MAREA: A Delay-Aware Multi-time-Scale Radio Resource Orchestrator for 6G O-RAN

TL;DR

This work tackles the challenge of delivering reliable, ultra-low-latency radio resource orchestration for multi-service 6G RAN deployments by introducing MAREA, an O-RAN-compliant framework that employs a Martingales-based delay bound to allocate guaranteed resources per uRLLC service. It integrates two control loops across near-RT and RT timescales, realized through four xApps and a Controller dApp, leveraging real traffic and capacity data via the E2 interface. Key contributions include a Martingales-based model for tight delay bounds, a backward-compatible orchestration architecture, and a learning-based estimator (MDN) for RB utilization, enabling more efficient per-service guarantees and shared RB across services. Simulations show up to an order of magnitude reduction in delay-violation probability and improved service accommodation compared with SNC-based and baseline methods, highlighting MAREA’s potential for scalable, real-world O-RAN deployments.

Abstract

The Open Radio Access Network (O-RAN)-compliant solutions often lack crucial details for implementing effective control loops at various time scales. To overcome this, we introduce MAREA, an O-RAN-compliant mathematical framework designed for the allocation of radio resources to multiple ultra-Reliable Low Latency Communication (uRLLC) services. In the near-real-time (RT) control loop, MAREA employs a novel Martingales-based model to determine the guaranteed radio resources for each uRLLC service. Unlike traditional queueing theory approaches, this model ensures that the probability of packet transmission delays exceeding a predefined threshold -- the violation probability -- remains below a target tolerance. Additionally, MAREA uses a real-time control loop to monitor transmission queues and dynamically adjust guaranteed radio resources in response to traffic anomalies. To the best of our knowledge, MAREA is the first O-RAN-compliant solution that leverages Martingales for both near-RT and RT control loops. Simulations demonstrate that MAREA significantly outperforms reference solutions, achieving an average violation probability that is x10 lower.

Figures (12)

• Figure 1: Integration of MAREA (i.e., green blocks) in the O-RAN architecture. For simplicity, only the northbound and southbound interfaces of the Controller dApp are shown. Other dApps would have similar interfaces.
• Figure 2: Illustrative example of how MAREA performs dynamic RB allocation at near- and scales.
• Figure 3: Evaluation of $K_{a,m}^{\prime}$ and $K_{s,m}^{\prime}$ for a service $m \in \mathcal{M}$ with a specific number of guaranteed $N_m^{min}$. This example uses the traffic pattern and channel conditions defined in the experimental setup (see Section \ref{['sec:PerformanceResults']}).
• Figure 4: Overview of the considered Mixture Density Network (MDN). Copied from Adamuz2024.
• Figure 5: Transitions of the finite-state machine to control the allocation for service $m$. From Adamuz2024.