Multicast Routing in Wireless Sensor Networks: A Distributed Reinforcement Learning Approach

Wireless Sensor Networks (WSNs) are consist of independent distributed sensors with storing, processing, sensing and communication capabilities to monitor physical or environmental conditions. There are number of challenges in WSNs because of limitation of battery power, communications, computation and storage space. In the recent years, computational intelligence approaches such as evolutionary algorithms and swarm intelligence are applied successfully to solve many problems in WSNs. Most important of these problems are data aggregation, energy-aware routing, duty cycle scheduling, security and localization. These problem are in form of distributed so distributed approaches are required to solve them. Reinforcement learning is one of the most widely used and most effective methods of computational intelligence. In this paper, we used the reinforcement learning to solve multicast Quality of Service (QoS) routing. The simulation results showed that reinforcement learning is a suitable approach to solve this problem. The algorithm is implemented easy, it has the great flexibility in topology changes and it leads to optimized results. Distributed reinforcement learning provides compatibility mechanisms that show the intelligence behavior in complicate and dynamic environment such as WSNs. Using reinforcement learning, sensors behave autonomously, independently and flexibly during topology and scenario changes.