A Nonclairvoyant Real-Time Scheduler for Ambient Energy Harvesting Sensors

Ambient energy harvesting also known as energy scavenging is the process where energy is obtained from the environment, converted, and stored to power small devices such as wireless sensors. We present a variant of EDF scheduling algorithm called EH-EDF (Energy Harvesting-Earliest Deadline First). Decisions are taken at run-time without having prior knowledge about the future energy production and task characteristics. We gauge the performance of EH-EDF by means of simulations in order to show its benefits. We evaluate and compare several variants of EH-EDF in terms of percentage of feasible task sets. Metrics such as average length of the idle times are also considered. Simulations tend to demonstrate that no online scheduler can reach optimality in a real-time energy harvesting environment.