Communications: directive radio wave systems and devices (e.g. – Synthetic aperture radar
Reexamination Certificate
2003-07-30
2004-08-24
Lobo, Ian J. (Department: 3662)
Communications: directive radio wave systems and devices (e.g.,
Synthetic aperture radar
C342S195000, C342S196000
Reexamination Certificate
active
06781541
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of Invention
This invention is in the field of signal processing of phase correction for focusing images obtained from Synthetic Aperture Radar (SAR).
2. Description of the Related Art
Synthetic aperture radar (SAR) systems image fixed, stationary radar scatterers (targets) at a range on the ground by coherently integrating information from returns derived from multiple radar pulses. SAR renders a high resolution ground map of the radar scatterers (targets) by coherently combining reflected radar return signals acquired with a moving antenna having a significant component of velocity in the cross range direction.
Coherent integration of returns from multiple pulses acquired while moving in the cross range direction is theoretically equivalent to carrying a large antenna having an extent in the cross range direction equivalent to the distance traveled during the pulse acquisition. Coherent integration of returns from pulses, and subsequent formation of focused SAR images requires accurate information on platform position and velocity. The returns from each transmitted pulse need to be aligned, or motion compensated to a high degree to maintain the phase coherent aspect of the SAR imaging process. Uncompensated motion error results in smearing (or de-focusing) of the SAR image. Thus, to obtain better identification of ground objects, improving phase compensation for better image resolution is a main goals in the advanced SAR systems.
Motion compensation based on accelerometers, inertial navigation and typical navigation motion information is typically not sufficient to assure coherent integration for a well focused image. Thus, autofocus functions, depending on a prominent radar scattering target within the image itself, estimate residual phase error from partially processed video phase history (VPH) data. This reduces the residual phase error for a focused image.
For the formation of SAR images using autofocus mechanisms has been used in SAR systems. The autofocus is based on the compensation of platform motion (typically antenna phase center) to a stationary, reference target on ground. One example is the Polar Format Algorithm (PFA). Problems with PFA, however, are geometric distortion and spatially variant de-focusing increasing with distance from the reference target. To limit the de-focusing related to these problems, the prior art divides the imaged area in range for sub-swaths and in azimuth for sub-patches, reducing the distance from the reference target. Separate motion compensation using a reference target is used in each sub-swath and sub-patch. This prior art sub-swath and sub-patch processing, however, does not maintain useful magnitude and phase continuity over all the sub-swaths and sub-patches required of typical interferometric SAR (IFSAR) processing, or coherent change detection (CCD).
Another approach of the prior art to the problems of geometric distortion and limited depth of focus over larger image sizes is the range migration algorithm (RMA). RMA requires motion compensation along a line in the focus plane and allows imaging of a relatively large map without subswath/sub-patch processing. One of the difficulties with the RMA is in the implementation of autofocus function during the batch processing as normally done with the PFA. The difficulty comes from the fact that phase histories from different targets after range compression do not align in time. Because of this alignment difficulty, not observed in the case of PFA, some new systems implement the autofocus function before actual RMA processing starts at the cost of additional image formation processing based on PFA or range-Doppler algorithm. This additional computing burden reduces increases the cost of the SAR imaging processor and/or time delay for image rendition.
Another approach of implementing autofocus function is to use partially processed data by RMA during batch processing with additional processing. This requires shifting of signal support depending on the target location in azimuth. When image is oriented in the direction other than range-azimuth, data skew is additionally required, further posing a computing burden.
Using search mode images formed using RMA, azimuth coverage in a single patch is large compared with that in spot mode, while resolution is lower. Shifting of signal support for a target is proportional to the displacement of targets in azimuth. Thus, data length after shifting signal support is typically too large for practical application with existing memory/computer capability.
SUMMARY OF THE INVENTION
Above limitations are minimized by a method for acquiring synthetic aperture images of stationary targets using a radar system, said radar system located on a moving platform, the platform moving in the along track direction, comprising the steps of:
converting a plurality of radar signals reflected from stationary targets located on a ground patch into digital values;
storing said digital values;
computing an along track Fourier transform of said digital values;
performing matched filter functions;
motion compensating said digital values stored within said memory to a first approximation of motion of said moving platform;
interpolating said digital values to generate interpolated values, generally using a Stolt interpolator;
skewing said interpolated values to reorient distortions arising from said first approximation of motion in a preferred direction thereby generating an image data, said image data descriptive of said stationary targets in a range direction and a cross-track direction;
dividing said image data into a plurality of overlapping sub-patches in, preferably, the azimuth direction, each sub-patch containing a portion of said image data and overlapping data, said overlapping data part of said image data and common between two or more of said overlapping sub-patches;
focusing each of said overlapping sub-patches individually using usual SAR autofocus means to obtain focused sub-patches having a phase using such methods as the phase gradient autofocus algorithm or a multiple discrete autofocus algorithm;
adjusting said phase of each of said focused sub patches to obtain a continuous phase with respect to one or more of said focused sub-patches, thereby obtaining smooth phase transition sub-patches;
extracting said synthetic aperture images of stationary targets after concatenating said smooth phase transition sub-patches.
Skewing said interpolated values is first computed in one direction and subsequently in a second direction, where the first direction is perpendicular to the second direction.
For smoothing purposes across two or more sub-patches, the phase of a each sub-patch is calculated by integrating twice the second derivatives of target phase descriptive of all of said stationary targets contained within each of said focused sub-patches and computing the weighted average of said second derivatives.
Another method for smoothing purposes for computing phase is calculated by removing constant terms and linear terms in said phase using said overlapping data.
REFERENCES:
patent: 5248976 (1993-09-01), Niho et al.
patent: 6018306 (2000-01-01), Serbin
patent: 6492932 (2002-12-01), Jin et al.
patent: 6661369 (2003-12-01), Cho
patent: 6670907 (2003-12-01), Cho
Alkov Leonard A.
Lobo Ian J.
Raytheon Company
LandOfFree
Estimation and correction of phase for focusing search mode... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Estimation and correction of phase for focusing search mode..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Estimation and correction of phase for focusing search mode... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3292713