Energy Estimator for Weather Forecasts Dynamic Power Management of Wireless Sensor Networks

摘要

Emerging Wireless Sensor Networks (WSN) consist of spatially distributed autonomous sensors. Although an embedded battery has limited autonomy, most WSNs outperform this drawback by harvesting ambient energy from the environment. Nevertheless, this external energy is very variable and mainly depends on weather evolution. Therefore, including weather at design stage and weather forecasts at runtime is essential for autonomy management. This paper presents a Power Estimator to simulate node autonomy for various weather conditions and locations. This work also addresses the integration of Weather Forecasts in the Dynamic Power Management policy (WF-DPM). These two contributions significantly improve the system scaling and the energy availability prediction to help to achieve better node autonomy duration. Experimental results compared various locations to study the weather impact on the system autonomy.