Work

Environmental-aware Reinforcement Learning-based Scheduler for Trustworthy 6G in the Factory Floor

Year: 2024

Type: conference-paper

Abstract: 6G networks will play a crucial role in Industry 4.0 advancing further smart manufacturing and trustworthy communication. This paper introduces a reinforcement learning scheduler that considers environmental knowledge designed to improve reliability and security aspects of trustworthiness by favoring to serve nodes when in line-of-sight. We evaluate the performance in a factory floor scenario for two different heights of remote-radio heads and various blockage densities. The results show that the proposed method outperforms traditional schedulers such as round robin and proportional fair in terms of reliability, availability and fairness enhancing thus reliability and security aspects of trustworthiness when the environment is prone to a mixture of line-of-sight and non-line-of-sight.

Source: Proceedings of the 28th Annual International Conference on Mobile Computing And Networking

DOI: 10.1145/3636534.3696728

Authors: Fjolla Ademaj-Berisha, Roya Khanzadeh, Andreas Springer, Hans-Peter Bernhard

Institutions: Silicon Austria Labs (Austria), Johannes Kepler University of Linz

Language: English

Topic: Advanced Wireless Network Optimization

Subfield: Electrical and Electronic Engineering

Field: Engineering

Domain: Physical Sciences

SDG: Industry, innovation and infrastructure

Open Access status: bronze

Funder: HORIZON EUROPE Framework Programme