Section: New Software and Platforms

PHYSALIS

Keywords: Source localization - Cosparsity

Scientific Description: PHYSALIS (Physics-Driven Cosparse Analysis) gathers algorithms for (joint) source localization and estimation, expressed as inverse problems and addressed with co-sparse regularization. A particular emphasis is put on the acoustic and EEG settings.

Functional Description: PHYSALIS is distributed as a set of Matlab routines to reproduce experimental results from the Ph.D. thesis of Srdan Kitic.

News Of The Year: In 2017, the code of PHYSALIS has been packaged at the occasion of the writing of an overview chapter on co-sparse source localization.

• Participants: Laurent Albera, Nancy Bertin, Rémi Gribonval and Srdan Kitic

• Contact: Rémi Gribonval

• Publications: Physics-driven inverse problems made tractable with cosparse regularization - Cosparse regularization of physics-driven inverse problems - Versatile and scalable cosparse methods for physics-driven inverse problems - Hearing behind walls: localizing sources in the room next door with cosparsity - Sparse Acoustic Source Localization with Blind Calibration for Unknown Medium Characteristics - The best of both worlds: synthesis-based acceleration for physics-driven cosparse regularization

• URL: http://cosoloc.gforge.inria.fr/