A Survey on Utilizing Reinforcement Learning in Wireless Sensor Networks Routing Protocols

Ali Forghani Elah Abadi, Seyedeh Elham Asghari, Sepideh Sharifani, Seyyed Amir Asghari, Mohammadreza Binesh Marvasti · 2022

This article reviews the control and routing methods in Wireless Sensor Networks. These methods are able to increase the energy efficiency by using the reinforcement learning technique, considered as one of the means of machine learning. It is based on the reward and punishment technique which has a behavior similar to the learning process in children. Appropriate energy management and therefore lifespan increase in wireless sensor networks is one of the main issues in these types of networks due to the energy consumption limitation in its nodes. The purpose of writing the current article is to get acquainted with the relative methods provided. In this article, various means which are trying to use the reinforcement learning process to improve the behavior of wireless sensor networks and to make them smarter are presented and analyzed. Meanwhile, the evolution of these methods and the ratio of the superiority of each in comparison to the others have been examined.

Read the paper · More papers on PaperTik