DRL-based Joint Resource Scheduling of eMBB and URLLC in O-RAN

TL;DR

This paper introduces a distributed DRL framework using Thompson sampling for real-time resource scheduling of eMBB and URLLC in O-RAN networks, enhancing QoS and resource efficiency.

Contribution

It proposes a novel distributed DRL-based resource allocation method specifically designed for O-RAN architectures, enabling online decision-making at network edges.

Findings

Effective in meeting QoS for eMBB and URLLC users

Improves resource utilization in dynamic environments

Demonstrates real-time decision-making capability

Abstract

This work addresses resource allocation challenges in multi-cell wireless systems catering to enhanced Mobile Broadband (eMBB) and Ultra-Reliable Low Latency Communications (URLLC) users. We present a distributed learning framework tailored to O-RAN network architectures. Leveraging a Thompson sampling-based Deep Reinforcement Learning (DRL) algorithm, our approach provides real-time resource allocation decisions, aligning with evolving network structures. The proposed approach facilitates online decision-making for resource allocation by deploying trained execution agents at Near-Real Time Radio Access Network Intelligent Controllers (Near-RT RICs) located at network edges. Simulation results demonstrate the algorithm's effectiveness in meeting Quality of Service (QoS) requirements for both eMBB and URLLC users, offering insights into optimising resource utilisation in dynamic wireless environments.