Delay analysis of mixed fronthaul and backhaul traffic under strict priority queueing discipline in a 5G packet transport network

Abstract

Virtualization of the base station for the purpose of centralization is being actively studied and researched as an implementation option for 5G mobile networks. Proposed as Cloud radio access network, the technology is expected to facilitate easier operation and maintenance than regular radio access networks. However, the base stations traffic has stringent delay requirements. In this paper, we explore the possibility of multiplexing fronthaul traffic and traditional backhaul traffic as it traverses over the metropolitan network while keeping the average fronthaul queueing delay and jitter under control. We analyze and simulate the cases of a single fronthaul flow and multiple fronthaul flows arriving at the packet switch assuming strict priority for the fronthaul queue. We propose a fronthaul frame aggregation strategy to improve the packet transmission efficiency while keeping the average fronthaul queueing delay and jitter constant regardless of the percentage of fronthaul traffic. While the criteria for aggregation is different for the 2 cases, we show that the optimal number of basic frames to aggregate is between 3-10 frames assuming the Common Public Radio Interface protocol.