Application of machine learning (reinforcement learning) for routing in Wireless Sensor Networks (WSNs)

Traditionally, protocols and applications in the networking domain have been designed to work in large-scale heterogeneous, hierarchically organized networks with low failure rate. In a Wireless Sensor Network (WSN) scenario, new problems arise and traditional routing protocols cannot be successfully applied. Additionally, in energy-restricted environments like WSNs the overhead of keeping routing information fresh becomes unbearable. In this problem context problem context, many researchers have turned their attention to the domain of machine learning (ML). The goal of this paper is to analyze the application of the Reinforcement Learning (specifically Q-learning) for an energy- aware routing scenario.