Auto-focusing correction for rotational acceleration effects on inverse synthetic aperture radar images

• US 5,343,204 A
• Filed: 07/29/1993
• Issued: 08/30/1994
• Est. Priority Date: 07/29/1993
• Status: Expired due to Term

• Current Assignee: Lockheed Martin Corporation (Martin Marietta Corporation)
• Original Assignee: Unisys Corporation
• Inventors: Hatlestad, John D., Farmer, Michael E.
• Primary Examiner(s): Hellner, Mark
• Application Number:
  US08/098,917
• Time in Patent Office: 397 Days
• Field of Search: 342/25, 342/195, 342/196
• US Class Current: 342/25.00F
• CPC Class Codes: G01S 13/9019 Auto-focussing of the SAR signals
• CPC Class Codes: G01S 13/904 SAR modes

Abstract

Inverse synthetic aperture radar imaging systems that are used to image targets that undergo rotational accelerations are enhanced by signal processing techniques which provide correction factors to reduce the frequency-shift and frequency-stretch errors that occur due to such accelerations. A target scan is subdivided into a number of sub-apertures and a Fast Fourier Transfer (FFT) is performed on the data for each sub-aperture to provide associated frequency sub-images. The FFT is then subdivided in "sub-swaths" of amplitude versus frequency for a plurality of frequency bands. The change-of-frequency, or frequency shift, data between prominent scattering points of each sub-image are processed to provide a phase correction factor that incorporates range changes and translational acceleration distortions for additional scattering points other than the most prominent scattering point. A frequency-stretch correction factor that compensates for rotational acceleration distortion is also provided.

Claims

1. A method for improving an inverse synthetic aperture radar image by enhancing the radar return image signals that are utilized to form an image of a target that is undergoing rotational acceleration and that may also be undergoing range change and translational acceleration so that range change, translational acceleration and rotational acceleration distortion components that result from the relative target motion may be present in said radar return image signals comprising the steps of:

(a) sampling said radar return image signals at a plurality of initially equal time intervals to provide a plurality of digital sub-image data representations of said radar return image signals wherein said sub-images collectively span said image of said target,

(b) forming a plurality of sub-aperture digital data storage areas for each of said sub-images into which said digital sub-image data representations are respectively collected over said image of said target,

(c) computing a frequency spectrum for each sub-aperture to produce a magnitude versus frequency data plot for each sub-aperture,

(d) subdividing said frequency spectrum of each sub-image into a plurality of magnitude versus frequency plots of sub-swaths which collectively span said associated sub-image,

(e) estimating and storing change-of-frequency data in each sub-swath between selected pairs of sub-apertures,

(f) generating a change-of-frequency versus frequency data plot for all of said selected sub-swaths for each selected pair of sub-apertures,

(g) selecting a plurality of times each of which is a mid-point time for a selected sub-aperture pair,

(h) curve-fitting said change-of-frequency versus frequency data to form a plurality of data plots which are each associated with one of said mid-point times,

(i) measuring the change-of-frequency intercept value for each of said data plots on the change-of frequency axis of said change-of-frequency versus frequency data plot,

(j) measuring the slope value of each of said data plots of the change-of-frequency versus frequency data plot,

(k) providing a polynomial fitting of the polynomial represented by said intercept values versus scan-time,

(l) providing a polynominal fitting of the polynominal representation by said slope values,

(m) twice integrating said intercept polynomials and slope polynomials with respect to time, and

(n) replacing said radar return image signals with image signals that have;

(1) range change and translational acceleration distortion components eliminated by utilizing the values obtained from the twice-integrated change-of-frequency intercept values versus time data plots to develop phase correction signals for said radar return image signals, and

(2) rotational acceleration distortion components eliminated by utilizing the values obtained from the twice-integrated slope values versus time data plot to modify said initial equal sampling time intervals to compensate for frequency-stretch errors of said radar return image signals.

2. A method as claimed in claim 1 comprising the selecting the peak points of said magnitude versus frequency plots in said sub-swaths to provide peak point data which are representative of significant scattering points of said target image according to predefined criteria.

3. A method as claimed in claim 2 comprising the further step of:

weighting said peak-point data of said sub-images as a function of the magnitudes of said peak points.

4. A method as claimed in claim 1 wherein each of said sub-swaths overlap at least one other of said sub-swaths.

5. A method as claimed in claim 4 wherein said sub-swaths are arranged into first and second groups of substantially equal bandwidth sub-swaths and each sub-swath image of said first group is frequency-overlapped by at least one sub-swath of said second group, and each frequency overlap range is approximately equal to one-half of the bandwidth of said sub-swath.

6. A method as claimed in claim 1 wherein there is no time gap between the sub-apertures of at least one sub-aperture pair.

7. A method as claimed in claim 3 wherein each of said sub-swaths overlap at least one other of said sub-swaths.

8. A method as claimed in claim 7 wherein said sub-swaths are arranged into first and second groups of substantially equal bandwidth sub-swaths and each sub-swath image of said first group is frequency-overlapped by at least one sub-swath of said second group, and each frequency overlap range is approximately equal to one-half of the bandwidth of said sub-swath.

9. A method as claimed in claim 8 wherein there is no time gap between the sub-apertures of at least one sub-aperture pair.

10. A method as claimed in claim 9 wherein each of said sub-apertures overlap at least one other of said sub-apertures.

11. A method as claimed in claim 6 wherein each of said sub-apertures overlap at least one other of said sub-apertures.

12. A method as claimed in any of the claims 1-6 or 10-11 wherein said computing step is accomplished by using a Fast Fourier Transformation.