Published Paper Details:

The rapid growth of smart home Internet of Things (IoT) devices has increased the demand for intelligent, real-time data processing while operating under strict battery constraints. Edge Artificial Intelligence (Edge AI) enables local data processing on IoT devices, reducing latency and dependency on cloud infrastructures. However, continuous AI inference can significantly drain battery resources, limiting device lifetime and reliability. This paper proposes a Battery-Aware Edge AI framework for Energy-efficient smart home IoT devices. The framework dynamically adapts AI workloads based on a battery state, device usage patterns, and environmental context. By integrating battery-aware scheduling, lightweight machine learning models, and adaptive inference mechanisms, the proposed approach significantly reduces energy consumption while maintaining acceptable performance. Experimental evaluation using simulated smart home scenarios demonstrates improvements in battery lifetime, reduced power consumption, and efficient AI-driven decision-making. The proposed framework offers a practical solution for a sustainable and long-lasting smart home IoT systems.