CN-122017757-A - Method and system for correcting channel phase deviation of multi-channel SAR

CN122017757ACN 122017757 ACN122017757 ACN 122017757ACN-122017757-A

Abstract

The invention provides a channel phase deviation correction method and system of a multichannel SAR, comprising the steps of inputting azimuth multichannel SAR echo data, carrying out distance compression on each channel echo data, calculating a correlation coefficient between instantaneous adjacent channels and an optimal correlation coefficient between channels at adjacent azimuth moments, adaptively weighting and estimating Doppler center frequency, inverting the phase deviation between channels by utilizing the Doppler center frequency, correcting echo or imaging data by utilizing the phase deviation, and outputting corrected data. The invention only utilizes the self statistical property of echo data, does not need an external calibration source, obviously improves the accuracy and the robustness of phase deviation estimation, effectively solves the problem of channel mismatch in the staggered pulse repetition interval mode, and provides a reliable technical basis for high-resolution wide-swath imaging and moving target detection.

Inventors

FAN JIXIA
LIU YANYANG
Lv Zongsen
CHEN QUAN
XUE LINGLING
ZHAO YANG

Assignees

上海卫星工程研究所

Dates

Publication Date: 20260512
Application Date: 20260202

Claims (10)

1. A channel phase deviation correction method of a multi-channel SAR is characterized by comprising the following steps: s1, acquiring original echo data of an interleaved pulse repetition interval azimuth multi-channel SAR; Step S2, respectively performing distance compression processing on the original echo data of each channel obtained in the step S1 to obtain echo data of each channel after distance compression; S3, estimating a correlation coefficient between instantaneous adjacent channels and an optimal correlation coefficient between channels at adjacent azimuth moments based on the echo data of each channel after the distance compression obtained in the step S2; Step S4, carrying out self-adaptive weighted estimation of Doppler center frequency by combining the instantaneous adjacent inter-channel correlation coefficient obtained in the step S3 and the optimal inter-channel correlation coefficient at the adjacent azimuth moment; S5, inverting to obtain phase deviation values among all channels based on the Doppler center frequency estimation result obtained in the step S4; s6, carrying out phase correction processing on the echo data after the distance compression in the step S2 or the imaging data generated based on the echo data by utilizing the phase deviation value among channels obtained by inversion in the step S5; and S7, outputting the echo data or the imaging data subjected to the phase deviation correction in the step S6, and finishing the phase deviation estimation and correction flow between the whole channels.
2. The method for correcting channel phase deviation of multi-channel SAR according to claim 1, wherein said step S3 specifically comprises the steps of: Step S3.1, estimating a correlation coefficient between the instantaneous adjacent channels, wherein the estimation expression of the correlation coefficient between the instantaneous adjacent channels is as follows: Wherein, the Indicating azimuth time A correlation coefficient between the mth receiving channel and the (m+1) th receiving channel; Indicating azimuth time A distance compressed signal at an nth distance sampling point of an mth receiving channel; Indicating azimuth time Distance compressed signals at the (m+1) th receiving channel and the (n) th distance sampling point; Superscript of (2) Represents conjugation; Is an integer in the range of 1 to , Representing the total number of channels; Representing summing the distance compressed signals at all the distance sampling points of the receiving channel; step S3.2, estimating the optimal correlation coefficient between channels at adjacent azimuth moments, and defining azimuth moments First, the Individual channels and azimuth moments First, the The correlation coefficient of each channel is optimal, and then the estimated expression of the optimal correlation coefficient between the channels at adjacent azimuth moments is: Wherein, the Indicating azimuth time First, the Each channel and azimuth moment First, the Correlation coefficients between the individual channels; Indicating azimuth time Location No The distance compressed signal at the nth distance sampling point of the receiving channel; Indicating azimuth time Location No The distance compressed signal at the nth distance sampling point of the receiving channel; Superscript of (2) Represents conjugation; Indicating azimuth time Pulse repetition interval with next azimuth moment; And Is an integer in the range of 1 to , Indicating the total number of channels.
