CN-121995378-A - SAR periodic missing signal recovery method based on echo structure priori knowledge

CN121995378ACN 121995378 ACN121995378 ACN 121995378ACN-121995378-A

Abstract

The invention discloses a SAR periodic missing signal recovery method based on priori knowledge of an echo structure, which comprises the steps of obtaining an original echo signal obtained by sampling a synthetic aperture radar, dividing the original echo signal along a distance direction based on the sampling structure of the original echo signal to obtain a plurality of distance blocks, determining a target distance unit with the maximum signal-to-noise ratio in each distance block, calculating the recovery coefficient of each target distance unit, preprocessing the recovery coefficient of each target distance unit to obtain the target recovery coefficient of the distance block where each target distance unit is located, and determining the missing signal in the original echo signal according to the target recovery coefficient of each distance block, a predetermined target estimation error compensation parameter and the original echo signal. The invention can be popularized to the full distance section only by estimating the recovery coefficient of part of the distance units, and can greatly reduce the operation time while ensuring the pulse recovery precision.

Inventors

LI YIMING
ZHANG MINGZHU
XING MENGDAO

Assignees

西安电子科技大学

Dates

Publication Date: 20260508
Application Date: 20260108

Claims (8)

1. The SAR periodic missing signal recovery method based on the priori knowledge of the echo structure is characterized by comprising the following steps: acquiring an original echo signal obtained by sampling a synthetic aperture radar; based on the sampling structure of the original echo signal, the original echo signal is segmented along the distance direction to obtain a plurality of distance blocks, wherein each distance block comprises a plurality of sampling signals; determining a target distance unit with the maximum signal-to-noise ratio in each distance block, and calculating the recovery coefficient of each target distance unit; Preprocessing the recovery coefficient of each target distance unit to obtain a target recovery coefficient of a distance block where each target distance unit is located; And determining a missing signal in the original echo signal according to the target recovery coefficient of each distance block, a predetermined target estimation error compensation parameter and the original echo signal.
2. The method for recovering a SAR periodic missing signal based on a priori knowledge of echo structures according to claim 1, wherein said determining a missing signal in said original echo signal based on a target recovery coefficient of each of said distance blocks, a predetermined target estimation error compensation parameter, and said original echo signal, comprises: Based on a least square method, constructing an objective function comprising objective recovery coefficients of the distance blocks, sampling signals in the original echo signals and estimation error compensation parameters to be solved, and solving to obtain objective estimation error compensation parameters; And determining a missing signal in the original echo signal according to the target recovery coefficient of each distance block, the target estimation error compensation parameter and the sampling signal in the original echo signal.
3. The method for recovering a SAR periodic missing signal based on a priori knowledge of an echo structure according to claim 1, wherein said preprocessing the recovery coefficients of each of said target distance units to obtain the target recovery coefficients of each of said target distance units comprises: removing jump values in the recovery coefficients of the target distance units, and replacing the jump values by adopting the average value of the recovery coefficients of the target distance units in the left and right adjacent distance blocks to obtain the preprocessing recovery coefficients of the target distance units; and normalizing the preprocessing recovery coefficient of each target distance unit based on the radar slope distance of the predetermined reference distance unit and the radar slope distance of each target distance unit to obtain the target recovery coefficient of the distance block where each target distance unit is located.
4. The method for recovering a SAR periodic missing signal based on a priori knowledge of an echo structure according to claim 1, wherein said partitioning the original echo signal along a distance direction based on the sampling structure of the original echo signal to obtain a plurality of distance blocks comprises: Performing distance pulse pressure processing on the original echo signal to obtain a preprocessed echo signal; Based on the sampling structure of the original echo signals, the preprocessed echo signals are segmented along the distance direction, so that a plurality of distance blocks are obtained.
5. The method for recovering a periodic missing SAR signal based on a priori knowledge of echo structures according to claim 1, wherein the calculation formula of the recovery coefficient of each target distance unit is expressed as: Wherein, the Representing the recovery coefficients of each target distance cell, Representing the echo after the radar coherent detection, Representation and absence signal Distance from each other The sampling signals corresponding to the distance units of the azimuth sampling interval, Representing the length of the known azimuth sequence used for signal recovery, The ordinal number representing the known signal, The time of the fast-time period is indicated, Indicating a slow time period for which the time period is slow, The azimuth sampling time interval is represented as, Representation of Is a conjugate of (c).
6. The method for recovering a SAR periodic missing signal based on a priori knowledge of echo structure according to claim 2, wherein said objective function is expressed as: Wherein, the Indicating the number of distance blocks, Represent the first Target distance unit in each distance block The corresponding sampled signal is used to determine the signal, Represent the first Target distance units in distance blocks Is used for the radar tilt distance of the (a), Represents the radar tilt corresponding to the reference range bin, Representing the length of the known azimuth sequence used for signal recovery, Representing target distance units Is used for the target coefficient of restitution of (c), Represent the first In distance blocks and missing signals Distance from each other The sampling signals corresponding to the distance units of the azimuth sampling interval, The sampling time interval representing the azimuth direction, And Representing the estimated error compensation parameters to be solved.
7. The method for recovering a periodic missing SAR signal based on a priori knowledge of echo structures according to claim 6, wherein the missing signal in the original echo signal is expressed as: Wherein, the Representing a missing signal in the original echo signal, Representation and absence signal Distance from each other The sampling signals corresponding to the distance units of the azimuth sampling interval, Representing the azimuth to target recovery factor of the range bin, And Representing the target estimation error compensation parameter.
8. The method for recovering a SAR periodic missing signal based on a priori knowledge of echo structures according to claim 3, wherein the target recovery coefficients of the distance blocks where each of the target distance units is located are expressed as: Wherein, the Represent the first The target recovery coefficients for the distance blocks, Represent the first Target distance units in distance blocks Is used for the radar tilt distance of the (a), Represents the radar tilt corresponding to the reference range bin, Representing target distance units Is used for preprocessing recovery coefficients.

