Abstract: A system and a method involve acoustic source localization for multiple acoustic sources using passive sonar. Tracks are constructed based on source location maps (SLMs) that reveal the locations of the sources over a grid of tentative source locations. The proposed estimator for the SLMs performs coherent processing of broadband acoustic measurements, capitalizes on the sparse structure of the SLMs, and uses the prior SLM estimate to capture temporal information about the locations of the sources.

Abstract: A system and method involve acoustic source localization using passive sonar and capitalizing on the sparse nature of a source location map (SLM). Two types of sparsity are exploited, namely sparsity in the support of the SLMs and sparsity in the innovations across consecutive SLMs. The first type is motivated by the desire to construct SLMs whose non-zero entries corresponded to locations where sources are present. The second type of sparsity is motivated by the observation that few changes occur in the support of consecutive SLMs. Per time instant, an SLM is obtained as the solution of a regularized least-squares problem, where the regularization terms are chosen to encourage the desired sparse structures in each SLM and the innovations. Each SLM may be obtained via a specifically-tailored, computationally-efficient proximal gradient algorithm.

Abstract: A system and a method involve acoustic source localization for multiple acoustic sources using passive sonar. Tracks are constructed based on source location maps (SLMs) that reveal the locations of the sources over a grid of tentative source locations. The proposed estimator for the SLMs performs coherent processing of broadband acoustic measurements, capitalizes on the sparse structure of the SLMs, and uses the prior SLM estimate as a means to capture temporal information about the locations of the sources.

Abstract: A system and method involve acoustic source localization using passive sonar and capitalizing on the sparse nature of a source location map (SLM). Two types of sparsity are exploited, namely sparsity in the support of the SLMs and sparsity in the innovations across consecutive SLMs. The first type is motivated by the desire to construct SLMs whose non-zero entries corresponded to locations where sources are present. The second type of sparsity is motivated by the observation that few changes occur in the support of consecutive SLMs. Per time instant, an SLM is obtained as the solution of a regularized least-squares problem, where the regularization terms are chosen to encourage the desired sparse structures in each SLM and the innovations. Each SLM may be obtained via a specifically-tailored, computationally-efficient proximal gradient algorithm.

Abstract: A method involves acoustic source localization by capitalizing on the sparse nature of a source location map (SLM). Sparsity arises naturally since one seeks the location of K sources in a grid of G tentative locations where G>>K. The source localization problem is cast as a regularized LS regression problem with a sparsity constraint whose solution yields the SLM. An iterative solver based on block coordinate descent (BCD) is used. BCD leads to scalar closed-form updates rendering the method's computational complexity per iteration linear with respect to the grid size. The disclosed method enables high resolution location estimation with fewer array measurements than classical matched field processing methods.

Abstract: A method involves acoustic source localization by capitalizing on the sparse nature of a source location map (SLM). Sparsity arises naturally since one seeks the location of K sources in a grid of G tentative locations where G>>K. The source localization problem is cast as a regularized LS regression problem with a sparsity constraint whose solution yields the SLM. An iterative solver based on block coordinate descent (BCD) is used. BCD leads to scalar closed-form updates rendering the method's computational complexity per iteration linear with respect to the grid size. The disclosed method enables high resolution location estimation with fewer array measurements than classical matched field processing methods.