Synchronization Uncertainty Versus Power Efficiency in LoRaWAN Networks

The availability of low-power wide area networks solutions opens new scenario in many application sectors. Among these technologies, long-range wide area network (LoRaWAN) was originally designed for consumer internet-of-things systems, but now it is investigated for distributed measurement systems applied to more demanding applications, including industry. In order to fulfill real-time requirements, the LoRaWAN end devices (EDs) are frequently requested to share a common timebase (i.e., to be synchronized). This paper deals with the tradeoff between the desired synchronization uncertainty and the energy available in the ED, which is usually battery powered. A new approach based on two algorithms for a posteriori synchronization and uncertainty estimation is proposed. After experimental validation of the proposed algorithms, a set of optimal energy/uncertainty tradeoff curves is obtained with the help of a specific tradeoff algorithm. Finally, the proposed tradeoff methodology is applied to two use cases concerning the design of: an industrial time division multiple access system and a predictive maintenance framework.