Soft-information for localization and sensing

Abstract

Accurate localization and sensing are essential in modern context-aware applications. Traditional localization relies on deterministic single-value metrics, such as range and angle estimates, with simplified measurement models. However, these approaches often falter in challenging environments with noise, non-line-of-sight conditions, and complex propagation effects.
This seminar introduces the Soft Information (SI)-based approach for localization and sensing, a recent probabilistic framework that integrates data from diverse intra- and inter-node measurements, as well as contextual data, to enhance localization accuracy and robustness. By employing efficient methods for extracting, learning, and fusing SI, this approach outperforms classical techniques in accuracy with different levels of computational complexity. The SI framework represents a key advancement in AI-based localization and sensing, poised to support the next generation of resilient and efficient positioning technologies.

About the Speaker

Stefania Bartoletti is Associate Professor at the Department of Electronic Engineering of the University of Rome, Tor Vergata, where she is also coordinating the ERC Starting Grant FIND-OUT (2023–2028).
Sge received the Ph.D. degree in Engineering Science–Information Engineering and the Laurea degree (Summa Cum Laude) in Electronics and Telecommunications Engineering from the University of Ferrara, in 2015 and 2011, respectively.
From 2019 to 2022 she was a researcher at the Institute of Electronics, Computer and Telecommunication Engineering of the National Research Council of Italy (IEIIT-CNR).  She has been a Marie-Skłodowska Curie Global Fellow (MSCA-IF-GF, Horizon 2020) at the Wireless Information & Network Science Laboratory of the Massachusetts Institute of Technology (MIT) and the University of Ferrara (2016–2019). She was a Visiting Ph.D. Student at the Laboratory for Information & Decision Systems at MIT (2013–2014).
Her research interests include theory and experimentation of statistical signal processing for wireless communication systems, with a particular focus on sensing and localization aspects and vehicular communications.