Reinforcement Learning-Based User Scheduling and Resource Allocation for Massive MU-MIMO System

User Scheduling and resource allocation in massive multi-user multiple-input multiple-output (MU-MIMO) systems can be regarded as a multi-objective optimization problem from the perspective of spatial domain and time-frequency domain. An appropriate scheduling scheme is crucial for the multiple users in MU-MIMO system to efficiently exploit spectrum resources and improve system sum rate. This paper proposes a user scheduling and resource allocation scheme based on reinforcement learning (RL), which has the decision making ability to decide whether a candidate user should be scheduled on a given resource block (RB). Specifically, we modeled the user scheduling problem in MU-MIMO system as a Markov Decision Process (MDP), characterizing complex problems with sequential decisions. Simulation results demonstrate that the proposed RL based scheduling and resource allocation scheme yields observable performance gains compared with other scheduling algorithms.