CN112799062A

CN112799062A - A high-resolution wide swath SAR motion compensation method based on prior information

Info

Publication number: CN112799062A
Application number: CN202110395192.0A
Authority: CN
Inventors: 陈溅来; 熊毅; 金养昊; 梁步阁; 杨德贵
Original assignee: Central South University
Current assignee: Central South University
Priority date: 2021-04-13
Filing date: 2021-04-13
Publication date: 2021-05-14

Abstract

The invention discloses a high-resolution wide swath SAR motion compensation method based on a priori information, the steps of which include: step S1: performing distance segmentation to ensure that the distance in each distance block does not change vacantly, which is used for subsequent RD Estimation of phase error; Step S2: Use PGA to estimate the phase error of each distance block, and perform azimuth pulse pressure; Step S3: If the phase error is correctly estimated, the output parameter P is 1, otherwise, P is 0. The invention has the advantages of simple principle, capable of improving estimation accuracy, and expressing the RD phase error curve more completely and accurately.

Description

High-resolution wide swath SAR motion compensation method based on prior information

Technical Field

The invention mainly relates to the technical field of radars, in particular to a high-resolution wide swath SAR motion compensation method based on prior information.

Background

Due to the influence of various factors such as atmospheric motion, the motion trajectory of an airborne Synthetic Aperture Radar (SAR) is changed from an ideal straight-line trajectory to a nonlinear trajectory, and during SAR imaging, motion compensation (MOCO) is often required to compensate the nonlinear trajectory to a linear trajectory. While the RD phase error is a factor that must be considered for motion compensation of high-resolution and wide-band airborne SAR, since the PGA estimation is based on strong scattering points, a situation of erroneous phase error estimation may occur for some observation scenarios without strong scattering points.

In the conventional scheme, after phase error data is obtained, a phase error curve is generally estimated by using a fitting method, which includes a least square method (hereinafter, referred to as LS) and a weighted least square method (hereinafter, referred to as WLS). But the estimation curve has a certain deviation due to the existence of wrong estimation points. The existing methods for estimating the RD phase error comprise LS and WLS, because of the existence of wrong phase error estimation points, a large estimation error occurs when the LS is used, a phase error curve cannot be accurately represented, although the WLS passes through weighting processing, an estimation error which cannot be ignored still exists, and the two methods cannot automatically identify and remove the wrong phase error estimation points, so that the estimation accuracy is poor. In summary, because there are wrong phase error estimation points, both LS and WLS have no function of self-detecting and eliminating the wrong estimation points, firstly, a single use of LS will result in a large estimation error when estimating RD phase error, and secondly, while WLS can reduce the error through weighting processing, but cannot completely eliminate the influence of the wrong phase error, and also will result in a non-negligible estimation error.

Disclosure of Invention

The technical problem to be solved by the invention is as follows: aiming at the technical problems in the prior art, the invention provides the high-resolution wide swath SAR motion compensation method based on the prior information, which has a simple principle, can improve the estimation precision and more completely and accurately express the RD phase error curve.

In order to solve the technical problems, the invention adopts the following technical scheme:

a high-resolution wide swath SAR motion compensation method based on prior information comprises the following steps:

step S1: distance partitioning is carried out, the distance in each distance block is guaranteed not to be changed in a space mode, and the distance is used for subsequent estimation of RD phase errors;

step S2: estimating the phase error of each distance block by using PGA (programmable Gate array), and carrying out azimuth pulse pressure;

step S3: the output parameter P is 1 if the phase error is estimated correctly, otherwise P is 0.

As a further improvement of the process of the invention: the elimination of the estimated points of possible error in the estimated phase error in step S2 includes:

step S201: each distance block is subjected to two imaging processes, namely imaging after PGA estimation and compensation and imaging without PGA estimation and compensation;

step S202: the imaging quality is expressed by specific numerical quantization, and the entropy value of imaging is selected as the standard for judging the imaging quality; setting the entropy value of the imaging image estimated and compensated by using PGA as E1, and the entropy value of the imaging image estimated and compensated by using PGA as E2;

step S203: comparing the image entropy values of the two cases in step S202, if E1> E2, it indicates that the imaging quality estimated by PGA is better, i.e. the phase error of the range block is correctly estimated, otherwise, the phase error of the range block is incorrectly estimated.