3. The method for correcting channel phase deviation of multi-channel SAR according to claim 2, wherein in step S3.2, the azimuth time is set First, the Individual channels and azimuth moments First, the The selection of each channel satisfies the shortest space-time distance of the channel, and the expression is: Wherein, the Represent the first The first channel The channel space-time distance of each channel is argmin I which is a function taking the minimum value; is azimuth time Time No Three-dimensional spatial positions of the individual channels; Indicating the azimuth time as Time No Three-dimensional spatial positions of the individual channels; Indicating azimuth time Pulse repetition interval with next azimuth moment; And Is an integer in the range of 1 to , Indicating the total number of channels.
4. The method for correcting the channel phase deviation of the multi-channel SAR according to claim 2, wherein said step S4 specifically comprises the steps of: s4.1, estimating the instantaneous Doppler center and azimuth time Doppler center estimate at time The method comprises the following steps: Wherein, the Is the azimuth moment The corresponding pulse repetition frequency, angle {.cndot }, is a radial angle function, For an optimal coherence coefficient between adjacent time channels, In order to take the sign of the continuous multiplication, Is the correlation coefficient between adjacent channels; and S4.2, carrying out weighted fusion on the Doppler center of the full scene, wherein the expression is as follows: Wherein, the Is azimuth time The estimated value of the doppler center at the time, Representing time for all orientations The corresponding terms are summed up and, Is azimuth time The Doppler center estimated value weight is calculated according to the following formula: Wherein, the Is the azimuth moment The corresponding pulse repetition frequency is set to be the same, For an optimal coherence coefficient between adjacent time channels, Is the correlation coefficient between adjacent channels.
5. The method for correcting channel phase deviation of multi-channel SAR according to claim 2, wherein said step S5 specifically comprises the steps of: step S5.1, calculating the coherence coefficient of the adjacent channels of the whole scene The calculation formula is as follows; Wherein, the In order to be the number of azimuth moments, Is the azimuth moment Correlation coefficient estimation values between the mth and m+1th channels; step S5.2, calculating phase deviation between adjacent channels The calculation formula is as follows: Wherein, the For the full scene correlation coefficient between the mth and the m +1 th channels, j represents the imaginary unit, Is the full scene doppler center estimate, For the spatial separation between the phase centers of adjacent channel antennas, Is satellite speed; Step S5.3, calculating the phase deviation of each channel relative to the reference channel, and the phase deviation of the mth channel relative to the 1 st channel The method comprises the following steps: Wherein, the Is the phase offset between the kth channel and the k+1th channel.
6. A channel phase offset correction system for a multi-channel SAR, comprising: the method comprises the steps of M1, acquiring original echo data of an interleaved pulse repetition interval azimuth multichannel SAR; the module M2 is used for respectively carrying out distance compression processing on the original echo data of each channel acquired in the module M1 to acquire echo data of each channel after the distance compression; The module M3 is used for estimating the correlation coefficient between the instantaneous adjacent channels and the optimal correlation coefficient between the channels at the adjacent azimuth moment based on the echo data of each channel after the distance compression obtained by the module M2; the module M4 is used for carrying out self-adaptive weighted estimation of Doppler center frequency by combining the instantaneous adjacent inter-channel correlation coefficient obtained by the module M3 and the optimal inter-channel correlation coefficient at the adjacent azimuth moment; The module M5 is used for inverting to obtain phase deviation values among the channels based on the Doppler center frequency estimation result obtained by the module M4; The module M6 is used for carrying out phase correction processing on echo data after distance compression in the module M2 or imaging data generated based on the echo data by utilizing the phase deviation value among channels obtained by inversion of the module M5; And the module M7 is used for outputting the echo data or imaging data subjected to phase deviation correction in the module M6 to complete the phase deviation estimation and correction flow between the whole channels.
