Buffer-aware Wireless Scheduling based on Deep Reinforcement Learning

TL;DR

This paper introduces a deep reinforcement learning approach for wireless downlink scheduling that optimizes throughput, fairness, and packet drop rate, outperforming baseline algorithms without requiring future information.

Contribution

The paper proposes a novel DRL framework with A2C algorithm tailored for wireless scheduling, enhancing efficiency and adaptability over existing heuristic methods.

Findings

DRL outperforms baseline algorithms in scheduling tasks.

Achieves performance comparable to genie-aided methods.

Improves sampling and training efficiency in scheduling optimization.

Abstract

In this paper, the downlink packet scheduling problem for cellular networks is modeled, which jointly optimizes throughput, fairness and packet drop rate. Two genie-aided heuristic search methods are employed to explore the solution space. A deep reinforcement learning (DRL) framework with A2C algorithm is proposed for the optimization problem. Several methods have been utilized in the framework to improve the sampling and training efficiency and to adapt the algorithm to a specific scheduling problem. Numerical results show that DRL outperforms the baseline algorithm and achieves similar performance as genie-aided methods without using the future information.