Spreading Factor Analysis for LoRa Networks: A Supervised Learning Approach

Today, the Internet of Things (IoT) has been introduced in our lives, giving a variety of solutions and applications. The critical requirements for devices connected to the IoT are long battery life, long coverage, and low deployment cost. Some applications require the transmission of data over long distances, thus Low Power Wide Area Networks (LPWAN) have emerged, with LoRa being one of the most popular players of the market. In order, to improve energy consumption and connectivity problems, machine learning can be used in LoRa networks. In this paper, we intend to improve the energy consumption of end nodes by using machine learning models. For this reason, we present a comparison of classification algorithms, specifically, the k-NN, the Naive Bayes, and Support Vector Machines (SVM), for the Spreading Factor (SF) assignment in LoRa networks. The simulation results indicate that, both energy efficiency and reliability in IoT communications could be significantly improved using the proposed learning approach. These promising results, which are achieved using lightweight learning, make our solution favorable in many low-cost low-power IoT applications.