As a further improvement of the process of the invention: the higher the entropy value of the image, the worse the imaging quality.

As a further improvement of the process of the invention: estimating an error phase point from the variance in step S2, including: calculating the variance of the estimated phases of the adjacent distance blocks, and if the variance of a certain distance block and the variance of two adjacent distance blocks exceed the set threshold value

Then the phase of this distance block is estimated incorrectly.

As a further improvement of the process of the invention: estimating an error phase point according to a threshold in step S2, including: setting a maximum threshold for phase estimation, the estimated phase exceeding the threshold

Then the phase of the distance block is deemed to be erroneously estimated.

As a further improvement of the process of the invention: the distance blocking in step S1 means: a certain distance range is evenly divided into a plurality of distance sections.

As a further improvement of the process of the invention: in step S3, when all the distance blocks are completely estimated, the WLS is used to estimate the RD phase error, and the wrong estimation point is affected by the parameter P and filtered out, so that only the correct estimation point is left, thereby obtaining an accurate RD phase error estimation curve.

As a further improvement of the process of the invention: before the step S1, data with RCM residues eliminated is obtained.

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

the invention discloses a high-resolution wide swath SAR motion compensation method based on prior information, which is developed around how to improve the estimation accuracy of RD phase errors. The method can effectively solve the influence of insufficient RD error estimation precision caused by the existence of error phase errors.

Drawings

FIG. 1 is a schematic flow diagram of the process of the present invention.

Fig. 2 is a schematic diagram of phase errors estimated by different methods in a specific application example of the present invention.

Detailed Description

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

For ease of understanding, the present invention makes clear the important key terms and standard names mentioned below. PGA: a phase gradient autofocus algorithm. The algorithm can be used to estimate and compensate for low order, high order, and random phase errors. LS: the least squares method. WLS: a weighted least squares method. PI-WLS: weighted least squares method based on prior information (method of the invention). RD: the distances are related. RCM: and (4) moving distance. The change of the distance between the airborne synthetic aperture radar and the target due to the relative motion between the two is called range migration.

As shown in fig. 1, the high resolution wide swath SAR motion compensation method based on prior information of the present invention includes the steps of:

step S1: distance partitioning is carried out; in order to estimate the RD phase error, distance blocking must be performed to ensure that the distance in each distance block does not change spatially, so as to be used for subsequent RD phase error estimation;

In a specific application example, the distance blocking in step S1 means: a certain distance range is evenly divided into a plurality of distance sections. For example, the range of 0-200m is divided into two blocks, the first block is 0-100m, and the second block is 100-200 m. The distance blocking is carried out according to actual needs, and the number and the size of the blocks are selected according to the actual needs.

In a specific application example, in the phase error estimated in step S2, there may be a wrong estimation point, and the present invention further adopts the following procedure to eliminate the wrong estimation point:

determining a criterion: if the estimated point is estimated correctly, the imaging quality after estimation and compensation using the PGA should be better than that without the PGA, but if the estimated point is estimated incorrectly, the imaging quality with the PGA is worse than that without the PGA due to the incorrect estimation point.

step S202: the imaging quality is expressed by specific numerical quantification, and the entropy value of imaging is selected as a standard for judging the imaging quality. Note that: the higher the entropy value of the image, the worse the imaging quality. The entropy value of the imaging image estimated and compensated by the PGA is set as E1, the entropy value of the imaging image estimated and compensated by the PGA is set as E2;

Compared with the traditional method of directly estimating the phase error by using LS or WLS, the method has the advantages that a detection function is generated in advance, namely whether the phase error estimation point is correct or not is judged, the wrong estimation point can be effectively filtered, the estimation precision is improved, and the RD phase error curve is expressed more completely and accurately.

