Estimating target location from scattered field data (Q5906631)

The paper deals with the problem of estimation of unknown target parameters from monostatic or bistatic scattering returns. The main goal is to obtain a maximum likelihood estimation procedure of target location through Cramér-Rao bounds for the variance of the estimation error. The general problem of estimating the location of known target from a set of noisy observations without referring to a particular geometry is considered. The model of signal includes measurements for different probing. The noise is considered white, Gaussian, unncorrelated. By using frequency domain approach a translation formula and expressions for log likelihood function and the Fisher information matrix are received. As it is mentioned in the paper, the expressions are exact and in particular, are not limited to weak scattering formulations such as the Born or Rytov approximations. On the example of CR bound computing on estimate of location of a semitransparent cylinder from single view and single frequency transmission of reflection data, the significant improvement in performance of the estimator is illustrated. The result is obtained in case when the target is positioned less than one wavelength from the receivers. The paper is addressed to scientists working in the field of radar signal processing and diffraction tomography.