| Time reversal techniques have been exploited to achieve super resolution focusing in acoustics, radiotherapy medicine, and telecommunications. In this thesis, we develop analytical models to explain the super resolution focusing achieved with time-reversal, matched field processing. Our analytical model is based on multipaths introduced due to total internal reflections of the transmitted signal within a random medium, which is used to develop robust detection algorithms for the determination of the number and locations of passive scatterers embedded in an unknown medium. Using the analytical multipath model, we prove that the total number of scatterers equals the number of nonzero dominant eigenvalues of the system response matrix. The locations of the scatterers are estimated by extending the MUSIC algorithm to incorporate the analytical multipath model. A setup consisting of two transducer arrays, perpendicularly oriented from each other, is used to accurately estimate the location of the passive scatterers embedded in an unknown medium. (Abstract shortened by UMI.). |
