Distributed Learning for Reliable and Timely Communication in 6G Industrial Subnetworks

2025-06-13

AI Summary

To address untimely and unreliable transmission of event-driven critical control traffic in 6G industrial subnetworks—caused by constrained wireless resources, dynamic device activity, and high mobility—this paper proposes a distributed learning-based random access protocol. Our method introduces a novel implicit cross-subnetwork coordination mechanism, leveraging lightweight neural networks and online reinforcement learning. Coordination is achieved implicitly across subnetworks via competition signature signals broadcast by access points, eliminating explicit signaling overhead while dynamically optimizing access configurations to suppress collisions and improve timeliness. Evaluated in a dense industrial scenario with 60 subnetworks and 5 shared channels, the proposed protocol increases the timely delivery probability of critical events by 21% over classical multi-armed bandit baselines. It achieves superior performance in terms of latency, reliability, and scalability, while maintaining low communication overhead, strong adaptability to dynamic environments, and high robustness against mobility and resource fluctuations.

Technology Category

• Planning, Routing, and Scheduling: Scheduling under Uncertainty
• Search and Optimization: Distributed Search
• Game Theory and Economic Paradigms: Coordination and Collaboration

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Energy management for devices in mobile Web and WoT environments
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems

Abstract

Emerging 6G industrial networks envision autonomous in-X subnetworks to support efficient and cost-effective short range, localized connectivity for autonomous control operations. Supporting timely transmission of event-driven, critical control traffic is challenging in such networks is challenging due to limited radio resources, dynamic device activity, and high mobility. In this paper, we propose a distributed, learning-based random access protocol that establishes implicit inter-subnetwork coordination to minimize the collision probability and improves timely delivery. Each subnetwork independently learns and selects access configurations based on a contention signature signal broadcast by a central access point, enabling adaptive, collision-aware access under dynamic traffic and mobility conditions. The proposed approach features lightweight neural models and online training, making it suitable for deployment in constrained industrial subnetworks. Simulation results show that our method significantly improves the probability of timely packet delivery compared to baseline methods, particularly in dense and high-load scenarios. For instance, our proposed method achieves 21% gain in the probability of timely packet delivery compared to a classical Multi-Armed Bandit (MAB) for an industrial setting of 60 subnetworks and 5 radio channels.

Problem

• Minimize collision probability in 6G industrial subnetworks
• Improve timely delivery of critical control traffic
• Adapt to dynamic traffic and high mobility conditions

Innovation

• Distributed learning-based random access protocol
• Lightweight neural models with online training
• Contention signature signal for adaptive access