TY - JOUR
T1 - UD-WCMA: An Energy Estimation and Forecast Scheme for Solar Powered Wireless Sensor Networks
AU - Dehwah, Ahmad H.
AU - Elmetennani, Shahrazed
AU - Claudel, Christian
N1 - KAUST Repository Item: Exported on 2020-10-01
Acknowledgements: The research reported in this manuscript is supported by King Abdullah University of Science and Technology (KAUST).
PY - 2017/4/11
Y1 - 2017/4/11
N2 - Energy estimation and forecast represents an important role for energy management in solar-powered wireless sensor networks (WSNs). In general, the energy in such networks is managed over a finite time horizon in the future based on input solar power forecasts to enable continuous operation of the WSNs and achieve the sensing objectives while ensuring that no node runs out of energy. In this article, we propose a dynamic version of the weather conditioned moving average technique (UD-WCMA) to estimate and predict the variations of the solar power in a wireless sensor network. The presented approach combines the information from the real-time measurement data and a set of stored profiles representing the energy patterns in the WSNs location to update the prediction model. The UD-WCMA scheme is based on adaptive weighting parameters depending on the weather changes which makes it flexible compared to the existing estimation schemes without any precalibration. A performance analysis has been performed considering real irradiance profiles to assess the UD-WCMA prediction accuracy. Comparative numerical tests to standard forecasting schemes (EWMA, WCMA, and Pro-Energy) shows the outperformance of the new algorithm. The experimental validation has proven the interesting features of the UD-WCMA in real time low power sensor nodes.
AB - Energy estimation and forecast represents an important role for energy management in solar-powered wireless sensor networks (WSNs). In general, the energy in such networks is managed over a finite time horizon in the future based on input solar power forecasts to enable continuous operation of the WSNs and achieve the sensing objectives while ensuring that no node runs out of energy. In this article, we propose a dynamic version of the weather conditioned moving average technique (UD-WCMA) to estimate and predict the variations of the solar power in a wireless sensor network. The presented approach combines the information from the real-time measurement data and a set of stored profiles representing the energy patterns in the WSNs location to update the prediction model. The UD-WCMA scheme is based on adaptive weighting parameters depending on the weather changes which makes it flexible compared to the existing estimation schemes without any precalibration. A performance analysis has been performed considering real irradiance profiles to assess the UD-WCMA prediction accuracy. Comparative numerical tests to standard forecasting schemes (EWMA, WCMA, and Pro-Energy) shows the outperformance of the new algorithm. The experimental validation has proven the interesting features of the UD-WCMA in real time low power sensor nodes.
UR - http://hdl.handle.net/10754/623267
UR - http://www.sciencedirect.com/science/article/pii/S108480451730142X
UR - http://www.scopus.com/inward/record.url?scp=85018625851&partnerID=8YFLogxK
U2 - 10.1016/j.jnca.2017.04.003
DO - 10.1016/j.jnca.2017.04.003
M3 - Article
SN - 1084-8045
VL - 90
SP - 17
EP - 25
JO - Journal of Network and Computer Applications
JF - Journal of Network and Computer Applications
ER -