RL-MAC: a reinforcement learning based MAC protocol for wireless sensor networks

This paper introduces RL-MAC, a novel adaptive Media Access Control (MAC) protocol for Wireless Sensor Networks (WSN) that employs a reinforcement learning framework. Existing schemes centre around scheduling the nodes' sleep and active periods as means of minimising the energy consumption. Recent protocols employ adaptive duty cycles as means of further optimising the energy utilisation. However, in most cases each node determines the duty cycle as a function of its own traffic load. In this paper, nodes actively infer the state of other nodes, using a reinforcement learning based control mechanism, thereby achieving high thoughput and low power consumption for a wide range of traffic conditions. Moreover, the computational complexity of the proposed scheme is moderate rendering it pragmatic for practical deployments.