DC04

Description

The DC will pioneer the integration of federated and distributed learning frameworks over dense wireless network targeting RF sensing applications. You'll develop novel breakthrough technologies that combine distributed learning tools with sensing and dense communication networks to enhance remote healthcare technologies and patient monitoring. This position offers the opportunity to drive innovation in AI and machine learning for healthcare.

Collaborate with leading researchers to develop cutting-edge federated, distributed, and generative AI technologies for remote healthcare. Benefit from MSCA’s prestigious funding, global mobility, and professional development opportunities.

Consiglio Nazionale delle Ricerche (CNR) is the largest Italian public research organisation, nationwide distributed in a network of 100+ Institutes, cross-disciplinary fostering science and innovation.The Institute of Electronics, Computer and Telecommunication Engineering (IEIIT) participating in SMARTTEST DN action carries out advanced scientific and technological research in distributed signal processing, radar technology and radio sensing/vision advanced applications. IEIIT is currently offering a PhD position with a focus on federated and distributed learning tools in integrated sensing and communication. Position is funded by the prestigious Horizon Europe Marie Skłodowska-Curie Doctoral Networks 2024 initiative.

Host institution

CNR

Country

(IT)

Supervisor

Dr. Stefano Savazzi & Dr. Sanaz Kianoush (CNR, IT)

Co-supervisors

Prof. Monica Nicoli (POLIMI, IT), Prof. Sofie Pollin (KUL, BE)

Objectives

To design and develop advanced distributed learning mechanisms (i.e., Bayesian learning, federated learning (FL), self-supervised) for accurate and efficient multi-target mapping, localization and tracking in indoor scenarios, collaborative sensing and association of node/link, band, time. The  effectiveness/efficiency of the proposed approaches will be demonstrated in mixed static/moving persons. The DC candidate will explore novel Bayesiasignal processing approaches as well as recent advances in physics-driven generative AI tools designed to reconstruct the effects of body motions on wireless signals. These tools  have recently opened new perspectives for the development of novel RF sensing systems. 

Expected Results

Documented design and verification of efficient data-driven learning (AI) localization and mapping algorithms.

PhD enrolment

POLIMI (IT)

Planned secondments

KUL, BE (M18-20, mentor & co-supervisor: Prof. Sofie Pollin): train DC04 on localization using coms system and massive MIMO testbed.

RDI, NL (M31-33, mentor: Loek Colussi): train DC04 on over-the-air radio experimentation

Candidate profile

The applicant should hold a MSc degree in telecomunication engineering, or/and computer science, or/and automation and control engineeting or/and electronics or/and engineering physics. The applicant should also have a background in wireless communication, machine learning and signal processing. International publications will be considered favorably.

Desirable skills and interests

machine learning, signal processing, wireless networks