In another embodiment, further, the present invention can also estimate the error phase point according to the variance, as follows:

according to the variance: calculating the variance of the estimated phases of the adjacent distance blocks, and if the variance of a certain distance block and the variance of two adjacent distance blocks exceed the set threshold value

Then the phase of this distance block is estimated incorrectly. Here, the problem is illustrated by an example: a, B, C are three adjacent distance blocks. The invention can calculate the variance of two adjacent distance blocks, if the B point is estimated by error, the variance of two adjacent distance blocks can be calculated

And

are all greater than

And if not, the step (B),

and

are all less than

。

In another embodiment, further, the present invention may also estimate the error phase point according to a threshold, specifically as follows:

setting a maximum threshold for phase estimation, the estimated phase exceeding the threshold

Then the phase of the distance block is deemed to be erroneously estimated.

In a specific application example, in step S3, further, when all the distance blocks are completely estimated, the WLS may be used to estimate the RD phase error, and the wrong estimation point may be filtered out due to the influence of the parameter P, and only the correct estimation point is left, so as to obtain an accurate RD phase error estimation curve.

In a specific application example, before step S1, the residual RCM data may be further eliminated.

Referring to fig. 2, a schematic diagram of curves for estimating phase errors by using various methods is shown, it is apparent from the diagram that phase estimation errors at abscissa 5 and abscissa 10 are erroneous, and LS and WLS are both affected by erroneous estimation points to some extent.

Therefore, by adopting the scheme of the invention, the estimated phase errors of the distance blocks can be detected to be wrong, wrong phase error estimation points are removed when the phase errors are estimated, and only correct phase error estimation points are reserved, so that the precision of the estimated curve can be greatly improved, and the RD phase error curve can be completely and accurately represented.

The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above-mentioned embodiments, and all technical solutions belonging to the idea of the present invention belong to the protection scope of the present invention. It should be noted that modifications and embellishments within the scope of the invention may be made by those skilled in the art without departing from the principle of the invention.

Claims

1. a high-resolution wide swath SAR motion compensation method based on prior information, is characterized in that, step comprises:

Step S1: carry out distance segmentation to ensure that the distance in each distance block does not change, and is used for subsequent estimation of the RD phase error;

Step S2: use PGA to estimate the phase error of each distance block, and perform azimuth pulse pressure;

Step S3: if the phase error is correctly estimated, the output parameter P is 1, otherwise, P is 0;

In step S2, the error phase point is estimated according to the variance, including: calculating the variance of the estimated phases of the adjacent distance blocks, if the variance of a certain distance block and the two adjacent distance blocks exceeds the set threshold

, then the phase of this distance block is erroneously estimated; or, in step S2, the erroneous phase point is estimated according to the threshold, including: setting a maximum threshold for phase estimation, and the estimated phase exceeds the threshold

, the phase of the range block is considered to be misestimated.

2. the high-resolution wide swath SAR motion compensation method based on prior information according to claim 1, is characterized in that, in step S2, in the estimated phase error, the estimated point that may have error is deleted, including:

Step S201 : perform two imaging processes on each distance block, one is imaging after estimation and compensation using PGA, and the other is imaging without PGA estimation and compensation;

Step S202: Quantify the imaging quality with a specific numerical value, and select the entropy value of the imaging as the standard for judging the quality of the imaging; set the entropy value of the imaging image without PGA estimation and compensation to E1, and use PGA estimation and compensation. The entropy value of the imaging image is E2;

Step S203: Compare the image entropy values of the two cases in step S202, if E1>E2, it means that the image quality estimated by the PGA is better, that is, the phase error of the distance block is correctly estimated, otherwise, the phase error of the distance block is estimated. Errors are misestimated.

3 . The high-resolution wide swath SAR motion compensation method based on prior information according to claim 2 , wherein the higher the entropy value of the image, the worse the imaging quality. 4 .

4. The high-resolution wide swath SAR motion compensation method based on a priori information according to any one of claims 1-3, is characterized in that, carrying out distance block in described step S1 refers to: a certain distance range Evenly divided into several distance segments.

5. The high-resolution wide swath SAR motion compensation method based on a priori information according to any one of claims 1-3, characterized in that, in the step S3, when all the distance blocks are fully estimated, use WLS estimates the RD phase error, and the wrong estimated points will be filtered out due to the influence of the parameter P, leaving only the correct estimated points, so as to obtain an accurate RD phase error estimation curve.

6 . The high-resolution wide swath SAR motion compensation method based on a priori information according to any one of claims 1 to 3 , wherein, before the step S1 , the data after the RCM residuals are first eliminated are obtained. 7 .