Detection and Imaging from MSR Measurements

Abstract

The problem addressed in this chapter is to detect and localize point reflectors or small inclusions embedded in a medium from MSR measurements. We use random matrix theory tools and the results of Chap. 6 to study these problems in the presence of measurement noise. The measurement noise can be modeled by an additive complex Gaussian matrix with zero mean. We consider an SVD based detection test. By the Neyman-Pearson lemma we design the most powerful test for a given false alarm rate and provide the probability of detection of a point reflector hidden or not in noise. Then we build algorithms that estimate the number, the location, and the strength of points reflectors embedded in the medium. Using again the results in Chap. 6 we adopt these algorithms for small inclusion detection and localization.