MoBAIWL in a glimpse

The MoBAIWL project (Model-Based frugal AI for efficient WireLess communication systems) aims to design efficient data processing methods for future wireless communication systems (6G and beyond), using physical models to structure, initialize and train frugal artificial intelligence methods.

Objective

Wireless communication systems constantly evolve towards more sophistication in order to meet ever-growing efficiency requirements. This evolution entails more complex data processing, which can be handled via classical signal processing techniques or the more recent machine learning approaches. Signal processing tends to be computationally efficient but can rely on simplistic analytic models, while machine learning is data-adaptive by nature but requires heavy computations to be trained. MoBAIWL aims to take the best of both worlds by designing computationally efficient AI methods built on models usually used in signal processing.

Organization

The project will run from 2024 to 2027, with funding of €300k from the French National Research Agency (ANR). This funding will mainly be used to recruit graduate researchers.

The project relies on a team of permanent researchers whose skills are complementary:

• Luc Le Magoarou (scientific coordinator), INSA Rennes
• Matthieu Crussière, INSA Rennes
• Philipp del Hougne, CNRS
• Clément Elvira, Centralesupélec
• Cédric Herzet, ENSAI
• Philippe Mary, INSA Rennes

Publications

José Miguel Mateos-Ramos, Christian Häger, Musa Furkan Keskin, Luc Le Magoarou, Henk Wymeersch (2024). Unsupervised Learning for Gain-Phase Impairment Calibration in ISAC Systems. ICASSP 2025

Arad Gast, Luc Le Magoarou, Nir Shlezinger (2024). DCD-MUSIC: Deep-Learning-Aided Cascaded Differentiable MUSIC Algorithm for Near-Field Localization of Multiple Sources. ICASSP 2025

Cheima Hammami, Lucas Polo-López, Luc Le Magoarou (2024). Efficient Frequency Selective Surface Analysis via End-to-End Model-Based Learning. EuCAP 2025

Job offers

2025

• 6 months internship starting early 2025 (leading to PhD): Description
• 6 months internship starting early 2025: Description

2024

• Filled PhD position starting fall 2024: Description
• Filled 6 months internship starting early 2024 (leading to PhD): Description
• Filled 6 months internship starting early 2024: Description