Internet of Things applications have gained widespread recognition for their efficacy in typical scenarios, such as smart cities and smart healthcare. Nonetheless, there exist numerous unconventional situations where IoT technologies have not yet been massively applied, though they can be extremely useful. One of such domains is the underground mining sector, where enhancing automation monitoring through wireless communications is of essential significance. In this paper, we focus on the development, implementation, and evaluation of a LoRa-based multi-hop network tailored specifically for monitoring underground mining environments, where data traffic is sporadic, but energy efficiency is of paramount importance. We hence define a synchronization framework that makes it possible for the nodes to sleep for most of the time, waking up only when they need to exchange traffic. Notably, our network achieves a sub 40us proven synchronization accuracy between parent-child pairs with minimum overhead for diverse topologies, rendering it highly viable for subterranean operations. Furthermore, for proper network dimensioning, we model the interplay between network's throughput, frame size, and sampling periods of potential applications. Moreover, we propose a model to estimate devices' duty cycle based on their position within the multi-hop network, along with empirical observations for its validation. The proposed models make it possible to optimize the network's performance to meet the specific demands that can arise from the different subterranean use cases, in which robustness, low power operation, and compliance with radio-frequency regulations are key requirements that must be met.

翻译：物联网应用在智慧城市和智慧医疗等典型场景中已获得广泛认可。然而，在众多非传统场景中，尽管物联网技术能够发挥极大作用，却尚未得到大规模应用。地下采矿领域便是其中之一，通过无线通信增强自动化监测在该领域具有至关重要的意义。本文聚焦于专门针对地下采矿环境监测的LoRa多跳网络的开发、部署与评估，此类环境数据流量稀疏，但能源效率至关重要。为此，我们定义了一种同步框架，使节点能够在大部分时间处于休眠状态，仅在需要交换流量时唤醒。值得注意的是，在网络最小开销条件下，我们的网络在多种拓扑结构中实现了亲子节点间低于40微秒的同步精度，使其高度适用于地下作业。此外，为合理进行网络规划，我们建模了网络吞吐量、帧大小与潜在应用采样周期之间的相互作用关系。同时，我们提出了一种基于节点在多跳网络中位置估算设备占空比的模型，并通过实验观测验证了该模型。所提出的模型能够优化网络性能，以满足不同地下应用场景中出现的特定需求，其中鲁棒性、低功耗运行以及符合射频法规要求是必须满足的关键条件。