises the following steps:

where diag denotes a function used in FreeMatlab, Matlab to construct a diagonal matrix.

A＝[b_-Ia_-I … b_ia_i … b_Ia_I] (9)

Wherein the content of the first and second substances,

b_i＝diag[ε_1,i … ε_m,i …ε_M,i] (10)

wherein epsilon_m,iAnd (3) phase terms brought by the baseline error to the channel m and the azimuth fuzzy component i are shown.

Wherein, a_iThe steering vector corresponding to the azimuth ambiguity with index i is shown.

a_i＝[κ_1,i … κ_m,i … κ_M,i]^T (12)

Wherein, κ_m,iThe phase terms brought to the channel m and the azimuth ambiguity component i by the antenna base line are shown.

Wherein, F_i' representing the sub-images obtained by imaging the corresponding blurred components of the frequency bands with different serial numbers, where theoretical analysis is used to represent the actual acquisition of the multi-channel image set F_cCan be made up of the set of images.

It is to be understood that, after the step S3 is executed, the subsequent processing steps are performed in the complex image domain.

And S4, performing channel phase error estimation on the multi-channel image set by using an image subspace method, and performing phase correction on each channel image according to the estimation result of the channel phase error.

In an optional embodiment, the performing channel phase error estimation on the multi-channel image set by using an image subspace method includes the following steps:

a1, a first covariance matrix corresponding to the multi-channel image set is obtained.

Matrix F_cThe analysis process of the covariance matrix mainly comprises the following expressions:

wherein E {. is } represents an averaging operation; { }^HRepresenting a conjugate transpose operation on a matrix;a power representing noise; i is_nRepresenting an identity matrix.

In the embodiment of the invention, the matrix F is used for distinguishing conveniently_cIs named as the first covariance matrix.

A2, decomposing the eigenvalues of the first covariance matrix to obtain a first noise subspace corresponding to the multi-channel image set.

Decomposing the first covariance matrixThe feature value of (2) can be used to obtain an image space. According to the size of the characteristic value obtained by decomposition, the obtained image space can be divided into signal subspaces U of the image_sNoise subspace U of sum image_n. The dividing basis is that each characteristic value is compared with a preset characteristic value threshold, a signal subspace of the image is formed by characteristic vectors corresponding to the characteristic values larger than the characteristic value threshold, a noise subspace of the image is formed by the characteristic vectors corresponding to the characteristic values smaller than the characteristic value threshold, and the signal subspace and the noise subspace are orthogonal theoretically. In the embodiment of the present invention, for convenience of distinction, the signal subspace Us obtained in the step is named as a first signal subspace, and the noise subspace U obtained in the step is named as a second signal subspace U_nNamed first noise subspace, it is understood that Us and U_nIn the form of a matrix.

The decomposition of the characteristic value meets the following two conditions: characteristic value lambda_mSatisfy the requirement ofCharacteristic value lambda_mThe corresponding feature vector is u_mWherein M is 1,2, …, M; each column of the steering vector matrix of signals is orthogonal to the first noise subspace U_nEach column of (1), U_n＝[u_2L+2,u_2L+3,...,u_M]. For a specific process of feature value decomposition, please refer to related prior art, which is not described herein.

And A3, solving the optimal solution of the pre-constructed first function based on the first noise subspace to obtain a channel phase error estimation value matrix.

Wherein the first function is an optimal objective function of a channel phase error estimation problem.

An expression of the first function, comprising:

wherein Γ represents a diagonal matrix composed of phase error terms between channels; a is_iIndicating a steering vector corresponding to the azimuth ambiguity with the sequence number i; argmin {. } represents a variable value at which an objective function in { } takes a minimum value; i represents the change serial number of the orientation fuzzy quantity, and the value range of I is [ -I, I]I is larger than 0, and (2I +1) azimuth ambiguities are counted; { }^HRepresenting a conjugate transpose operation on a matrix; u shape_nRepresenting a first noise subspace.

(16) Equivalent to the formula:

δ＝ve＝Γ (18)

where vec represents the operation of the function in matlab on the diagonal elements of the matrix.

D_i＝diag(a_i) (19)

Under the linear constraint of δ H_w＝1,w＝[1,0,0,…,0]^TThen, an optimal solution of the first function can be obtainedWhich represents a matrix of channel phase error estimates.

Then, the performing phase correction on each channel image according to the channel phase error estimation result may include:

by compensating the obtained estimated value of the phase error to each channel image, equation (7) can be changed to:

F′_c(τ,t_a)＝AF'+N (22)

wherein, F'_c(τ,t_a) Representing a phase corrected multi-channel image set.

