Reinforcement Learning (RL) for optimal power allocation in 6G Network
An RL-based approach in which the agent will select the access point with the highest SNR for establishing a communication link with the user is proposed, which helps in effective spectrum utilization and minimizing power wastage.
Abstract
The article introduces a Reinforcement learning (RL) based approach for optimal power allocation to the users based on the application demand. The density of users in a 6G network will be very high. The need for high data rates will also increase tremendously. Various applications like haptic communication, telerobotics, telesurgery, holographic imaging, etc., will require very high data rates and reliable connections for their realization. This paper proposes an RL-based approach in which the agent will select the access point with the highest SNR for establishing a communication link with the user. The optimal power is then allocated based on the application demand. The collection of rewards depends on the quality of service and the quality of experience. The agent always works for reward maximization by ensuring the quality of service. It is analyzed from the simulation results that the optimal power allocation enhances the energy efficiency and reduces the power utilization of the network. The proposed approach helps in effective spectrum utilization and minimizing power wastage.