LENA (non-LinEar sigNal processing for solving data challenges in Astrophysics) is a research project as well as a team financed by the ERC Starting Grant program, with the aim of developing methods for solving inverse problems in signal and image processing. LENA researchers explore the uses of sparse signal modelling, proximal algorithms and machine learning; extending these methods to the non-linear world as well as astrophysical applications. For real world signal processing applications, these new methods are required to deal with corrupted data and outliers to the expected model. A wide range of astrophysical problems can also be targeted, such as component separation problems for the cosmic microwave background or epoch of reionisation signal (disentangling the cosmological signal from Galactic emissions), the estimation of the weak lensing effect from the measurement of galaxy shapes (highly sensitive to instrumental effects and noise) and feature-learning from Galaxy spectral energy distributions.
More information about this project can be find at this location.