Energy-Aware IoT and Edge Computing for Decentralized Smart Infrastructure in Underserved U.S. Communities

This paper presents a novel energy-aware Internet of Things (IoT) and edge computing architecture designed to support decentralized infrastructure in low-resource environments. The proposed framework combines virtual MIMO-enabled wireless sensor networks, lightweight edge AI inference models, and nanomaterial-based photovoltaic systems to autonomously manage public utility systems including waste, water, and energy. The system enables localized decision-making, reduces dependency on cloud services, and optimizes energy usage for off-grid deployment. A prototype implementation in a simulated rural setting demonstrated a 28% reduction in energy consumption compared to conventional IoT architectures, with average decision latency reduced to 800 milliseconds and uptime reaching 97.5% over a 30-day period. These results validate the feasibility of deploying scalable, autonomous infrastructure systems in environments with limited connectivity and power availability.