05953441

a

05953441 Garapon, Pierre Resolution limits in source localization and small inclusion imaging. Ammari, Habib (ed.) et al., Mathematical and statistical methods for imaging. NIMS thematic workshop, Inha University, Incheon, Korea, August 10--13, 2010. Providence, RI: American Mathematical Society (AMS) (ISBN 978-0-8218-5289-7/pbk). Contemporary Mathematics 548, 21-30 (2011). 2011 Providence, RI: American Mathematical Society (AMS) EN inverse problems imaging resolution diffraction multiscale expansions Summary: We explicit the notions of resolution limits arising in the source localization problem. In the simple case of the Helmlholtz equation with homogeneous medium, we show that if the optimal estimator for the location of a source has a unique minimum, it has finite resolution if the field is recorded far away from the source. We observe that this is no longer the case if the data is observed everywhere. We then try to connect this simple formalism in the case of the identification of a small defect in an homogeneous background. Although this is not simple because the problem has a lot more unknowns, we observe the same features of resolution limits, in boundary recordings, and accuracy in the case of internal data.