Designing Energy Efficient Reinforcement Learning Based Routing Protocol for Next Generation Large Area WSN: EE-RLRP

Within the healthcare domain, HealthCare (HC) has adopted cost-effective, tiny wireless sensors to transmit extensive patient data across networks. The primary objective of this paper centers around the creation of an energy-efficient (EE) Reinforcement Learning (RL)-based routing protocol (RP), denoted as EE-RLRP, specifically for the context of large, densely deployed sensor networks. To optimize the performance of IoT networks with exceedingly high device density, the paper introduces a method for learning and selecting energy enhancement parameters optimally. The approach involves adjusting critical network parameters such as the transmission range, node density, and overall network area, leveraging RL techniques to enhance routing protocol efficacy. In this scheme, the sink node assumes the default central position for network scaling. This allows for the network density to be scaled up by a factor of 3 to 4, maintaining the same ratio as the initially considered large network area. By fine-tuning the design parameters to extend network lifespan, improve energy efficiency, and enhance network scalability, it becomes feasible to assess the effectiveness of the recommended protocol in making superior routing decisions.