IoT

We design lightweight AI models that empower IoT (Internet of Things) and edge devices to operate efficiently, portably, and with extended battery life. Optimized for energy consumption and resource optimization, these models process data locally, minimizing reliance on cloud infrastructure and reducing latency.

To ensure privacy and scalability, the framework integrates federated learning (FL) , enabling devices to learn collaboratively without sharing raw data. This decentralized approach preserves security and accuracy compared to classical centralized or isolated learning approaches.

Key features include:

• Lightweight models for devices with limited resources
• Adaptive optimization balancing performance and energy efficiency
• Secure connectivity for seamless integration into larger networks

Applications span personalized healthcare and predictive maintenance, demonstrating how compact AI, FL, and energy-aware design unlock intelligent, autonomous, and sustainable connected environments.