Topic of the workshop

Ultra-low power wireless sensor networks are at the heart of the next wave of smart applications that will benefit from omnipresent, ambient technologies and networks for joined sensing and communication. Yet these wireless devices are heavily resource-constrained in terms of energy, memory, and compute. Wirelessly transmitting raw sensor data to the cloud for processing is often impractical due to power, latency, bandwidth, and privacy limitations.

Embedded AI offers a compelling alternative: by processing data locally at the sensor node, systems can reduce communication overhead, extend battery lifetime, and react in real time while protecting sensitive information.

In this workshop, we will present hardware and software technology innovations for bringing intelligence directly onto constrained wireless devices, as well as best practices for design trade-offs and optimization techniques that enable practical deployment. Through live demos our presenters will show how embedded AI can turn tiny, power-limited nodes into smart building blocks of the Artificial Intelligence of Things (AIoT).

This workshop is co-organised with EWSN, the International Conference on Embedded Wireless Systems and Networks, and is accessible to all enthusiasts in wireless technologies and embedded AI.

The programme includes presentations, (live) demos and networking opportunities.

Programma

12:00Registration & lunch
13:20Introduction
Kris Hermus, Coordinator Wireless Community & Innovation Program Manager Flanders, imec
 PART I – GENERAL PRINCIPLES
13:30Enabling AI at the Extreme Edge
Marian Verhelst, Professor, KU Leuven – ESAT
14:30COFFEE BREAK
 PART II – EMBEDDED AI
14:50TinyKubeML: Orchestrating TinyML Models on Far-Edge Clusters
João Oliveira, Fernando Rego, Filipe Sousa (Fraunhofer AICOS); Luis Almeida (CISTER / FEUP – University of Porto)
15:10TinyML as a Service on Multi-Tenant Microcontrollers
Bastien Buil (Cnam / Orange); Chrystel Gaber (Orange); Samia Bouzefrane (Cnam); Emmanuel Baccelli (Inria)
15:30PEARL: Power- and Energy-Aware Multicore Intermittent Computing
Khakim Akhunov (imec); Eren Yildiz (Georgia Institute of Technology); Kasim Sinan Yildirim (University of Trento); Khakim Akhunov (University of Trento)
15:50COFFEE BREAK
 PART III – DEMO TOUR
16:10Guided Demo Tour
 
  • Demo 1 – NXP Semiconductors:
    embedded AI for audio applications
  • Demo 2 – VersaSense:
    edge intelligence in practice: applications of condition monitoring in smart industries
  • Demo 3 – Forcebit:
    high-throughput wireless sensing for linear & rotary drivetrains
  • Demo 4 – Qorvo
    battery-less wireless sensor with Matter and Thread
  • More demos will be announced soon
    ...
 PART IV – INDUSTRY PERSPECTIVE
17:00Enabling AI on edge devices: insights from audio applications
Bruno Defraene, R&D Engineer – Machine Learning and Signal Processing, NXP Semiconductors
17:30A tutorial on the state-of-the-art in (batteryless) wireless sensor networking (with Thread and Matter)
Brecht Van Cauwenberghe, Software engineer, Qorvo
18:00Plenary Q&A session
18:10Networking reception
19:30End of this workshop

Participation only possible after registration.
Please confirm your attendance by completing the

Online registration form