Intelligent Resource Allocation in LoRaWAN Using Machine Learning Techniques

With the ubiquitous growth of Internet-of-things (IoT) devices, current low-power wide-area network (LPWAN) technologies will inevitably face performance degradation due to congestion and interference. The rule-based approaches assign adapt device parameters are insufficient in dynamic massive IoT scenarios. For example, adaptive data rate (ADR) algorithm LoRaWAN has been proven inefficient outdated for large-scale networks. Meanwhile, new solutions involving machine learning (ML) reinforcement (RL) techniques shown be very effective solving resource allocation dense In this article, we propose a concept using two independent allocating spreading factor (SF) transmission power devices combination decentralized centralized approach. SF is allocated RL contextual bandit problem, while assigned centrally by treating it as supervised ML problem. We compare our approach with existing state-of-the-art algorithms, showing significant improvement both level goodput energy consumption, especially large highly congested