Deadline-Aware Multi-Agent Reinforcement Learning for TSN-Based Vehicular Edge Networks
MAPPO agents schedule TSN queues in vehicular edge networks, cutting latency versus centralized baselines.
The paper proposes multi-agent reinforcement learning for queue-level scheduling in time-sensitive networking for vehicular edge computing. Each TSN queue has an agent that jointly learns service order and time-slot duration to reduce deadline misses under speed-dependent latency requirements, using multi-agent proximal policy optimization. Compared with centralized single-agent methods, MAPPO reduces service latency by up to 66.2 percent and improves reliability by up to 271.8 percent. It also reports lower inference time than other multi-agent reinforcement learning methods.
- Each TSN queue is controlled by an autonomous MAPPO agent.
- Agents jointly learn service order and time-slot duration.
- Latency falls by up to 66.2 percent versus centralized methods.
- Reliability improves by up to 271.8 percent in reported tests.
From the original article
1 min read at arxiv.orgVehicular edge computing (VEC) enables latency-sensitive applications by bringing computing and networking resources closer to vehicles. However, existing approaches often overlook network contention among co-located services with heterogeneous and dynamic latency requirements.
By Bernardo A. C. Pereira · arxiv.org
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.