5G NR downlink scheduling time delay optimization system based on reinforcement learning

According to a 5G NR downlink scheduling time delay optimization system based on reinforcement learning, the downlink scheduling process of a base station is modeled into a partially observable Markov control problem according to the condition that the state is not completely observable in an actual scene, and the problem is solved through an Actor-Critic reinforcement learning algorithm framework. The system specifically comprises: a network monitoring module which is used for collecting related input of a downlink scheduler; a resource scheduler module which is used for simulating a fine-grained scheduling process of the base station through a simulator; a POMDP construction module which is used for processing the state of each time slot into a partially observable state and establishing an intelligent agent for different time scale tasks; a core controller module which is used for helping the POMDP construction module to complete action strategy formulation of the agents aiming at different time scale task