Energy Prediction Model Based on Kernel Partial Least Squares for Energy Harvesting Wireless Sensor Network
In energy harvesting wireless sensor network (EH-WSN), many energy harvesting technologies are developed to sustain the long-term operation of wireless sensor network. However, the prediction of harvested energy plays an important role for energy management. In this paper, we focus on energy prediction for EH-WSN. We first analyze the factors of affecting energy harvesting and the characteristic of the solar array. Then, the kernel partial least squares (KPLS) is proposed as the energy prediction model. According to the difference of energy intake for the days, months, season and year, the four energy prediction models are established. By extensive experimental analysis for real solar data in different areas, the proposed prediction model improves prediction accuracy than existing energy prediction algorithms in EH-WSN.
KeywordsEnergy prediction Kernel partial least squares Energy harvesting wireless sensor network
This research is supported by the National Natural Science Foundation of China (Grant No. 61401189), Natural Science Foundation of Jiangxi, China (Grant No. 20161BAB212036), and Natural Science Fund of Nanchang Institute of technology (Grant No. 2014KJ016).
- 3.Kansal, A., Hsu, J., Zahedi, S., Srivastava, M.B.: Power management in energy harvesting sensor networks. ACM Trans. Embed. Comput. Syst. (TECS 2007) 6(4) (2007)Google Scholar
- 4.Piorno, J., Bergonzini, C., Atienza, D., Rosing, T.: Prediction and management in energy harvested wireless sensor nodes. In: Proceedings of Wireless VITAE 2009, Aalborg, Denmark, pp. 6–10 (2009)Google Scholar
- 6.Hassan, M., Bermak, A.: Solar harvested energy prediction algorithm for wireless sensors. Qual. Electron. Des. 48(1), 178–181 (2012)Google Scholar
- 8.Cammarano, A., Petrioli, C., Spenza, D.: Pro-energy: a novel energy prediction model for solar and wind energy-harvesting wireless sensor networks. In: IEEE International Conference on Mobile Ad-Hoc and Sensor Systems, pp. 75–83 (2012)Google Scholar