7. The system for correcting channel phase deviation of multi-channel SAR according to claim 6, wherein said module M3 comprises in particular the following modules: the module M3.1 estimates the correlation coefficient between the instantaneous adjacent channels, and the estimation expression of the correlation coefficient between the instantaneous adjacent channels is: Wherein, the Indicating azimuth time A correlation coefficient between the mth receiving channel and the (m+1) th receiving channel; Indicating azimuth time A distance compressed signal at an nth distance sampling point of an mth receiving channel; Indicating azimuth time Distance compressed signals at the (m+1) th receiving channel and the (n) th distance sampling point; Superscript of (2) Represents conjugation; Is an integer in the range of 1 to , Representing the total number of channels; Representing summing the distance compressed signals at all the distance sampling points of the receiving channel; Module M3.2 estimating the optimal correlation coefficient between channels at adjacent azimuth moments, defining azimuth moments First, the Individual channels and azimuth moments First, the The correlation coefficient of each channel is optimal, and then the estimated expression of the optimal correlation coefficient between the channels at adjacent azimuth moments is: Wherein, the Indicating azimuth time First, the Each channel and azimuth moment First, the Correlation coefficients between the individual channels; Indicating azimuth time Location No The distance compressed signal at the nth distance sampling point of the receiving channel; Indicating azimuth time Location No The distance compressed signal at the nth distance sampling point of the receiving channel; Superscript of (2) Represents conjugation; Indicating azimuth time Pulse repetition interval with next azimuth moment; And Is an integer in the range of 1 to , Indicating the total number of channels.
8. The system for correcting channel phase bias of multi-channel SAR according to claim 7, wherein in said module M3.2, the azimuth time is First, the Individual channels and azimuth moments First, the The selection of each channel satisfies the shortest space-time distance of the channel, and the expression is: Wherein, the Represent the first The first channel The channel space-time distance of each channel is argmin I which is a function taking the minimum value; is azimuth time Time No Three-dimensional spatial positions of the individual channels; Indicating the azimuth time as Time No Three-dimensional spatial positions of the individual channels; Indicating azimuth time Pulse repetition interval with next azimuth moment; And Is an integer in the range of 1 to , Indicating the total number of channels.
9. The system for correcting channel phase deviation of multi-channel SAR according to claim 7, wherein said module M4 comprises the following modules: module M4.1 estimating the instantaneous Doppler centre, azimuth time Doppler center estimate at time The method comprises the following steps: Wherein, the Is the azimuth moment The corresponding pulse repetition frequency, angle {.cndot }, is a radial angle function, For an optimal coherence coefficient between adjacent time channels, In order to take the sign of the continuous multiplication, Is the correlation coefficient between adjacent channels; And a module M4.2, wherein the weighted fusion is carried out on the Doppler center of the full scene, and the expression is as follows: Wherein, the Is azimuth time The estimated value of the doppler center at the time, Representing time for all orientations The corresponding terms are summed up and, Is azimuth time The Doppler center estimated value weight is calculated according to the following formula: Wherein, the Is the azimuth moment The corresponding pulse repetition frequency is set to be the same, For an optimal coherence coefficient between adjacent time channels, Is the correlation coefficient between adjacent channels.
10. The system for correcting channel phase deviation of multi-channel SAR according to claim 7, wherein said module M5 comprises in particular the following modules: module M5.1 calculating the coherence coefficient of the adjacent channels of the whole scene The calculation formula is as follows; Wherein, the In order to be the number of azimuth moments, Is the azimuth moment Correlation coefficient estimation values between the mth and m+1th channels; module M5.2 calculating phase deviation between adjacent channels The calculation formula is as follows: Wherein, the For the full scene correlation coefficient between the mth and the m +1 th channels, j represents the imaginary unit, Is the full scene doppler center estimate, For the spatial separation between the phase centers of adjacent channel antennas, Is satellite speed; module M5.3 calculating the phase offset of each channel relative to the reference channel, the phase offset of the mth channel relative to the 1 st channel The method comprises the following steps: Wherein, the Is the phase offset between the kth channel and the k+1th channel.

