RELIANCE: A Reliable and Power Efficient LoRaWAN Optimization Approach for IoT Systems

Long Range Wide Area Network (LoRaWAN) protocol has been proposed as an alternative for deploying secure communication networks in Industrial Internet-of-Things ecosystems. Typically, such systems run on batteries demanding low energy consumption, depend on reliable communications, and are deployed using a low-cost infrastructure based on gateways that connect to the LoRaWAN server in a simple star network topology. For example, indoor and outdoor assets localization and tracking may be critical in harsh environments, where industrial processes and operating machinery introduce noise and interference, leading the transmission channel to be less reliable. In this work, we present RELIANCE, a first approach to demonstrate the technical viability of using LoRaWAN for optimizing packet loss and energy efficiency when operating in dynamic and harsh environments. For the first time, RELIANCE takes advantage of image processing techniques to estimate the path loss of the environment. By correct path loss estimation, the proposed method determine the most optimal Transmission Power (TP) for the end nodes, based on their distance to the LoRaWAN gateways. Results have shown that, based on the proposed approach, an accurate path loss estimation method with an error of up to 2 dB is feasible for optimal TP prediction.