发表机构
Universidade Federal de Minas Gerais; School of Electrical Engineering and Computer Science, University of Ottawa（米纳斯吉拉斯联邦大学; 渥太华大学电气工程与计算机科学学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对TSN网络中队列级XR流量调度的现有方法存在队列依赖捕捉能力弱、对XR流量适应性差的问题，本文提出基于多智能体Transformer的多智能体强化学习方法，仿真显示其延迟与故障率降低效果显著。
AI 中文摘要
时间敏感网络（TSN）与移动边缘计算（MEC）为扩展现实（XR）等时间敏感应用实现超可靠低延迟通信提供了巨大潜力。然而，XR的广泛应用因MEC环境中服务共存而带来重大挑战，导致共享网络资源出现竞争；此外，XR流量类型在时序要求方面具有不同特性和关键性，进一步增加了此类环境的复杂性与动态性。尽管强化学习在动态网络场景的TSN调度优化中展现出潜力，但现有方法依赖集中式或高层级多智能体设计，且通常针对周期性、可预测的工业流量定制，限制了其对XR工作负载的适用性。这些方法存在两个缺陷：一是因控制粒度较粗，捕捉队列间依赖关系的能力有限；二是对高度动态且异构的XR流量适应性较差。为解决上述问题，本文提出一种用于队列级XR流量调度的多智能体强化学习方法，采用多智能体Transformer（MAT）通过对智能体的观测与动作进行注意力计算来建模队列间依赖关系，实现异构共存XR应用间的隐式协同。仿真结果表明，所提方法优于基线方法，实现了最高71.42%的延迟降低、最高83.2%的故障率降低，同时在所有队列中均能保持高可靠性。
英文摘要
Time-Sensitive Networking (TSN) and Mobile Edge Computing (MEC) hold strong potential for enabling ultra-reliable low-latency communication for time-sensitive applications, such as eXtended Reality (XR). However, the widespread adoption of XR introduces significant challenges due to co-located services in MEC environments, leading to contention for shared network resources. Moreover, XR traffic types have distinct characteristics and criticality in terms of timing requirements, further increasing the complexity and dynamics of such environments. Although reinforcement learning has shown promise for TSN scheduling optimization in dynamic network scenarios, existing approaches rely on centralized or high-level multi-agent designs and are typically tailored to periodic and predictable industrial traffic, limiting their applicability to XR workloads. As a result, these approaches suffer from (i) limited ability to capture inter-queue dependencies due to coarse-grained control, and (ii) poor adaptability to highly dynamic and heterogeneous XR traffic. To address these gaps, we propose a multi-agent reinforcement learning approach for queue-level XR traffic scheduling. We adopt the multi-agent transformer (MAT) to model inter-queue dependencies via attention over agents' observations and actions, enabling implicit coordination across heterogeneous co-located XR applications. Our simulation results show that the proposed method outperforms baselines, achieving up to 71.42% latency reduction and up to 83.2% reduction in failure rate, while consistently achieving high reliability across all queues.
