Distributed Air-Time Reduction In Multi-Hop LoRa Networks With Multiple Spreading Factors

Abstract

LoRa is a physical-layer communication technology usually used for Internet of Things (IoT). LoRa has one specific parameter not found in typical wireless technology, called spreading factors (SF). The lowest SF,SF7, has the highest data transfer rate, so it is common to conFigure all the end devices to use SF7. On the other hand, for a LoRa network that spans a large area, higher SFs can be used to cover larger distances. LoRa’s radio modulation is based on the chirp spread spectrum technique which allows multiple devices using different SFs to transmit data in simultaneously. Even though LoRa supports long-distance data transmission (up to 15 km line-of-sight and 1.5 km in outdoor urban scenarios), single-hop networks do not support the coverage range required by certain scenarios such as smart cities, in which multi-hop networks are more appropriate. However, managing devices in multi-hop networks requires more effort. Centralized, manual configuration of devices become impractical in large-scale deployments. In this paper, we propose a distributed algorithm that improves the capacity of a multi-hop LoRa network. The algorithm arranges the end devices in a tree topology and assigns different SFs to different branches of the tree so that transmissions can be carried out in parallel at much as possible, hence minimizing total time on air. From simulation experiment, the algorithm can reduce time on air of the system up to 41%.