Description

SAR periodic missing signal recovery method based on echo structure priori knowledge Technical Field The invention belongs to the technical field of radar signal processing, and particularly relates to an SAR periodic missing signal recovery method based on priori knowledge of an echo structure. Background In an ideal case, the imaging echo of the synthetic aperture radar (SYNTHETIC APERTURE RADAR, SAR) is obtained by periodically and uniformly sampling the radar along a preset track in a distance-azimuth two-dimensional mode, however, due to the influence of various factors in reality, such as single radar multi-task switching, time-frequency synchronous signal transmission, same-frequency-band signal serial reception and the like, the situation that part of the echo is periodically lost or unavailable may occur, and the recorded signal is in a non-ideal state at the moment. If the non-ideal signal is simply subjected to zero setting, the coherence of the uniform sampling signal is destroyed, so that a large amount of high-energy grating lobe ghosts appear in the direction of the missing signal, original scene information is covered, and the visibility and the interpretation of the SAR image are affected. In an actual application scenario, if the quality of the obtained SAR image is not damaged, a fast and effective missing signal recovery algorithm is needed to restore the coherence of the signal. Because the azimuth sampling time is far longer than the distance sampling time, the azimuth missing signal recovery problem is mainly considered. In recent years, due to high performance and high robustness, the compressed sensing algorithm can still robustly reconstruct signals under the conditions of poor signal-to-noise ratio and high signal loss rate, so that the compressed sensing algorithm becomes a new research hot spot, and mainly comprises the steps of carrying out coefficient representation on signals to be reconstructed, constructing an observation matrix, and completing reconstruction on the signals by utilizing a proper algorithm, wherein an OMP algorithm and a st-OMP algorithm for reducing iteration times in a segmentation mode are typical. However, although these algorithms have excellent recovery effects, the cost is that the estimation matrix needs iterative optimization, the number of times is variable, the operation amount is huge, and the algorithm can be applied only at the post-processing end without considering the time consumption of the algorithm. In addition to compressed sensing, the mainstream algorithms can be divided into two modes, interpolation and spectral estimation. Interpolation algorithms are often used to do the equivalent homogenization of non-uniformly sampled data. Representative examples of the interpolation include sinc interpolation, cubic interpolation, linear optimal unbiased interpolation, and the like. Interpolation-like algorithms typically determine interpolation weights by relative positional relationships between data or optimization objectives based on interpolation accuracy, and complement the region to be interpolated by a combination of peripheral regions and interpolation weights. Interpolation algorithms can usually achieve good accuracy when the signal meets the nyquist sampling theorem, however, in order to reduce data redundancy, the SAR azimuth sampling rate is usually set to be slightly larger than the azimuth bandwidth, so in a local area lacking data, because the nyquist sampling rate is no longer met, the low-order algorithm such as interpolation is not enough to acquire information. The spectral estimation class algorithm typically performs spectral parameter estimation based on the non-missing part signal, and directs the recovery of the signal based on information obtained by the spectral estimation. The Burg algorithm is a stable auto-regression algorithm, and obtains a linear prediction coefficient of a signal based on minimizing power spectrum average power errors of forward and backward predictions of each order. The GAPES algorithm additionally reserves phase information by estimating complex amplitude, so that the method has higher estimation accuracy, and interpolation recovery is carried out on missing data by a least square method. The MIAA algorithm also estimates the complex amplitude of the signal, calculates the optimal weight for each frequency point to be estimated, and iterates and alternately performs the spectrum estimation and signal recovery processes until the recovery effect reaches an ideal value. These spectral estimation algorithms require sequential estimation from missing locations one by one, and the total computation remains too great for the number of range-azimuth samples of the SAR echo. In summary, there are many existing algorithms for processing the problem of data missing, but the algorithms effective for low-oversampling rate data are limited by high operand, and a fast and effective missing signal recovery