Julie Grollier, a CNRS researcher at the Laboratoire Albert Fert, has been awarded an ERC Proof of Concept (PoC) 2026 grant

To discover

Discover Julie Grollier’s profile by CNRS Physique

Julie Grollier, a CNRS researcher at the Laboratoire Albert Fert (CNRS/Thales/Université Paris-Saclay), has just been awarded an ERC Proof of Concept 2026 grant for her project RF-ImagingNet. The goal: to develop intelligent radiofrequency sensors capable of analysing microwave signals directly at the point of acquisition using spintronic components, without going through a full digitisation process that is costly in terms of energy and computing power

© image fournie par Julie Grollier
The ERC Proof of Concept is a supplementary funding scheme designed to support the initial stages of translating research results into practice, in order to explore their commercial and societal innovation potential. It is open exclusively to researchers who currently hold, or have previously held, an ERC grant, and who are responsible for an ERC Starting, Consolidator, Advanced or Synergy project that is either ongoing or completed within the past year. This first 2026 funding round represents a total of €27.3 million, with each grantee receiving €150,000 over a maximum period of 18 months.
Julie Grollier has been awarded funding for her project RF-ImagingNet – Spintronic RF Neural Front-Ends for Embedded Microwave Sensing. The project explores a spintronic neural interface capable of processing radiofrequency signals directly at the point of acquisition. By moving part of the intelligence upstream of full digitisation, RF-ImagingNet aims to develop embedded sensing systems that are faster, more compact and more energy-efficient.
Sensors using radiofrequency or microwave signals can probe environments that light cannot easily penetrate. However, in current systems, the measured signals are often fully digitised before being analysed, which requires significant energy, memory and computing power. The RF-ImagingNet project proposes moving part of this intelligence closer to the physical signal itself. It aims to develop an RF front-end chip based on spintronic components capable of processing microwave signals directly, before full digitisation takes place. These components act as frequency-sensitive artificial synapses: they weight and combine the information carried by radio waves, in order to rapidly produce simplified, decision-relevant information. The aim is to pave the way for a new generation of embedded sensors that are faster, more compact and less energy-hungry, capable of performing an initial analysis locally, without relying on heavy computing hardware or transmitting large volumes of data.

Plus de news Award

  • News
15 May 2024