{Reference Type}: Journal Article
{Title}: A Learned-SVD Approach to the Electromagnetic Inverse Source Problem.
{Author}: Capozzoli A;Catapano I;Cinotti E;Curcio C;Esposito G;Gennarelli G;Liseno A;Ludeno G;Soldovieri F;
{Journal}: Sensors (Basel)
{Volume}: 24
{Issue}: 14
{Year}: 2024 Jul 11
{Factor}: 3.847
{DOI}: 10.3390/s24144496
{Abstract}: We propose an artificial intelligence approach based on deep neural networks to tackle a canonical 2D scalar inverse source problem. The learned singular value decomposition (L-SVD) based on hybrid autoencoding is considered. We compare the reconstruction performance of L-SVD to the Truncated SVD (TSVD) regularized inversion, which is a canonical regularization scheme, to solve an ill-posed linear inverse problem. Numerical tests referring to far-field acquisitions show that L-SVD provides, with proper training on a well-organized dataset, superior performance in terms of reconstruction errors as compared to TSVD, allowing for the retrieval of faster spatial variations of the source. Indeed, L-SVD accommodates a priori information on the set of relevant unknown current distributions. Different from TSVD, which performs linear processing on a linear problem, L-SVD operates non-linearly on the data. A numerical analysis also underlines how the performance of the L-SVD degrades when the unknown source does not match the training dataset.