And S5, channel baseline error estimation is carried out on the multichannel image set after phase correction, and updating along the orientation baseline value is carried out by utilizing the estimated channel baseline error.

In an alternative embodiment, the performing the channel baseline error estimation on the phase-corrected multi-channel image set includes the following steps:

and B1, obtaining a second covariance matrix corresponding to the phase-corrected multi-channel image set.

And B2, decomposing the eigenvalue of the second covariance matrix to obtain a second noise subspace corresponding to the multi-channel image set after phase correction.

B1-B2 represent sets F 'of phase-corrected multi-channel images'_c(τ,t_a) The image space is counted and spatial decomposition is performed, and the specific process is similar to a1 and a2, which are not described herein.

Let U be the noise subspace corresponding to the MACs SAR system after the residual constant phase compensation, which is named as the second noise subspace for the sake of distinction in the embodiments of the present invention.

And B3, solving the optimal solution of the pre-constructed second function by using a least square method based on the second noise subspace to obtain a channel baseline error estimation value matrix.

Wherein the second function is pre-constructed in order to estimate the measurement error matrix along the azimuthal baseline. The second function is an optimal objective function of the measurement error estimation problem along the azimuth baseline.

An expression of the second function, comprising:

wherein, Δ X represents a channel baseline error estimate vector; a is_iIndicating a steering vector corresponding to the azimuth ambiguity with the sequence number i; argmin {. } represents a variable value at which an objective function in { } takes a minimum value; i represents the change serial number of the orientation fuzzy quantity, and the value range of I is [ -I, I]I is greater than 0, in total

(2I +1) azimuth ambiguities;^Hrepresenting a conjugate transpose operation on a matrix; u denotes the second noise subspace.

Equation (23) is equivalent to minimizing the following function J:

J＝min||X+YΔd|| (24)

wherein min represents the minimum value; solving 2 norm by | l | · | |; x represents; y represents; and deltad represents a diagonal matrix of channel baseline errors.

Namely:

X＝-YΔd (25)

wherein:

solving equation (25) by using least square method to obtain channel baseline error estimation result, namely channel baseline error estimation value matrixThe channel baseline error estimation value matrix comprises a plurality of channel baseline error estimation values used for representing the measurement error along the azimuth baseline.

Wherein real is the operation of extracting the real part of the complex variable.

Then, the updating along the azimuth baseline value by using the estimated channel baseline error may include:

and correcting the measured value of the channel baseline by using the obtained error estimated value of the channel baseline to obtain an accurate value along the azimuth baseline. The process can be represented as follows:

wherein the content of the first and second substances,the updated value along the orientation baseline of the mth channel is shown, namely the accurate value along the orientation baseline is obtained; x is the number of_mA measurement along the azimuth baseline, i.e., a measurement of the channel baseline, representing the mth channel;representing the channel baseline error estimate for the mth channel.

And S6, repeatedly performing channel phase error estimation on the multi-channel image set by using the image subspace method by using the updated value along the orientation baseline until the step of updating the value along the orientation baseline is performed until the iteration termination condition is met.

The step is to repeat the steps S4-S5 until the iteration termination condition is met by using the updated along-orientation baseline value. The iteration termination condition may be that the number of iterations reaches a predetermined number, for example, in practice, two iterations are generally selected to achieve higher estimation accuracy.

Using updated values along the azimuthal baselineMore recent formula (7) and x referred to later in the formula_mAnd re-performing channel phase error estimation and channel baseline error estimation. The specific process should be understood in conjunction with the foregoing steps, and will not be described in detail herein.

And S7, obtaining the HRWS SAR image by using the value along the orientation baseline and the channel phase error when the iteration is terminated.

The step is to use the finally obtained phase error to compensate multi-channel data, use the finally obtained along-direction baseline value to reconstruct Doppler spectrum without direction ambiguity of the multi-channel data, and use SAR imaging to obtain HRWS SAR image. The specific process of this step is understood by referring to the related art and will not be described in detail here.

Fig. 2 is a simplified flow chart of the method according to the embodiment of the present invention, and the detailed contents thereof are understood in conjunction with the foregoing description, and are not repeated herein.

