Energy-Aware Packet Schedulers for Battery-Less LoRaWAN Nodes

TL;DR

This paper proposes energy-aware packet scheduling algorithms for battery-less LoRaWAN IoT nodes, demonstrating that such designs can significantly improve transmission frequency and efficiency by optimizing scarce energy use based on real harvesting data.

Contribution

It introduces novel energy-aware scheduling algorithms tailored for battery-less IoT nodes, validated through simulations with real energy harvesting measurements.

Findings

Energy-aware scheduling increases packet transmissions.

Improves transmission frequency based on energy harvesting capabilities.

Significant gains depend on harvesting efficiency.

Abstract

The Internet of Things (IoT) enables a wide variety of applications where large sensor networks are deployed in remote areas without power grid access. Thus, the sensor nodes often run on batteries, whose replacement and disposal represent important economical and environmental costs. To realize more sustainable IoT solutions, it is therefore desirable to adopt battery-less energy-neutral devices that can harvest energy from renewable sources and store it in super-capacitors, whose environmental impact is much lower than that of batteries. To achieve the energetic self-sustainability of such nodes, however, communication and computational processes must be optimized to make the best use of the scarce and volatile energy available. In this work, we propose different energy-aware packet scheduling algorithms for battery-less LoRaWAN nodes, and compare them in various simulated scenarios, using actual energy-harvesting measurements taken from a testbed. We show that an energy-aware design can significantly increase the number of transmitted packets, also lowering the mean time between packet transmissions, though (as predictable) the gain strongly depends on the harvesting capabilities of the nodes.