Posts

A new preprint entitled Model-based Deep Learning for High-Dimensional Periodic Structures just appeared on arXiv. It shows that model-based learning using equivalent circuits can be leveraged to make the simulation of frequency selective surfaces (FSS) faster.

A new preprint entitled Model-based learning for location-to-channel mapping just appeared on arXiv. It shows that model-based learning using physical propagation models leads to better results than classical neural network on the mapping linking the location of a given user to its wireless channel.

A new preprint entitled Experimentally realized physical-model-based wave control in metasurface-programmable complex media just appeared on arXiv. This is a joint work with Philipp del Hougne and his team, in which a physical model is used to infer the channel parametrized by RIS configurations from a small subset of measured ones.

A project for which I am the PI has been accepted for funding by the french national research agency (ANR), within the young researcher track (JCJC). The MoBAIWL project (2024-2027, 300k€) will aim at designing efficient data processing methods for future wireless communication systems (6G and beyond), using physical models to structure, initialize and train frugal artificial intelligence methods.