作者: Haeyoung Lee , Seiamak Vahid , Klaus Moessner
DOI: 10.1007/978-3-030-25748-4_3
关键词:
摘要: While ultra-reliable and low latency communication (uRLLC) is expected to cater emerging services requiring real-time control, such as factory automation autonomous driving, the design of uRLLC stringent requirements would be very challenging. Among novel solutions satisfy uRLLC’s requirements, interface diversity widely regarded an efficient enabler connectivity. When mobile devices are connected multiple base stations (BSs) different radio access technologies (RATs) same data transmitted via links simultaneously, transmission reliability can improved. However, duplicate causes increase in traffic loads, leading resource shortage. Considering it, configuration multi-connectivity (MC) for important. In this paper, RAT selection scheme including MC proposed. By adopting distributed reinforcement learning (RL), each device could learn policy select appropriate RATs. Simulation results show that 20.8% improvements over single connectivity observed. Comparing method configure all time, 37.6% improvement achieved at high loads.