Description

Method and system for correcting channel phase deviation of multi-channel SAR Technical Field The invention relates to the technical field of synthetic aperture radar signal processing, in particular to a channel phase deviation correction method and system of a multi-channel SAR, and particularly relates to a channel phase deviation correction method suitable for staggered pulse repetition interval azimuth multi-channel SAR. Background The azimuth multi-channel SAR technique is an effective way to achieve high resolution wide swath imaging, but its performance is limited by phase inconsistencies between the receive channels. Such phase deviations introduced by hardware differences can disrupt signal coherence, leading to reduced imaging quality and advanced application failure. Particularly, when complex patterns such as staggered Pulse Repetition Interval (PRI) are used, the signal space-time relationship is complicated, and the phase error problem is more remarkable. The traditional correction method relies on an external scaler or manually selecting a point target, is complex in process and difficult to process in real time, and severely restricts the practical development of the multichannel SAR system. Therefore, the inter-channel phase deviation estimation technology suitable for the staggered pulse interval azimuth multi-channel SAR has important theoretical and application values, and can break through the fundamental contradiction that the resolution and the mapping bandwidth are mutually restricted in the traditional single-channel SAR system. The space-time cross correlation coefficient (STCCC) algorithm proposed in literature On the Baseband Doppler Centroid Estimation for Multichannel HRWS SAR IMAGING (Yanyang Liu et al, IEEE Geosciene and Remote SENSING LETTERS, vol.11, no.12, 2014) makes full use of the correlation between channels, and can estimate the phase deviation between channels while effectively estimating the azimuth multi-channel SAR baseband doppler center. The STCCC method is only valid in a fixed pulse repetition Period (PRF) mode, which is no longer true in an interleaved PRI system based on the assumption of a fixed spatial-temporal relationship, resulting in phase estimation bias or even failure. Document "Robust Channel Phase Error Calibration Algorithm for Multichannel High-Resolution and Wide-Swath SAR Imaging"（L. Zhang, Y. Gao and X. Liu, IEEE Geoscience and Remote Sensing Letters, vol. 14, no. 5, pp. 649-653） proposes a robust channel phase error calibration algorithm based on minimum variance distortion free response (MVDR) beamformer output power maximization. The output power of the MVDR beamformer will be maximized with an accurate steering vector. The method directly searches for a phase error estimation value capable of maximizing the MVDR output power by constructing an optimization problem without performing feature decomposition necessary for the conventional subspace method. It uses the inverse of the covariance matrix and the nominal steering vector to construct an optimization function to estimate the channel phase error. Although this approach avoids subspace decomposition and does not require channel redundancy, its performance relies on accurate estimation of the covariance matrix, which requires a sufficient number of range-doppler cell samples. In cases where low signal-to-noise ratios or scene features are not apparent, the estimation of the covariance matrix may be inaccurate, thereby affecting the accuracy of the phase error estimation. But this method is only effective in fixed PRF mode, which will no longer hold in an interleaved PRI system based on the assumption of a fixed spatio-temporal relationship, resulting in phase estimation bias or even failure. Document "A Novel Channel Errors Calibration Algorithm for Multichannel High-Resolution and Wide-Swath SAR Imaging"（H. Huanget al., IEEE Transactions on Geoscience and Remote Sensing, vol. 60, pp. 1-19） proposes a multi-channel high-resolution wide swath synthetic aperture radar (HRWS-SAR) channel error calibration algorithm based on orthogonal projection theory. The method comprises the steps of estimating and compensating a distance synchronous time error through a Total Least Squares (TLS) technology, calibrating an amplitude error through local cross correlation processing, and estimating and compensating a phase error among channels by constructing an orthogonal projection weight vector in a Doppler domain so as to output a total power maximization criterion. The algorithm avoids covariance matrix eigenvalue decomposition in the traditional subspace method, so that the signal leakage problem is reduced under the condition of low signal-to-noise ratio (SNR), and the robustness and the accuracy of error estimation are improved. But this method is only effective in fixed PRF mode, which is no longer true in interleaved PRI systems based on the assumption of a fixed spatio-temporal relationship, r