IDENTIFICATION OF DISCONTINUITIES IN BLURRED NOISY SIGNALS

In this paper we address the problem of estimating the locations of di scontinuities in blurred and noisy signals, The discontinuity paramete rs (location and size) are embodied in a model describing the signal r egular and singular components. First, the well-posedness of the estim ation problem, i.e., uniqueness and robustness of the solution, is est ablished. Then dab are processed by a smoothing multiscale resolution technique that, besides providing some noise filtering, enhances the e ffect of the discontinuities. Hence, a maximum-likelihood method is us ed for parameter estimation, Numerical results are obtained bath on si mulated and real data.