An Efficient Method for Optimal Allocation of Resources in LPWAN Using Hybrid Coati-Energy Valley Optimization Algorithm Based on Reinforcement Learning

In Wide Area Networks (WANs), resource allocation is essential since computer networks manage complex optimization difficulties. The goal of optimal resource allocation is to improve total computing productivity. Distributing numerous specialized data and communication technology services inside a WAN can be challenging due to a wide range of application demands. A system to provide enhanced data transfer rate with minimum power transmission in a Low Power Wide Area Network (LPWAN) with the utilization of reinforcement learning technique is developed in this work. Here, the reinforcement learning method is used to determine the parameters involved for minimizing the transmission power. The resources in the networks, like the channel, spreading factor, and transmission power in the LPWAN is optimized. These parameters optimization is aided with the help of the newly developed Hybrid Coati with Energy Valley Optimization Algorithm (HC-EVOA). A large number of reinforcement learning agents are generated that match the terminal hubs in the server of the LPWAN. Once the reinforcement agent is generated, then the optimized transmission parameter is given to these terminal hubs of the network. The optimization is carried out to enhance the throughput and reduce the energy consumption rate by the equipment in the network.Simulations are conducted on the developed model to prove the system’s effectiveness. Based on the analysis, at transmission power of 8, the developed HC-EVOA-based optimal resource allocation system obtains energy consumption as 12.86% lower than GSO, 19.29% lower than ROA, 2.76% lower than COA, and 2.81% lower than EVO.