Dynamic Task Scheduling for Energy Harvesting Nodes in Wireless Sensor Networks

Abstract

The energy harvesting wireless sensor networks are explored in order to enable environmental monitoring without maintenance for a long term. In this paper, we propose a task scheduling algorithm which performs fair and periodic data collection from all energy harvesting sensor nodes. The goal of the proposed algorithm is to collect data effectively in the situation where the prediction accuracy of recharging profiles is low. In the proposed algorithm, tasks of sensing, reception and transmission for a sensor node are prioritized and scheduled according to the amount of harvested energy of the node. We evaluate the proposed algorithm by computer simulation. As for performance measures, we consider the average number of packets sent to the sink node per cycle for a node, the fairness index in terms of the node position. and the average transmission delay. Numerical examples show that the proposed algorithm increases the average number of packets sent to the sink node while it never degrades fairness index and the average transmission delay as compared with a conventional scheme especially when the amount of harvested energy is large.