A WAKE-UP INTERVAL ADJUSTMENT MAC PROTOCOL TO SUPPORT ENERGY EFFICIENCY AND QOS USING REINFORCEMENT LEARNING FOR WIRELESS SENSOR NETWORKS
Hye Jin Kim, Hye Yun Kim · Journal of Southwest Jiaotong University · 2023
In a WSN composed of sensors operated by batteries with limited capacity and difficult to replace, it is essential to reduce the energy consumption of sensor nodes to extend the network lifetime. In addition, as the industrial field to which WSN is applied rapidly expands, the demand for providing the quality of service (QoS) required by each application gradually increases. So far, many studies have been conducted on ways to reduce the energy consumption of sensor nodes and ways to support. In this paper, we proposed a multiple access control (MAC) protocol that reduces the energy consumption of the sensor and supports QoS in WSNs. The proposed protocol is an energy-efficient and QoS-supporting MAC protocol that considers the remaining energy of the sensor node, the amount of transmitted traffic, wake-up time prediction, and priority. The receiver and sender nodes predict each other’s wake-up time during packet transmission and reception and reduce their own energy consumption by reducing unnecessary idle listening time. In particular, the protocol uses reinforcement learning (RL) techniques to control the wake-up time of sensor nodes and support the priority transmission in WSN. The simulation results show that the transmission delay was reduced by up to 13.25%, and the lifetime was increased by up to 15.4% over the existing MAC protocol. Keywords: Wireless Sensor Networks, Multiple Access Control Protocol, Energy Efficiency, Quality of Service, Reinforcement Learning, Priority DOI： https://doi.org/10.35741/issn.0258-2724.58.4.81