Teoria de Jogos para Ajustes Dinâmicos de Parâmetros em Redes LoRaWAN

Abstract

Traffic managing in networks with a large number of nodes is challenging. Some nodes may transmit fewer messages to save energy; however, while other nodes would need to compensate for this, collaborating more with the network's objectives and transmitting more messages. LoRaWAN networks make this dynamic adjustment more complex since nodes must be kept inactive most of the time to save energy. As a result, each node's decision-making is generally based on outdated information, which makes convergence on the network's state very difficult. This paper proposes using the Gur Game, a game theory-based technique suitable for dealing with the individual decisions of each node and balancing their selfish and altruistic behavior. Simulations were performed using different scenarios and showed that this proposal can dynamically adjust network traffic in a massive LoRaWAN network, allowing each node to autonomously adapt its periodicity.