Understanding the Computational Requirements of Virtualized Baseband Units using a Programmable Cloud Radio Access Network Testbed

摘要

Cloud Radio Access Network (C-RAN) is emerging as a transformativearchitecture for the next generation of mobile cellular networks. In C-RAN, theBaseband Unit (BBU) is decoupled from the Base Station (BS) and consolidated ina centralized processing center. While the potential benefits of C-RAN havebeen studied extensively from the theoretical perspective, there are only a fewworks that address the system implementation issues and characterize thecomputational requirements of the virtualized BBU. In this paper, aprogrammable C-RAN testbed is presented where the BBU is virtualized using theOpenAirInterface (OAI) software platform, and the eNodeB and User Equipment(UEs) are implemented using USRP boards. Extensive experiments have beenperformed in a FDD downlink LTE emulation system to characterize theperformance and computing resource consumption of the BBU under variousconditions. It is shown that the processing time and CPU utilization of the BBUincrease with the channel resources and with the Modulation and Coding Scheme(MCS) index, and that the CPU utilization percentage can be well approximatedas a linear increasing function of the maximum downlink data rate. Theseresults provide real-world insights into the characteristics of the BBU interms of computing resource and power consumption, which may serve as inputsfor the design of efficient resource-provisioning and allocation strategies inC-RAN systems.