Phase correction system for automatic focusing of synthetic aperture radar

• US 4,924,229 A
• Filed: 09/14/1989
• Issued: 05/08/1990
• Est. Priority Date: 09/14/1989
• Status: Expired due to Fees

• Current Assignee: Sandia Corporation (Martin Marietta Corporation)
• Original Assignee: United States Department of Energy (Government of the United States of America)
• Inventors: Jakowatz, Charles V. Jr., Eichel, Paul H., Ghiglia, Dennis C.
• Primary Examiner(s): Tarcza, Thomas H.
• Assistant Examiner(s): Hellner, Mark
• Application Number:
  US07/407,088
• Time in Patent Office: 236 Days
• Field of Search: 342/25, 342/191, 342/175
• US Class Current: 342/25.00E
• CPC Class Codes: G01S 13/9017 with time domain processing of the SAR signals in azimuth G01S13/9005 takes precedence
• CPC Class Codes: G01S 13/9019 Auto-focussing of the SAR signals

Abstract

A phase gradient autofocus system for use in synthetic aperture imaging accurately compensates for arbitrary phase errors in each imaged frame by locating highlighted areas and determining the phase disturbance or image spread associated with each of these highlight areas. An estimate of the image spread for each highlighted area in a line in the case of one dimensional processing or in a sector, in the case of two-dimensional processing, is determined. The phase error is determined using phase gradient processing. The phase error is then removed from the uncorrected image and the process is iteratively performed to substantially eliminate phase errors which can degrade the image.

Claims

1. A synthetic aperture radar comprising:

a directional radar antenna directed at a target area to be imaged and displaying relative movement with respect thereto;

means for supplying radar interrogation pulses to said antenna;

image formation processing means for receiving position indicating reflections from said target area and comparing said position indicating reflections with said interrogation pulses produced by said means for supplying, said image formation processing means using multiple samples of said position indicating reflections to develop target range and azimuth data representative of a target image;

at least one of said target range and azimuth data exhibiting arbitrary phase error between said samples, resulting in smearing of the target image; and

phase correction autofocus means for estimating and subtracting said phase error from said target image, said phase correction autofocus means including,

autofocus processor means for,

(a) dividing said target image into a plurality of target image portions,

(b) locating the strongest highlight in each of said target image portions,

(c) estimating the width of the smearing of each highlight in each dimension of said phase error for each of said target image portions,

(d) estimating of the phase error in each said dimension thereof from said estimated smear width of said highlights, and

(e) compensating for said phase error in said target image to produce compensated data for said target image.

2. The synthetic aperture radar of claim 1 wherein said radar is aircraft borne,

said phase error being arbitrary phase error causing smear in said target image in said azimuthal direction.

3. The synthetic aperture radar of claim 1 wherein said radar is space borne.

4. The synthetic aperture radar of claim 2 wherein said target image portions divided from said target image by said autofocus processor means are range bins of said target image.

5. The synthetic aperture radar of claim 4 wherein said autofocus processor means further circularly shifts said strongest highlight to a central part of each said target image portion;

said autofocus processor means zeroing the parts of each image portion outside the estimated smear width thereof to reduce extraneous errors in said autofocus processor.

6. The synthetic aperture radar of claim 1 wherein said autofocus processor means iteratively performs said steps b-e;

said autofocus processor means further compares the estimated phase error of each iteration of steps b-e to a threshold, ending the iteration and producing said compensated data for said target image when said estimated phase error is less than said threshold.

8. The synthetic aperture radar of claim 1 wherein said image formation processing means compresses at least on of said target range and azimuth data by performing Fourier transformation to the frequency domain.

9. The synthetic aperture radar of claim 1 wherein both said range and azimuth data developed by said image formation processing means exhibit phase error.

10. The synthetic aperture radar of claim 9 wherein said target image portions divided from said target image by said autofocus processor means are rectangular cells C.

11. The synthetic aperture radar of claim 10 wherein said autofocus processor means further circularly shifts said strongest highlight to a central part of each said target image portion;

said autofocus processor means zeroing the parts of each image portion outside the estimated smear width thereof to reduce extraneous errors by said autofocus processor means.

7. The synthetic aperture radar of claim further comprising display means for displaying said target image from said compensated data.

12. A method of producing an image of a target object from radar signals transmitted and received from a radar antenna exhibiting movement relative to said target comprising:

(a) transmitting radar interrogation signals from said antenna moving relative to said target object;

(b) receiving position indicating reflections from said target object produced from said radar interrogation signals;

(c) comparing said position indicating reflections with said interrogation signals and using multiple samples of said position indicating reflections to develop target range and azimuth data representative of a target image, at least one of said target range and azimuth data exhibiting phase errors between said samples, resulting in smearing of the target image; and

(d) estimating and subtracting said phase error from said at least one of said target range and azimuth data to produce a phase corrected target image by,

(1) dividing said target image into a plurality of target image portions,

(2) locating the strongest highlight in each of said target image portions,

(3) estimating the width of the smearing of each highlight in each dimension of said phase error for each of said target image portions,

(4) estimating of the phase error in each said dimension thereof from said estimated smear width of said highlights, and

(5) compensating for said phase error in said target image to produce compensated data for said target image.

