NEUTRINAI: eNErgy-frUgal internet of Things Networking with generative AI modules

MSCA (Marie Skłodowska-Curie)HORIZON-TMA-MSCA-PF-EFID: 101151067
EC Contribution
€2,149
Consortium Size
1 orgs
Start Year
2024
Summary

This project aims to design energy-efficient communication technology and protocol for the 6th generation (6G) Internet of Things (IoT) communication system, in which energy efficiency is one of the essential Key Performance Indicators (KPIs) to realize sustainable IoT networking. The energy-efficient protocol designs for IoT devices serving 6G applications are more challenging than conventional wireless sensor networks, as we need to consider the energy cost caused by the introduction of the Artificial Intelligence (AI)/ Machine learning (ML) model as well as that of the primary radio circuit. However, most related works do not consider the energy cost of introducing the ML model, which affects the overall energy efficiency of the communication protocols. In order to realize the reduction of the overall energy consumption of AI-empowered IoT devices, this project newly introduces generative AI modules and wake-up radio technology into IoT devices operating with batteries, by which IoT devices can suppress unnecessary data transmission, and communication is taken place in an energy-efficient manner by cooperatively transmitting signaling, informative data, and ML model from devices to devices. This project will design a new communication framework integrating generative AI modules, investigate the system-level performance of G-AI-based communication systems, and clarify the gain brought by the introduction of G-AI modules in terms of energy efficiency and accuracy.

Consortium (1)

Project Results (7)

Source: CORDIS, the EU research results database.

Publications (6)
Coexistence of Push Wireless Access with Pull Communication for Content-based Wake-up Radios
GLOBECOM 2024 - 2024 IEEE Global Communications Conference· 2025DOI
Junya Shiraishi, Sara Cavallero, Shashi Raj Pandey, Fabio Saggese, Petar Popovski
Content-Based Wake-Up for Energy-Efficient and Timely Top- <i>k</i> IoT Sensing Data Retrieval
IEEE Transactions on Communications· 2025DOI
Junya Shiraishi, Anders E. Kalør, Israel Leyva-Mayorga, Federico Chiariotti, Petar Popovski, Hiroyuki Yomo
EcoPull: Sustainable IoT Image Retrieval Empowered by TinyML Models
GLOBECOM 2024 - 2024 IEEE Global Communications Conference· 2025DOI
Mathias Thorsager, Victor Croisfelt, Junya Shiraishi, Petar Popovski
Low-Power and Accurate IoT Monitoring Under Radio Resource Constraint
2025 IEEE 36th International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC)· 2025DOI
Takaho Shimokasa, Hiroyuki Yomo, Federico Chiariotti, Junya Shiraishi, Petar Popovski
Time-Constrained Federated Learning (FL) in Push-Pull IoT Wireless Access
IEEE Networking Letters· 2025DOI
Van Phuc Bui; Junya Shiraishi; Petar Popovski; Shashi Raj Pandey
TinyAirNet: TinyML Model Transmission for Energy-Efficient Image Retrieval From IoT Devices
IEEE Communications Letters· 2024DOI
Junya Shiraishi, Mathias Thorsager, Shashi Raj Pandey, Petar Popovski
Deliverables (1)
Data Management Plan