The rapid proliferation of Internet of Things (IoT) devices has unlocked transformative potential in healthcare, sports, industrial monitoring, and human-computer interaction. Despite significant advances, the practical deployment of wearable IoT systems faces key challenges related to energy eƯiciency, real-time processing, heterogeneous sensor integration, and scalable AI implementation. These issues are compounded by the need for lightweight, low-power solutions capable of delivering robust performance in resource-constrained environments.


Edge AI represents a paradigm shift, by enabling on-device data processing and decision-making. This in turn as potential to reduce latency, conserve bandwidth, and enhance privacy. This session explores cutting-edge research in Edge AI for low-power IoT systems, emphasising fundamental research challenges, algorithmic innovation, and system-level optimization.

Contributions can be application-agnostic and are invited on the following themes:

  • EƯicient Edge Learning Architectures: Development of novel optimised approaches for deployment of models onto low-power devices, including federated learning, knowledge distillation, etc.
  • Sensor Fusion and Multimodal Analysis: Research on integrating heterogeneous sensors for robust data interpretation and advanced context-awareness.
  • Adaptive AI for Edge Devices: Exploration of dynamic reconfiguration techniques that balance energy, accuracy, and computational load in real-time applications.
  • Edge AI Methodologies: Novel approaches in human activity recognition, time series analysis, physics-informed AI, and real-time edge analytics for resource-constrained
    devices.
  • Energy-Aware Hardware-Software Co-Design: Collaborative strategies for creating ultra-low-power devices with tailored AI accelerators and optimized communication protocols.
  • Scalability and Generalisation: Techniques for deploying IoT devices across diverse
    scenarios while maintaining reliability and interpretability.

This session aims bridge theoretical innovation with practical implementation across multiple domains, fostering collaborations that advance the state of Edge AI for wearable and IoT devices.