13. The method of claim 12 wherein said radar is aircraft borne,

said phase error being arbitrary phase error causing smear in said target image in said azimuthal direction.

14. The method of claim 13 wherein said target image portions divided from said target image are range bins of said target image.

15. The method of claim 14 wherein said step d of estimating and subtracting further includes circularly shifting said strongest highlight to a central part of each said target image portion, and

zeroing the parts of each image portion outside the estimated smear width of each image frame reduce extraneous errors.

16. The method of claim 12 further comprising:

(e) iteratively performing said sub-steps 2-5 of said step (d) of estimating, and

(f) comparing the estimated phase error of each iteration of sub-steps 2-5 of said step (d) of estimating to a threshold and ending the iteration and producing said compensated data for said target image when the estimated phase error is less than said threshold.

17. The method of claim 12 further compressing displaying said target image from said compensated data.

18. The method of claim 12 wherein both said range and azimuth data developed by said image formation processing means exhibit arbitrary phase error.

19. The method of claim 18 wherein said sub-step (1) of said step (d) of estimating divides said target image into image portions formed as rectangular cells C.

20. A system of correcting phase error in a synthetic aperture imaging system which develops range and azimuth data of individual image frames, at least one of said range and azimuth data exhibiting phase error resulting in smear of the image formed thereby, said system comprising:

input means for receiving said range and azimuth data from said synthetic aperture imaging system;

means for dividing said target image into a plurality of target sub-image portions;

highlight locating means receiving each sub-image portion developed by said means for dividing and locating the strongest highlight therein;

first means for estimating the amount of phase error in each sub-image portion by estimating the degree of smear of the brightest highlight contained therein;

second means for estimating the phase error in said image frame from the aggregate processing of the image error for each sub-image portion; and

correction means for removing the estimated phase error from the target image frame.

21. Then system of claim 20 further comprising error threshold determining means for determining whether said phase error estimated by said second means is less than a predetermined threshold;

said highlight locating means, first means, second means, and correction means iteratively operating until said error threshold determining means determines that phase error correction of that target image is completed.

22. The system of claim 20 further comprising display means for displaying each said target image.

23. The system of claim 21 wherein said phase error is one dimensional arbitrary phase error causing one dimensional smear in said target image;

said means for dividing forming said target sub-image portions from range bins of said target image.

24. The system of claim 23 wherein said highlight locating means further circularly shifts the strongest highlight in each sub-image portion to a central position thereof.

25. The system of claim 21 wherein said phase error is two dimensional phase error;

said means for dividing and forming said target image portions as rectangular cells C.

26. The system of claim 25 wherein said highlight locating means further circularly shifts the strongest highlight in each sub-image portion to a central position thereof.

27. A method of correcting phase errors in a synthetic aperture imaging system which develops range and azimuth data of individual image frames, at least one of said range and azimuth data exhibiting phase error resulting in smear of the image formed thereby, said method comprising:

(a) dividing said range and azimuth data representing a single image frame into a number of data portions defining sub-image portions;

(b) determining the brightest highlight in each sub-image portion;

(c) using the brightest highlight in each sub-image portion to estimate phase error induced image smear for each sub-image portion of said image frame; and

(d) estimating the phase error in said single image frame 15 from the aggregate processing of the estimated image smear of each said sub-image portion; and

(e) removing the estimated phase error from said range and azimuth data defining said single image frame to form a phase error corrected image.

28. The method of claim 27 further comprising:

(f) determining whether said phase error estimated by said step (d) of estimating is less than a predetermined threshold;

(g) iteratively performing said steps (b)-(f) until said step (f) determines that the estimated phase error is less than the predetermined threshold, thereby determining that phase error correction of that said target image is completed.

29. The method of claim 28 further comprising displaying each said target image after phase error correction is completed.

30. The method of claim 29 wherein said phase error is one dimensional arbitrary phase error causing one dimensional smear in said target image;

said step (a) of dividing forming said target sub-image portions from range bins of said target image.

31. The method of claim 30 wherein said step (b) of determining further circularly shifts the brightest highlight in each sub-image portion to a central position thereof.

32. The method of claim 27 wherein said phase error is two dimensional phase error;

said step (a) of dividing and forming said target image portions as rectangular cells C.

33. The method of claim 32 wherein said step (b) of determining further circularly shifts the brightest highlight in each sub-image portion to a central position thereof.