RSLAQ -- A Robust SLA-driven 6G O-RAN QoS xApp using deep reinforcement learning

TL;DR

RSLAQ presents a DRL-driven xApp for robust SLA-aware QoS management in O-RAN RAN slices. By translating operator SLA intents into target KPIs and embedding them into the DRL reward structure, RSLAQ optimizes cross-slice PRB distribution and per-slice scheduling while preserving slice isolation. System-level simulations demonstrate strong SLA adherence (often >95% reliability) across diverse network conditions and reveal RSLAQ’s advantage over traditional schedulers and non-ML baselines in maintaining QoS for eMBB, URLLC, and MTC services. The work highlights the practical viability and impact of integrating SLA policies into ML-based RAN control for next-generation networks, while outlining scalability, security, and SLA-extension considerations for future work.

Abstract

The evolution of 6G envisions a wide range of applications and services characterized by highly differentiated and stringent Quality of Service (QoS) requirements. Open Radio Access Network (O-RAN) technology has emerged as a transformative approach that enables intelligent software-defined management of the RAN. A cornerstone of O-RAN is the RAN Intelligent Controller (RIC), which facilitates the deployment of intelligent applications (xApps and rApps) near the radio unit. In this context, QoS management through O-RAN has been explored using network slice and machine learning (ML) techniques. Although prior studies have demonstrated the ability to optimize RAN resource allocation and prioritize slices effectively, they have not considered the critical integration of Service Level Agreements (SLAs) into the ML learning process. This omission can lead to suboptimal resource utilization and, in many cases, service outages when target Key Performance Indicators (KPIs) are not met. This work introduces RSLAQ, an innovative xApp designed to ensure robust QoS management for RAN slicing while incorporating SLAs directly into its operational framework. RSLAQ translates operator policies into actionable configurations, guiding resource distribution and scheduling for RAN slices. Using deep reinforcement learning (DRL), RSLAQ dynamically monitors RAN performance metrics and computes optimal actions, embedding SLA constraints to mitigate conflicts and prevent outages. Extensive system-level simulations validate the efficacy of the proposed solution, demonstrating its ability to optimize resource allocation, improve SLA adherence, and maintain operational reliability (>95%) in challenging scenarios.

Figures (6)

• Figure 1: QoS operation in O-RAN. The numbered steps exemplify the operation of QoS with network slicing and SLA assurance proposed by the O-RAN Alliance ORANWG12023UCASpecificationORANWG22024WG2Usecases.
• Figure 2: Intent-based SLA management. The high-level policies of the network operator are translated and enforced into the RAN using the proposed framework.
• Figure 3: Example of A1 policy.
• Figure 4: Reward function of RSLAQ for different SLAs and network conditions.
• Figure 5: RSLAQ performance vs. other mechanisms. RSLAQ enforces target KPIs (SLAs) in different traffic and SLA conditions considered in the design (sufficient resources). When the system resources are insufficient, URLLC cannot be fullfiled.