DC9: Adaptive algorithms for object detection in JCAS

Task: Adaptive algorithms for object detection in JCAS (WP3)

Host institution: UNITN

Country: Italy

Supervisor: Prof. P. Casari [UNITN]

Co-supervisors: Prof F. Granelli [UNITN]; Dr. Mikko Uusitalo [Nokia];

Objectives: To optimise object detection and sensing by minimising the signalling overhead of sensing through intelligent allocation of sensing signals; To exploit multi-agent reinforcement learning techniques to provide adaptive and near-optimal strategies in highly dynamic scenarios; To implement the proposed solutions in COTS sub-6GHz testbed at UNITN for performance evaluation.

Expected Results: Effective object detection and sensing is achieved with low signalling overhead; Proposed reinforcement learning techniques can adapt well in highly dynamic scenarios; Proposed techniques is implemented in sub-6GHz testbed.

PhD enrolment: Doctoral School of the University of Trento

Planned secondments: 

  • TU Delft (3 months, M17-M19): Adaptive object sensing using embedded platforms, with M. Zuniga (KPI: joint journal paper)

  • Nokia-FI (4 months, M24-M27): Effective JCAS scheduling in dense cellular networks, with M. Uusitalo (KPI: joint conference paper)

Candidate profile: Electrical Engineering, Telecommunications, Computer Science

Desirable skills/interests: Wireless networking, 4G/5G protocols, signal/array processing, machine learning, optimization, hands-on experience with hardware and systems (the applicant should be proficient in at least one or two of the skills).


Shortlisted applicants for this position will be prompted to complete the application on the University of Trento’s official web site, in order to comply with formal University rules.