Estimation of object location from diffraction tomographic intensity data

It is shown that an image of the log likelihood function for estimating the location of a known, weakly inhomogeneous object can be generated directly from noisy intensity distributions measured in a set of scattering experiments. This image is a close approximation to the exact log likelihood function generated from complete data (i.e., from both the magnitude and phase of the transmitted wavefields) as long as the measurements are performed sufficiently far from the object. The performance of the procedure is measured through calculation of the Cramer-Rao bound for the estimation error and is illustrated by a computer simulation study.>