Signal processing for image reconstruction in synthetic aperture radar (SAR) historically has been based on Fourier transform techniques. One reason for this is the fact that, at the time when SAR was invented in the early 1950's and for some time after that, the only way to process the huge amounts of data in a reasonably expeditious manner was to use optical techniques, such processors being based, at least in part, on Fourier optical principles. With the advent of digital processing in recent years, the existence of efficient algorithms for the computation of the discrete Fourier transform has continued to offer compelling reasons to use Fourier-type inversion methods. Additionally, geometrically-related simplifications in most analyses have engendered the assumption of plane waves being present over the ground patch being imaged, again encouraging the use of Fourier techniques. In applications where the distance of the radar from the ground patch is very large compared to the size of the ground patch to be imaged, such processing is appropriate, though still an approximation. In other cases, the wavefront curvature cannot be ignored, and other steps must be taken in order to yield high-quality imagery. Recent investigations into the use of convolution-backprojection algorithms modified from computer-aided tomography have proved fruitful in correcting for wavefront curvature in monostatic SAR. This paper reports on similar success in bistatic SAR. There appear to be other applications that could benefit from other adaptations of the convolution-backprojection idea.