In the scheme provided by the embodiment of the invention, the processing flow of the existing phase error estimation method is improved, a mechanism for processing multi-channel SAR data based on post-imaging correction channel errors is provided, and the multi-channel phase error and baseline error are jointly estimated in a complex image domain. Compared with the prior frequency domain estimation method, the clear region of the original data cannot be distinguished effectively, the embodiment of the invention adopts a processing mode of imaging first and then estimating, because the energy of the imaged signal is more concentrated than that before imaging and the echo data is localized, the blocking processing can be facilitated, and when the sub-image with high signal-to-noise ratio is selected to carry out the channel phase error estimation, the better error estimation effect can be obtained. Compared with the existing time domain estimation method, the method provided by the embodiment of the invention does not need to carry out cyclic search on the image, so that the calculated amount can be reduced, the baseline error estimation is introduced into the image domain and is combined with the channel phase error estimation, the reduction of multi-channel signal reconstruction and orientation ambiguity suppression performance caused by the inaccurate orientation baseline and the inconsistent channel phase can be reduced, the calculated amount of the image domain error estimation can be reduced, and the quality of the HRWS SAR image is improved.

Therefore, the multi-channel SAR data processing mechanism based on the post-imaging correction channel error separates imaging from error estimation, and has great significance for exploring new application scenes in the future.

In order to facilitate understanding of the relevant effects of the embodiments of the present invention, simulation experiment data will be described. As shown in fig. 3, fig. 3 is a schematic geometric configuration diagram of a MACs SAR system according to an embodiment of the present invention.

In the system, five-channel antenna arrays representing directions are uniformly distributed along the track direction, the distance between every two antenna arrays is d, a transmitting channel is positioned in the center of the antenna arrays and is represented by T, and each channel of the antenna arrays can be used as a channel for receiving echo signals. X, Y and Z represent spatial axes. R_bRepresenting radar flight R₁～R₅The slope of the trajectory into the scene. v denotes radar speed (flying platform speed). Scene represents a radar-illuminated ground Scene. P denotes a scene center point. The relevant parameters are shown in table 1.

TABLE 1 simulation parameters Table

Parameter(s)	Parameter value	Unit of
			In the sceneHeart slope distance	15	Km
Carrier frequency	10	GHz
			Flying platform velocity	180	m/s
Bandwidth of transmitted signal	100	MHz
			Pulse repetition frequency	240	Hz

Under the condition that the same channel phase error and antenna baseline error exist, the estimation performance of the multi-channel phase and baseline error joint estimation method provided by the embodiment of the invention is analyzed, and the simulation result is shown in table 2:

TABLE 2 simulation experiment channel phase error estimation results

TABLE 3 simulation experiment channel baseline error estimation results

As can be seen from the comparison of the estimation results of the channel phase error and the channel baseline error in tables 2 and 3, by using the method provided in the embodiment of the present invention, the estimated channel error is close to the actual channel phase error set in the simulation experiment, so that the validity of the multi-channel phase and baseline error joint estimation method provided in the embodiment of the present invention can be verified.

Referring to fig. 4 and 5, fig. 4(a) is a grayscale diagram of an imaging result obtained without using channel error estimation and compensation, and fig. 4(b) is a grayscale diagram of an imaging result obtained by the method according to an embodiment of the present invention. Comparing fig. 4(a) and fig. 4(b), it is found that the multi-channel imaging result obtained by the channel error compensation and the accurate baseline reconstruction estimated by the method of the embodiment of the present invention is clearer, and the fuzzy point along the azimuth direction is obviously suppressed.

To further illustrate the effect of the method of the embodiment of the present invention on the imaging, the 1360-th distance in fig. 4(a) and 4(b) is selected to be a slice along the azimuth direction, which corresponds to fig. 5(a) and 5(b), respectively. In fig. 5(a) and 5(b), the horizontal axis represents azimuth time, and the vertical axis represents signal amplitude. In contrast, the magnitude of the azimuth ambiguity point in fig. 5(b) is suppressed to about-57 dB and 76dB, relative to the magnitude of the azimuth ambiguity point in the result of fig. 5(a) without channel error compensation, and thus, the accuracy of the estimation of the channel phase error and the baseline error in tables 2 and 3 can be shown. Therefore, the method provided by the embodiment of the invention can reduce the degradation of multi-channel signal reconstruction and azimuth ambiguity suppression performance caused by the inaccurate azimuth baseline and the inconsistent channel phase by performing the joint estimation of the channel phase error and the baseline error in the image domain, and can improve the HRWS imaging quality.

It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.

All the embodiments in the present specification are described in a related manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the system embodiment, since it is substantially similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment.

The above description is only for the preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention shall fall within the protection scope of the present invention.