Development of A Novel Resource Allocation Algorithm for 5G Ran Using Reinforcement Learning and Game Theory

Abstract

In this, the implemented DRL based DDQN-CASA algorithm for resource allocation with to approximate Deep Q-function with uses of double deep neural network, that gains from existing experience and adapts to changing environments. The results prove that the proposed simulation delivers improved Resource Allocation with DRL based technique compared to existing models, leading to reduced latency and improved output. Effectively, the DRL-based DDQN algorithm provided 13% higher profit than existing techniques. This learning covers further way of development and research in progressive DRL based Admission Control and resource allocation methods used for 5G/6G, it is providing better resources utilization and allocation, their developing vertical services meets performance demands. In this, compare performance of implemented modules with various existing modules based on factor latency, utilization rate and throughput. At last, we also include the conclusion and future directions of our related work.