Is Cloud RAN a Feasible Option for Industrial Communication Network?

Abstract

Cloud RAN (C-RAN) is a promising paradigm for the next generation radio access network infrastructure, which offers centralised and coordinated base-band signal processing in a cloud-based BBU pool. This requires extremely low latency responses to achieve real-time signal processing. In this paper, we analysed the challenges to introduce cloud native model for signal processing in C-RAN. We studied the difficulties of achieving real-time processing in a cloud infrastructure by addressing its latency-constraint. To evaluate the performance of such a system, we mainly investigated a massive MIMO pilot scheduling process in a C-RAN infrastructure under a factory automation scenario. We considered the stochastic delays incurred by the cloud execution environment as the main constraint that has has impact on the scheduling performance. We use simulations to provide insights on the feasibility of C-RAN deployment for industrial communication, which has stringent criteria to meet Industry 4.0 standards under this constraint. Our experiment results show that, concerning a pilot scheduling problem, the CRAN system is capable of meeting the industrial criteria when the fronthaul and the cloud execution environment has introduced latency in the order of milliseconds.