Energy buffer dimensioning through energy-erlangs in spatio-temporal-correlated energy-harvesting-enabled wireless sensor networks

Abstract

Energy-harvesting-enabled wireless sensor networks (EHE-WSN), despite their disruptive potential impact, still present several challenges precluding practical deployability. In particular, the low power density and random character of the ambient energy sources produce slow deep fadings in the energy that nodes harvest. Unfortunately, the capacity of the energy buffers is very limited, causing that, at some times, the node might interrupt its operation due to lack of stored energy. In this context, a general purpose framework for dimensioning the energy buffer is provided in this work. To achieve this, a dynamics-decoupled, multi-source capable energy model is presented, which can handle fast random patterns of the communications and the energy harvesting, while it can capture slow variations of the ambient energy in both time and space. By merging both dynamics, the model can more accurately evaluate the performance of the sensor node in terms of the energy storage capacity and to estimate the expected energy of the neighboring nodes. In order to evaluate the performance of the sensor node, a statistical unit for energy harvesting resources, referred as the Energy-Erlang (E2), has been defined. This unit provides a link between the energy model, the environmental harvested power and the energy buffer. The results motivate the study of the specific properties of the ambient energy sources before the design and deployment. By combining them in this general-purpose framework, electronics and network designers will have a powerful tool for optimizing resources in EHE-WSNs.