Deconvolution Problems for Structured Sparse Signal

This dissertation studies deconvolution problems of how structured sparse signals appear in nature, science and engineering. We discuss about the intrinsic solution to the problem of short-and-sparse deconvolution, how these solutions structured the optimization problem, and how do we design an efficient and practical algorithm base on aforementioned analytical findings. To fully utilized the information of structured sparse signals efficiently, we also propose a sensing method while the sampling acquisition is expansive, and study its sample limit and algorithms for signal recovery with limited samples.