Service-Oriented Attention HAPPO for xApps Coordination in Digital Twin-Enabled AI-RAN

Abstract

Coordinating multiple heterogeneous xApps in Open Radio Access Networks (O-RAN) is challenging because real-time inter-xApp synchronization introduces communication overhead, latency, and scalability bottlenecks under live deployments. This paper proposes an attention-based Heterogeneous-Agent Proximal Policy Optimization (HAPPO) framework that coordinates four heterogeneous xApps—power control, resource-block allocation, access control, and beam selection—through a Digital Twin (DT)-enabled training pipeline. The framework combines 1) hybrid actor heads supporting mixed continuous/discrete actions; 2) task-aware self-attention for implicit coordination without predefined communication graphs; and 3) DT-based asynchronous experience collection with delay modelling, online correction, and safe rollback. We further validate that the learned attention aligns with the physical coupling between xApps (Pearson <inline-formula> <tex-math notation="LaTeX">$\rho = 0.78 \pm 0.04$ </tex-math></inline-formula>), establishing an interpretable link between the policy and underlying network dependencies. On a VIAVI O-RAN RIC emulator across 10 random seeds, the proposed scheme attains a 91.2% aggregate QoS score and 1.05 ms inference latency, improves per-slice SLA satisfaction by up to 6.4% over strong MARL baselines (FACMAC, MAPPO, CommNet), and reaches the 80%-of-final performance level <inline-formula> <tex-math notation="LaTeX">$1.49\times $ </tex-math></inline-formula> faster than the attention-free counterpart. Improvements are statistically significant (<inline-formula> <tex-math notation="LaTeX">$p\lt 0.05$ </tex-math></inline-formula>, paired two-sided Welch <inline-formula> <tex-math notation="LaTeX">$t$ </tex-math></inline-formula>-tests with Holm–Bonferroni correction across baselines; under fluctuating UE demands and diverse service-level requirements).