A
radar acquires a formed SAR image of
radar scatterers in an area around a central reference point (CRP). Target(s) are within the area illuminated by the
radar. The area covers
terrain having a plurality of elevations. The radar is on a moving platform, where the moving platform is moving along an actual path. The actual path is displaced from an ideal SAR
image acquisition path. The radar has a computer that divides the digital returns descriptive of the formed SAR image into multiple blocks, such as a first strip and an adjacent strip. The first strip is conveniently chosen, likely to generally align with a part of the area, at a first elevation. An adjacent strip covers a second part of the area at a second elevation. The first strip is overlapping the adjacent strip over an overlap portion. The first and second elevation are extracted from a
terrain elevation
database (DTED). Horizontal displacement of returns (range deviation) is computed for each strip using the elevation information from the
terrain elevation
database. Taylor series coefficients are computed for the horizontal displacement due to
terrain elevation using the ideal path, the actual path and central reference point. Actual flight
path deviation is available at each pulse position while
azimuth frequency is given in
azimuth angle off mid angle point. Remapping between indices in two arrays is also computed.
Phase error compensation and compensation in
azimuth (spacial frequency) is computed using the Taylor series coefficients, a
Fast Fourier Transform and an inverse
Fast Fourier Transform for each strip.
Phase error compensation is applied to the digital returns from each strip to obtain the SAR image. The SAR image is further improved by having the first strip corrected data and the second strip corrected data merged over the overlap portion to generate a relatively seamless SAR image.