Reinforcement Learning Based Cooperation Transmission Policy for HetNets with CoMP Technology.

In order to ensure the performance of cell edge users and meet the explosive growth of people’s data services, Coordinated Multiple Points technology (CoMP) has become a key technology in the evolution of wireless networks. Cooperative cell selection is a major support point of CoMP multi-point collaboration technology. However, multi-differentiated clusters and radio resource optimization at this stage still cannot meet the requirements of cell edge users for signal reception quality and rate, and improvement of system performance is also facing serious challenges. This paper focuses on the cooperative transmission technology of base stations in CoMP system. This paper focuses on the cooperative transmission technology of green micro base stations in CoMP system. Using the method of reinforcement learning (RL), a semi-dynamic collaboration scheme that can be learned autonomously is proposed. Simulations show that this solution has a significant effect on improving system throughput and improving edge user rates.